Operation Soteria Bluestone Year One Report (accessible version)

Report authored by Professor Betsy Stanko (OBE)

With special thanks to Joint Strategic Academic Lead Professor Katrin Hohl and to contributors; Professor Miranda Horvath, Dr Kari Davies, Dr Kelly Johnson, Dr Olivia Smith, Dr Emma Williams, Jo Lovett, Gavin Hales, Tiggey May, Robin Merrett, Lizzie Peters and Alice Light.

We would also like to thank all those who have shared their time, thoughts and personal experiences as part of this research. We are sincerely grateful for your contribution.

Foreword

In my role as National Police Chiefs’ Council Lead for Adult Sexual Offences, I work with many dedicated and committed colleagues. Among them the hard-working policing practitioners and leading academics who worked together on the first year of Operation Soteria Bluestone^{[footnote 1]}.

The publication of the Operation Soteria Bluestone Year 1 report is a milestone in our journey to improving outcomes for victims of rape and sexual offences. Many of the findings are challenging and some are concerning. But this is not a bad day for policing or for criminal justice. This report provides an evidence base for the action required for transformational change. I believe this is the best opportunity in a generation to truly solve the problem of investigation and prosecuting rape. We must engage with it with an open mind and a full heart.

Throughout this programme, we have benefited from political leadership, investment and cross-government coordination. The work to improve justice for survivors of rape and sexual offences must be done in partnership, bringing together the police, government, justice agencies, the voluntary sector and communities. I remain absolutely committed to that approach and I want to build on the work set out in the CPS and Policing Joint National Action Plan, the Government’s Rape Review and the completion of a Joint Inspection and the Home Affairs Select Committee inquiry.

Of all the rich and deep, and probably unprecedented insight we have gained, for me the most illuminating and significant is the importance of understanding human psychology – and how perpetrators use and exploit it and us – in the investigation and prosecution of rape and the continuum that is sexual violence. We need to turn the tables on them by exposing their tactics and using it against them. I believe building this kind of specialist knowledge, supported by critical thinking and a problem-solving mindset, is the single most important change we can make.

As we progress into Year 2 and the development of the National Operating Model for the investigation of Rape and other Sexual Offences, Professor Betsy Stanko, Professor Katrin Hohl and I welcome feedback and involvement from partners. Your expertise and support is critical to our ability to effect real, lasting, transformative change.

Operation Soteria Bluestone presents a critical opportunity to transform the way police respond to, and investigate, rape and sexual offences and to shape the criminal justice landscape for victim of these heinous crimes. We must seize it.

Chief Constable Sarah Crew (QPM)

Executive summary

Operation Soteria was launched by the Home Office in June 2021 as a core action in the UK Government’s End-to-End Rape Review to help deliver the ambition to more than double the number of adult rape cases reaching court by the end of this Parliament (May 2024)^{[footnote 2]}.

A police-academic collaborative programme with a strong research foundation, the mixed-methods research design was informed by the “Five Pillars” theoretical framework (Hohl and Stanko, 2022)^{[footnote 3]} and the learnings from a Home Office Police Science, Technology, Analysis and Research (STAR) funded pilot with Avon and Somerset Police called Project Bluestone between January – March 2021. Police practitioner knowledge combine with that of academic experts to pursue a shared goal: doing right by victims and developing a new National Operating Model for the investigation of rape and other sexual offences (RAOSO^{[footnote 4]}).

The Year 1 work programme consisted of research deep-dives in 4 police forces (known as ‘pathfinders’) between September 2021 and August 2022. The 4 pathfinder police forces are: the Metropolitan Police Service, Durham Constabulary, West Midlands Police and South Wales Police.

Over this first year, the programme examined a wide range of data in the pathfinder forces. This included: analysis of all police records of rape and other sexual offences reported between 2018 and 2020/21 (depending on the force) , and cases concluded with Outcomes 14, 15 and 16^{[footnote 5]}, police ethnography observing rape investigators and victim liaison officers on shift, review of Body Worn Video footage of first response to rape disclosures, review of video-recorded Achieving Best Evidence (ABE) interviews with rape victims, interviews, focus groups and surveys of officers of all ranks and roles involved in rape investigations, focus groups with Independent Sexual Violence Advisers (ISVA) and focus groups with rape survivors^{[footnote 6]}, observation and review of police training, as well as a review of force internal rape and sexual offences investigations guidelines, protocols and written procedures, including on digital forensics.

The breadth and depth of the empirical research was only possible because officers of all ranks and roles provided access to various forms of data, allowed researchers to shadow them, took part in case file reviews, interviews, focus groups and surveys. Officers have invested significant time and trust in the process. The academic team were humbled by bravery and personal risks many officers were willing to take in exposing difficulties, and even failings, in the hope this will help their force improve. Victims and their supporters, particularly ISVAs, gave of their time and shared their experiences in an open way to help shape the changes to police investigation processes.

Detailed findings are found on page 23. Key high-level findings are:

Police force investigators lack sufficient specialist knowledge about sexual offending. There is a need for specialism and research informed specialist investigative practice for rape and sexual offences.

• This lack of sufficient specialist knowledge about sexual offending, and evidence-informed specialist investigative practice for rape and sexual offences impacts on the quality and outcome of investigations and victim engagement. This includes significant deficits in;

• knowledge about sexual offending behaviours, the nature of rape contexts and its impact on different victims from different backgrounds informed by academic research

• integration of this knowledge into force-led learning and development for investigators and other police staff (for example, intelligence and performance analysts and police trainers), weakening the skills and investigative strategies of investigators.

• corporate resource and support to retain, refresh and review the capability and capacity to apply this specialist knowledge

• analytic overviews about recorded of rape and other sexual offences, the resources required to provide these analytic products, and the recognition of the value these overviews have for local prevention

• a mechanism to monitor and manage the personal impact investigating sexual offences has on officers’ lives at work and at home.

• Specialist knowledge about rape and sexual offending was also needed by other police staff in addition to investigators.

Disproportionate investigation effort was being put into testing the credibility of a victim’s account. There is a need to re-balance investigations to include a thorough investigation of suspects’ offending behaviours.

• Investigations should centre on investigating the sexual offence, including any grooming, manipulation and coercion tactics employed by the suspect directly relevant to the alleged offence, informed by evidence-based knowledge of sexual offending behaviour and its impacts on victims, thus re-balancing investigations away from firstly and primarily focussing on the credibility of the victim as a witness.

• Forces must improve the identification and disruption of repeat suspects through systematic review of criminal history and intelligence force data, better knowledge sharing and use of a range of tactics aimed to ensure investigators capture sexual offending behaviours.

Learning and development currently available to investigators lacks specialist knowledge about offending. The detrimental effects of this lack of capability are magnified by high workloads, the complexity of rape contexts, and understaffing, impeding the quality and outcome of investigations, and in particular the quality of victim engagement.

• Learning and development infused with the specialist knowledge as above, is fundamental for improvement. Reflective practice is key to creating a whole force learning culture, which will together address the well-being of investigators, better investigations, and better service to victims.

• There needs to be rapid improvement in digital forensics capability and capacity through learning and workforce development.

• We encountered great variability within police forces in terms of officer attitudes beliefs, customs and social behaviour (“culture”) towards rape and sexual offences. Challenging internal cultures which undermine fair and equitable rape investigations is necessary as a matter of urgency.

• In all pathfinder forces, victim confidence and engagement largely suffer from the above working context within which investigation of rape and other sexual offences takes place. When officers lack the knowledge and resources to make complex decisions to meet victims’ needs or understand the contexts within which sexual offences take place, they default to investigation procedures and processes. Wider knowledge about the difference different contexts bring to the investigative strategy undermines a whole view of offending and of victim impact.

There is a direct link between officer burnout, a lack of learning and development for officers and the confidence of officers in whether they are using the right investigative strategies when conducting investigations.

• This takes a significant personal toll on investigators (both knowledgeable and inexperienced) and lack of recognition, demonstrated through corporate support for managing workloads and well-being, to enable investigators’ work to be appropriately responsive to victims and challenging to suspects.

• Burnout symptoms for emotional exhaustion were shown from a bespoke survey of officers in the pathfinder forces to be higher than amongst NHS staff during the first year of the Covid-19 pandemic (see Pillar 4 report).

• It is a corporate responsibility to resource rape and sexual offence investigations such that there is a realistic prospect of officers carrying out an appropriate investigation and have time to engage with victims.

• Overwhelmingly, officers were committed to grasp the opportunity presented by the Operation Soteria Bluestone to help transform their rape investigations and victim engagement despite the very high levels of burn out, over-demanding workloads and under-resourcing.

Strategic analysis of recorded rape is fundamental to understanding the offending contexts in investigation strategies and in monitoring performance.

• Around one-third of police recorded rapes in the pathfinder forces and one in ten other sex offences are marked as domestic abuse related. The substantial overlap between domestic abuse and sexual offending is significant and requires consideration in investigative strategies, safeguarding and how officers engage victims of different backgrounds. At present domestic abuse and sexual offences are seen as separate areas of policing in some forces, as well as in some aspects of victim support.

• Analysis of pathfinder force data reveals the charge rate for offences of rape differed greatly depending on the relationship between the victim and the suspect (and the profile of relationship types varies by victim ethnicity).

• Lower charges for cases involving intimate and former intimate partners, may be linked to cases resulting from incidental disclosure (such as when a victim is responding to questions during a risk assessment following an unrelated incident of domestic abuse) rather than a victim making a complaint of rape to the police with a view to seeking a criminal justice outcome.

• Rape charge rates vary by local policing areas within the pathfinder forces. Forces should be curious about why these differences exist and aim to provide equal service to all victims as well as sensitive to different communities’ different levels of trust in policing.

• Outcome timescales, and therefore the average length of investigations, can differ widely by outcome and force.

None of the pathfinder forces had sufficient data systems, analysts or analytic capability to support good strategic analysis to improve rape and other sexual offences investigations, contribute to any focused local crime prevention activity or any forward planning to improve the demands on the investigative workforce.

• Across all pathfinder forces, police records were missing or contained incorrectly entered data, for example on victim ethnicity, the victim-suspect relationship and incorrectly applied outcome codes, in a significant proportion of cases. Poor quality police data is a limitation to a solid grasp of any differences in justice outcomes that might impact some groups of victims.

• Missing and incorrectly recorded information in police records, and the resulting poor quality of police data limits understanding of differences in justice outcomes which might impact some groups of victims. The case file reviews raised concerns about the reliability of outcome variables.

• Police forces lack the analytic capacity and capabilities to understand the nature of demand and their ability to respond to it. Using and understanding the police force’s own data on rape and sexual offences is critical for a strategic approach to improving rape and other sexual offence investigation and crime prevention.

• Building workforce capability and capacity requires a police force to grasp the strategic issues presented by their rape and other sexual offences local profile. This requires making strategic use of police data, which in turn requires good enough police data, data systems and strategic analytic capability.

In summary, the findings reveal that policing needs a capable, confident, and reflective workforce which is equipped with evidence-informed knowledge about the impact of rape and sexual offences on victims, the contexts within which rape and other sexual offending happen, the nature of sexual offending locally, and offending strategies used by sexual offenders.

This research has challenged the way in which the police workforce is professionalised. It is important that policing moves away from its history of ‘training’ and toward an iterative, more open, research-informed learning culture. Specialist knowledge on sexual offending is critical to investigative strategies, not only to understand the crime but to understand how offenders groom investigators.

At the end of each of the four deep dives the findings were shared with the police force. The forces then took responsibility for owning the findings and developing an action plan and identifying corporate processes and improved ways of working to address the issues discovered through the deep dives, in conversation with the academics. The aspiration for the second year is a wholescale transformation in the way police respond to rape and other sexual offences by developing a national operating model for the investigation of rape and other sexual offences.

While these findings are focused on policing, police forces cannot create sustainable improvement alone. They are located within a wider policing landscape, which is only one part of the criminal justice system.

Several enabling agencies are required to support force wide change and create sustainable improvements in working practices. In all the pathfinder forces, we found that the influx of inexperienced investigators managing rape and other sexual offences created problems, for victims as well as for the experienced investigators. These problems are directly linked to insufficient learning and development and forces must address this deficit in collaboration with the College of Policing. It is vital that the learning from Operation Soteria Bluestone is one of the aspects HMICFRS will take into account in determining the efficiency and effectiveness of forces. Police and Crime Commissioners are vital in providing oversight and in holding police forces to account, commissioning local services and potentially enabling better joined up interagency working. Police decision-making on investigations is dependent on the subsequent decisions by the Crown Prosecution Service to charge suspects. The CPS are part of Operation Soteria and are also striving to bring about improvements in the delivery of justice.

The finding that one in three reported rapes occurs in a domestic abuse context is an issue that needs to be addressed not only by the police. Community sector support often (but not solely) is organised through support offered to victims of kinds of sexual or physical violence. This discussion will help frame the police approaches to victims’ engagement and accountability.

The terms trust and confidence have been placed at the centre of Operation Soteria Bluestone’s work. Research confirms that not everyone has equal trust in policing. The lower rates of trust for different victims from different backgrounds^{[footnote 7]}, for example, of the disabled^{[footnote 8]}, blend in complex ways with the ways officers treat victims. Trust and confidence are not just terms, but are lived experiences, which arise through police interaction with victims. It is enhanced through better engagement with victims and strengthened when people feel they have been treated fairly, respected, and have a voice (Jackson et al., 2012). Placing victims’ voices at the centre of this work, and at the centre of transformation, is critical going forward.

Finally, the programme demonstrated the value of academic-police collaboration. There are many lessons from this collaboration that will be shared with both the police service and the academic community.

Introduction

1. Police are the primary gatekeepers to criminal justice. As representatives of the state, law and society, police also wield significant symbolic power in defining the faultline between behaviour that is deemed ‘right’ or ‘wrong’. Police actions matter to citizens’ sense of justice. These actions are the substance of trust and confidence in the rule of law.

2. Most experiences of sexual violence remain unreported, so never interface with law and legal decision-making in practice. When a victim, or someone on their behalf, reaches to the police for help following sexual violence the state has a clear mandate to making this justice space accessible and responsive to them. How police treat rape and other sexual offences and engage with victims and suspects – including within their own ranks – sends a powerful signal to wider society about what is and is not morally acceptable. What police do when victims tell them about harm is a powerful tool in validating people’s experiences of that harm. Harnessing this symbolic power of the police, by visibly improving the police response to rape, has the potential to aid the cause of ending sexual violence, and violence against women and girls more broadly (Hohl and Stanko, 2022: 222).

3. This report sets out Operation Soteria Bluestone’s first year’s programme of work on a unique academic-police collaboration for transforming the police response to rape investigation in England and Wales. Operation Soteria Bluestone is a core element of the UK Government’s End-to-End Rape Review Action Plan and the broader Operation Soteria programme.^{[footnote 9]} The theoretical framework underpinning this academic-police collaborative programme is set out in Hohl and Stanko (2022, Appendix 1).

4. This Year 1 Report details the practical implementation of the commissioned programme of work, including challenges, as well as key findings^{[footnote 10]}. These findings require fresh thinking about rape and other sexual offences disclosed to police, and how police must engage with suspects, victims, and other agencies.

5. Finally, the report previews the National Operating Model and the Year 2 programme of work for Operation Soteria Bluestone. The aspiration for the second year is a wholescale transformation in the way police respond to rape and other sexual offences by developing a national operating model for the investigation of rape and other sexual offences. The CPS are also developing a national operating model for the prosecution of rape.

6. The national landscape sets the stage on which the first year of Operation Soteria Bluestone operated (this is a summary of the full report in Appendix 4);

a) There has been a four-fold increase in rates of police recorded rape since 2012/13, although the scale and timing of increases has varied between police forces. There are also large differences in rates of police recorded rapes between forces.

b) The overall composition of all sexual offences – and rape and other sexual offences specifically – recorded by the police varies between police force areas, as therefore does the nature of demand.

c) There are wide variations in rape and other serious sexual offence charge rates by police force, confounded somewhat by differences in the time taken to conclude investigations and therefore differences in the proportion of cases that remain open at a given point in time.

d) National data shows interaction of charge and prosecution conviction rates: high charge rates can be eroded by low prosecution conviction rates, and vice-versa.

Aims and objectives: Transformational change in the investigation of rape

Programme aims and ambitions

7. The aims are to:

a) Enable truly transformative change in police investigations of rape, resulting in radically improved outcomes (including, but not limited to, justice outcomes);

b) Procedural justice for victims and suspects of all backgrounds;

c) Fair and balanced rape investigations to focus on the alleged offence, including the entirety of the suspect’s relevant behaviour, without over-investigating the victim;

8. The ambitions are to:

a) Contribute to ending sexual violence;

b) Radically and sustainably improve procedural and outcome justice of police responses to rape and other sexual offences;

c) Adequately resourced and supported specialist rape investigators: experts who feel proud of their work and are valued by their force

9. This programme is different; its learning is co-produced by the academic research team and police officers. This would not have been possible without the active participation of many police officers who are committed to improving their practice. Corporate, individual, and force-wide change is consequently research-informed enabling the local police force to understand its internal workings – which significantly influence rape investigations and rape investigators – through knowing;

a) the problem of rape in their force (what is the demand generated through reported rape and other sexual offences);

b) the force’s ability to investigate in a way that challenges offending behaviour and disrupts harm to victims; and

c) the capability, competence, and capacity of its whole workforce (investigators and police staff) to enable the best decision making and interface with justice and victims.

10. Transformational change requires people in each force to care about improvement. The research team met so many of those who are committed to this. Little can change unless there is personal leadership, corporate and personal commitment from people across a force to align in-force processes and agree to a vision of what good investigations and victim care looks like. Any change should be informed by victims, community support, and Police and Crime Commissioner oversight. It should be supported by wider agencies responsible for policing oversight and support, such as the College of Policing, HMICFRS, and the NPCC.

Year 1 objectives

11. The Year 1 objectives for Operation Soteria Bluestone include:

a) Build on the learning from the Project Bluestone (detailed below) using deep dive methodology to understand the current picture of ‘as is’;

b) Support the pathfinder forces, plus Avon and Somerset as the pilot force, to develop plans to address issues highlighted in the individual deep dives and set up peer networks to facilitate active communication between the pathfinder forces;

c) Knowledge Exchange by sharing the learning continuously with all police forces in England and Wales through a National Learning Network.

Background and proof of concept: Project Bluestone

12. The Government’s End-to-End Review of the Criminal Justice Response to Rape began in March 2019. In the same year, the findings of the London Rape Review (2019), commissioned by London’s Victims’ Commissioner, showed that 86% of attrition of rape reports in London were accounted for by police and victim actions^{[footnote 11]}. It also set down another challenge to what a radical rethink of the policing approach to rape might look like^{[footnote 12]}.

13. The Government’s End-to-End Rape Review, the London Rape Review and many other reports and a large body of academic research, has demonstrated that persistent obstacles are faced by many victims in seeking a justice outcome following a police-recorded rape. Across policing in England and Wales, three decades of government research, His Majesty’s Inspectorate Constabulary and Fire and Rescue Services (HMICFRS) inspection reports, and internal policing research have not resulted in sustainable improvement to the investigation of rape. Poor justice outcomes in rape and other sexual offences continued to deteriorate year on year. Victims’ voices about the damaging journey through the justice system have labelled the justice response as bad, or worse than sexual violation itself^{[footnote 13]}.

14. The London Rape Review’s findings caused some members of the Mayor’s Office for Policing and Crime (MOPAC)^{[footnote 14]} to reflect and think how new approaches could bring about embedded and sustainable improvements to the policing response to rape ^{[footnote 15]}. These officials, working with Professor Stanko and Professor Hohl, started to drive a conversation across London, and with policing and Government to suggest a transformative approach. The analytical framework for this approach embraces the global scholarly evidence on procedural justice^{[footnote 16]} and applies it in practice to victim experience. Fundamental is a keen understanding of policing practice in action.

15. The pilot project was sponsored and hosted by (then) Deputy Chief Constable (DCC) Sarah Crew in Avon and Somerset Constabulary in her role as National Police Chiefs’ Council (NPCC) lead for Adult Sexual Offences as a response to the declining numbers of suspects being charged and prosecuted.

16. In January 2021, Project Bluestone, a pilot approach of the emerging design, was awarded funding through the Home Office Police Science, Technology, Analysis and Research (STAR) programme. The pilot project was designed to bring together academic experts and operational policing to explore ways of improving, indeed transforming, the police response to rape and other sexual offences.

17. Pathfinder Project Bluestone, delivered in Avon and Somerset in early 2021, was the first practical application of this new approach. The research findings generated were unique in UK policing, in part because research took place within and in conjunction with operational police officers and harnessed the expertise of academic researchers who have specialized in studying the impact of rape and other sexual assault on survivors, the public at large, policing and justice. This fostered a trusted collaboration with police officer in practice and in situ, to align academic knowledge gained through decades of scholarship. This project aimed to embed the principles of procedural justice for victims and organizational justice for police officers investigating rape.

18. Improving investigations is tethered to five pillars^{[footnote 17]} of research-led inquiry, which in April 2022 became six pillars. The original five pillar approach (as per Project Bluestone) explored the organizational mechanisms which drive better, procedurally just outcomes for victims in rape and other sexual offences. These are:

• Pillar One -Foregrounding suspect-focused investigations; which

• Pillar Two – Identify, disrupt and challenge repeat suspects; embedding

• Pillar Three – Systematic procedural justice to victim engagement during the police process;

• Pillar Four – Placing officer learning and development on the specialist academic knowledge on rape and sexual offending, and officer well-being at the core of the investigation process; and

• Pillar Five – Supporting all the above with good police administrative data-led, performance-savvy monitoring and evaluation of investigative strategies and justice outcomes and better understanding of the problem of rape and other sexual offences through strategic analysis.^{[footnote 18]}

The pillars work together to deliver the corporate conditions within which investigations and investigators can operate conducive to procedural justice to victims. Improvement flows from the police force’s understanding of its ability to deliver the above, and internally setting in train a programme of work to change the way the force resources, learns, supervises, and monitors performance alongside better ways of responding to victims of rape and other sexual offences.

Learnings from Project Bluestone

19. The overwhelming consensus of the police staff and the academic teams participating in Project Bluestone was that the core vision of the five pillars sparked the right conversations to address corporate, local solutions in-force and required the police service to design and own new ways of working. The research found obstacles in processes and decision making during the investigation of rape and other sexual offences. It also confirmed the malalignment of the corporate support needed to enable the best investigation strategies victims deserve. Details of the findings are found in Appendix 5 and features of the initial ‘state of as is’ in Avon and Somerset are mirrored in the pathfinder police forces.

20. The five pillars’ findings enabled Avon and Somerset to set a clearer vision for what tangible changes would produce a procedurally just outcome for victims. Together, the collaborative police-academic pillar teams created a foundation for a whole system, organisational change programme using academically informed tools and products to set a course for improvement.

21. The next steps for improvement in Avon and Somerset were informed by the researchers’ products and insight gleaned on:

• Suspect-focused investigations, aided by a project-generated ‘investigation map’, focusing on four fundamental functions of the investigation: short-term investigation; engagement with the victims; police-CPS liaison and long-term suspect focus.

• Better disruption and challenging of repeat suspects. Using the investigation map and specialist knowledge to create both short- and long-term strategies for targeting offending collaborating with other policing teams and agencies to work more holistically and proactively.

• Systematic procedural justice to victim engagement, enhanced using a grounded and informed victim booklet and a victim communication plan grounded in and informed by victims’ experiences, created through extensive consultation with ISVAs and policing colleagues, continuous feedback, and dialogue with policing colleagues and the ISVA service in Avon and Somerset.

• Officer learning and development, promoted through specialist knowledge about victims’ responses to rape and offenders’ strategies for offending, paying specific attention to the well-being of investigative officers. The project team for example produced a plan to integrate academic knowledge into learning and development about rape and sexual offence investigations, paying specific attention to the well-being of investigators;

• To promote better administrative data-led, monitoring and evaluation of investigative strategies, the project team produced a data improvement plan, with maps to enable better tracking of case progression and outcomes, and a template for conducting a strategic problem profile to better understand the nature of rape and sexual offences, investigative responses and justice outcomes.

22. Improvement to the investigation of rape should be aligned to corporate enablers supporting good victims’ engagement. The researchers recognise that improvement is slow because there are factors that are beyond the ability of the local force to change quickly, such as:

a) The length of investigative timescales is often long

b) The innovative materials for learning and development about sexual assault and sexual offenders are not available locally

c) The well-being support for officers in a force is lacking or insufficient

d) The capacity to produce good analytical products is hampered by lack of analysts, and limited dedicated/specialist analytical resource

e) The day-to-day demand of high workloads

f) The information to monitor what works best for investigative strategies and case progression does not provide what forces need to pinpoint rapid improvement. But the shoots of change are already visible.

23. Transformational work is currently underway in Avon and Somerset Police and will provide a strong case study about the implementation of change during the next year of work. The academic team were humbled by bravery and personal risks many officers were willing to take in exposing difficulties, and even failings in the hope this will help their force improve. We were also extremely grateful for victims who were willing to share their experiences to help the force improve.

The national programme set up: Operation Soteria Bluestone Year 1

24. In June 2021, on the publication of the Government’s End to End Rape Review Report on Findings and Actions[^19], Ministers apologized unreservedly for failing victims of sexual offences and committed to develop a new national operating model for the investigation and prosecution of rape by June 2023 ^{[footnote 20]}. Operation Soteria Bluestone was established and is funded by the Home Office to develop the policing aspects of the national operating model, building on the theoretical framework and learnings from Project Bluestone.

25. The police service (NPCC) and the Crown Prosecution Service (CPS) had already initiated a joint national action plan (JNAP)^{[footnote 21]} in March 2021 which established a governance structure and an improvement plan to address the poor justice outcomes in the investigation and prosecution of rape and sexual offences. The findings from Project Bluestone helped shape this action plan and set the ground for the planning of Year 1 of Operation Soteria Bluestone.

26. Four deep dives in police forces, using the methodological and logistical approach designed through the learning from Project Bluestone were conducted in the first year of the programme (September 2021 to August 2022). The pathfinder forces are the Metropolitan Police Service, Durham Constabulary, West Midlands Police and South Wales Police.

27. Working with research team leaders across seven universities (see Appendix 3), the co-academic leads (Stanko and Hohl) and a small central team established a complex network of central MOPAC and university project support, coordinating over 40 individual researchers with a mixed set of disciplinary skills and expertise. Academic pillar leads^{[footnote 22]} steered skilled mixed methods research teams. Disciplines included forensic psychology, sociology, criminology, data science, feminist studies and law, with expertise in working within or with policing and victim-survivors for many of the team. Researchers refined their methods drawing on the learning from Project Bluestone.

28. The logistics of this first year’s endeavour were complex. Seven universities; over 20 Data Protection Impact Assessments (DPIA) and data sharing agreements with four different forces, each of which had entirely different systems, priorities, and decision-making mechanisms for enabling any of the researchers to ask questions or explore crime data; security vetting for all researchers, multiple times, as some forces did not recognize the vetting from other forces. Four force project managers; pillar leads in each force; logistics of arranging innumerable interviews, focus groups, observation sessions, viewing of body worn video or video recorded interviews with victims – and more. It is not an exaggeration to say that these processes in and of themselves sapped an extraordinary amount of time. Ethics approval was obtained from each university^{[footnote 23]}.

29. A key complaint of victims who report to police is the possible loss of their phones as the starting point for the investigation. As noted above, to address this the team added another research pillar (Pillar six) to explore the digital element of RAOSO investigations, including, the way police apply digital forensic tools to gather (mostly phone) digital evidence, with more DPIAs and data sharing agreements and university contracts. It is difficult to portray the extensive and complex logistical landscape the project was addressing in a short time frame.

30. Over this first year, the programme examined a wide range of data, gathered through the following data capture methods:

a) Analysis of all rape and serious sexual offences reported between January 2018 and January 2020 or November 2021 from four forces, 81,705 rape and other sexual offences were analysed by the teams.

b) 233 case file reviews, internally reviewed by at least two investigating officers (usually of different rank) from each of the four forces, using a coding frame set by the academic team who provided guidance to complete.

c) Bespoke research of 741 full investigation records from four police forces’ use of recorded crime outcome codes^{[footnote 24]}. These codes are set by the Home Office, as a means of classifying police officers’ rationale for not being able to charge suspects who may be identified with rape or other sexual offences. These hundreds of cases provided a lens into the investigation of rape not available to researchers or the Home Office previously, which will be addressed later in the report.

d) In person observations in police stations of officers working rape and other sexual offences investigation shifts (including ABE interviews).

e) In three forces, watching and analysing footage from Body Worn Video.

f) In all four forces, in-depth review of training courses offered to investigating officers, including the Specialist Sexual Assault Investigators Development Programme (SSAIDP) course^{[footnote 25]}.

g) Police officer surveys in four forces exploring officer well-being, learning and development needs and corporate support provided by each force.

h) Interviews and focus groups with officers of all ranks and roles, ISVAs, support service provides, CPS lawyers in each of the four forces, and lived experience panels.

i) Extensive document reviews of all forces’ internal RASSO-related documents provided to us, such as protocols, procedures, guidance and much more.

j) In-force feedback sessions with senior leaders in force and together with the force’s VAWG working group.

31. Operation Soteria Bluestone focuses primarily on the work of detective investigators of rape and other sexual offences. The pilot in Avon and Somerset alerted the team to explore the skills and knowledge of investigators in particular. Investigation is a specialist skill in policing. Its warranted officers are required to acquire specialist courses to attain the role of detective investigator to investigate crime.^{[footnote 26]}

32. At its heart, Operation Soteria Bluestone’s engagement with the police forces extends the pilot in Avon and Somerset and includes:

a) an understanding of contribution of victims-survivors’ lived experiences of rape and sexual assault and with the criminal justice system (the research team are all experts in studies of violence against women and community sector^{[footnote 27]} and police responses to these; survivors’ panels are part of the method of challenging approaches and products);

b) a foundation for improvement based on a research-informed pillar framework that aligns actions and activity with ‘procedural justice theory’– with benefits for victims, offenders, officers and staff;

c) an appreciation of the value of continuous dialogue between police forces and academics to offer direction on sustainable and novel solutions for local improvement, input on investigative strategies and guidance on force data improvement and use of research and analytics to steer understanding of outcomes;

d) the generation of novel analyses drawn from police records of rape and sexual offences, police case reviews, and the wealth of data collected through multi method social science research techniques by over 40 researchers.

33. The research team learned that understanding the local police force matters and engagement and feedback needed to be bespoke to be effective. The insight from all pathfinder forces (including Avon and Somerset Police) contributes to the building of the National Operating Model for the investigation of rape and other sexual offences^{[footnote 28]}.

Operation Soteria Bluestone: Key findings Year 1

34. This next section addresses key overarching findings from Operation Soteria Bluestone’s bespoke research and analytics conducted in Durham Constabulary, Metropolitan Police Service, South Wales Police and West Midlands Police. These cumulative findings help build a picture of policing nationally, with the aim of providing the evidence base for the development of the National Operating Model. As such, forces and force-level data have been anonymised, and particular analyses are presented as illustrative of the rape contexts and victim characteristics across the pathfinder forces, using data from one or more forces. Together, the observations and findings lay bare the complexities of recorded rape and demonstrate the challenges to investigations of rape and other sexual offences.

35. Data-led enquiry for Operation Soteria Bluestone uses a clear approach to aligning the data on recorded crime to outcomes. As there is no consensus across government, policing and the wider criminal justice system as to what defines ‘seriousness’ in sexual offences other than rape, the analytic findings presented here are aligned to the Home Office and ONS approach to publishing police-reported/recorded crime. Government reports and inspections, including Operation Soteria, use RASSO as the shorthand to refer to rape and serious sexual offences. As discussed by Pillar Five in Appendix 11, Operation Soteria Bluestone’s academic team uses RAOSO, rape and other sexual offences, as the term that most accurately captures the data reported by the Home Office and the Office of National Statistics (ONS).

Police force investigators lack sufficient specialist knowledge about sexual offending. There is a need for specialism and research informed specialist investigative practice for rape and sexual offences.

36. This lack of sufficient specialist knowledge about sexual offending, and evidence-informed specialist investigative practice for rape and sexual offences impacts on the quality and outcome of investigations and victim engagement. In addition, supervision of investigative decisions also demonstrates the lack of specialist knowledge or its use in rape and sexual offence investigations^{[footnote 29]}.

37. The research from Operation Soteria Bluestone suggests that police forces expertise in the investigation of rape and other sexual offences should include:

a) knowledge about sexual offending behaviour, the nature of rape contexts and its impact on different victims from different backgrounds informed by academic research^{[footnote 30]}

b) integration of this knowledge into force-led learning and development for investigators and staff (especially analysts)

c) appropriate digital forensic technologies and strategies

d) corporate resource and support to retain, refresh and review the capability and capacity to apply this specialist knowledge

e) analytic overviews about recorded of rape and other sexual offences, the resources required to provide these analytic products, and the recognition of the value these overviews have for local prevention.

f) a mechanism to monitor and manage the personal impact investigating sexual offences has on officers’ lives at work and at home.

38. Specialist knowledge about rape and sexual offending was also needed by other police staff in addition to investigators. The voices of victims and police officers interviewed by this project reinforce the need for first responding officers, and others within the force, to understand more about how best to communicate and treat victims of sexual crime. So too better knowledge of data and intelligence analysts (often civilian police staff) is critical to the delivery of smarter policing performance in the investigation of rape and other sexual offences.

Disproportionate investigation effort was being put into testing the credibility of a victim’s account. There is a need to re-balance investigations to include a thorough investigation of suspects’ offending behaviours.

39. Investigations should centre on investigating the sexual offence, including any grooming, manipulation and coercion tactics employed by the suspect directly relevant to the alleged offence, informed by evidence-based knowledge of sexual offending behaviour and its impacts on victims, thus re-balancing investigations away from firstly and primarily focussing on the credibility of the victim as a witness.

40. In every pathfinder force there was a disproportionate effort put into testing the credibility of a victim’s account. This was not only found through the work of Pillar One (see Appendix 7), exploring suspect focus investigations but also Pillar Five. This Pillar analysed 741 outcome code cases and found evidence of forces using victim credibility as part of their explanation when deciding to no further action an investigation. Pillar Six, exploring the approach to the extraction of digital material and phone seizures also found a much heavier focus on victims when the suspect was identified. Concerns are raised in the timing of the digital extract from victims (see Pillar Three Appendix 9). Further concerns are raised by a lack of consideration of sexual offending strategies, such as grooming, which may assist in validating the victim’s account. Grooming behaviours were even more evident in the cases involving repeat suspects.

41. Forces must improve identification of repeat suspects through systematic review of criminal history and intelligence force data, better knowledge sharing and use of a range of tactics to ensure investigators capture sexual offending behaviours. Checks of prior reports of the suspects and use of intelligence from prior sex offences in which the suspect has been named were not always conducted. In all pathfinder forces, more than half of the named suspects had come to police attention before, demonstrating the scope and breadth of the repeat suspects (see Pillar Two, Appendix 8). Poor identification of repeat suspects affects investigative strategy and decision making throughout investigations and misses opportunities to stop further offending.

Learning and development currently available to investigators lacks specialist knowledge about offending^{[footnote 31]}. The detrimental effects of this lack of capability are magnified by high workloads, the complexity of rape contexts, and understaffing, impeding the quality and outcome of investigations, and in particular the quality of victim engagement.

42. Learning and development infused with specialist knowledge is fundamental for improvement. Reflective practice is key to creating a whole force learning culture, which could together address the well-being of investigators, lead to better investigations, and a better service for victims.

43. Some investigators lack any introductory learning on or access to continuous development about rape and sexual offences. Further, they were overwhelmed by the volume and complexity of cases and lack of corporate support to enable their work to be more responsive to different victims and different contexts. This impacted both those who were extremely knowledgeable about rape and sexual assault, feeling guilty but ultimately too time pressured to provide sufficient mentoring and supervision, and others who were extremely inexperienced (or even not yet qualified as) detective investigators. This lack of corporate support is not just in reference to the insufficient number of officers deployed to investigate rape and other sexual offences. It refers to the lack of corporate support services, including analytic and intelligence, learning and development and particularly welfare support^{[footnote 32]}.

44. There needs to be rapid improvement in digital forensics capability and capacity through learning and workforce development. Pillar Six explored digital evidence and found that there were several significant challenges for the pathfinder forces, and for the police service as a whole. These are grouped under three headings: technological, institutional, and investigative. Some examples include:

• Digital evidential material within sexual offences investigations has increased exponentially, as have the challenges digital evidence poses to policing generally, and RAOSO investigations in particular.

• Digital policing is complex and ever evolving; the Police Services’ capability to conduct sexual offence related digital enquiries and investigations is hampered by funding, training, quality management, digital competence and capability, technological inadequacy and officer capability.

• The interpretation and analysis of conversations from applications such as Snapchat, WhatsApp, Tinder, Bumble, TikTok or simple text conversations is complex and adequate training is lacking.

• Officers and police staff were often unaware of national and force level digital evidence policies and procedures. Ways of working were different in every pathfinder force. Key issues such as managing the volume of data and storage of this data were often beyond the capacity of an individual force to fix.

• No case file management system found in any pathfinder force for digital evidence. This results in no data being easily retrieved on acquisition/seizure of digital devices, extraction, analysis, storage, transfer to CPS, or deletion, causing time delays and frustration of investigators. Consequently, the process must be done manually from several systems across each force.

• There is lots of work in the area designed to improve officers’ access to technology, however, without adequate training, resources, or appropriate strategies in place actioned by officers, more technology by itself will not solve issues of using digital material in RAOSO​.

• Resourcing is a challenge across policing more broadly, particularly relating to the investigative and technological aspects of the use of digital material, which has resulted in untrained staff, an overreliance on full phone downloads which the evidence suggests is highly stressful for victims and police, and poor-quality downloads of digital evidence. (See Pillar Six, Appendix 12).

45. In addition to the lack of learning and development on rape, sexual offences and sexual offending, we, as a research team, encountered great variability within police forces in terms of officer attitudes beliefs, customs and social behaviour (“culture”) towards RAOSO. Challenging internal cultures which undermine fair and equitable rape investigations is necessary as a matter of urgency.

46. In all the pathfinder police forces, some officers interviewed displayed a culture of disbelieving victims. There are serving officers today who don’t think that RAOSO should be a priority for policing. Some stated that they believed that most reports of rape are just examples of ‘regretful sex’, or that if victims presented additional issues, such as mental health problems or alcohol/substance misuse then this was the victim’s problem and the legal system was not obligated to safeguard them. Not surprisingly, these officers were less enthusiastic about whether there would be any successful change from the work of Operation Soteria Bluestone. These officers were found to be in the minority but nonetheless, they contribute to a considerable drag on progress, or obstruct progress (either intentionally or inadvertently), and influence how new officers are socialised into RAOSO work.

47. Challenging policing cultures that undermine progress is important to achieving transformation in this area. It requires officers within the police service to challenge each other’s attitudes and behaviour. New joiners must be exposed to good/best practice in engaging victims and investigating RAOSO crimes. Police forces must take corporate responsibility to support them in their learning as well as in their supervision to become champions and advocates in this area of policing.

48. When officers lack the knowledge and resources to make complex decisions to meet victims’ needs or understand the contexts within which sexual offences take place, they default to investigation procedures and processes rather than connection with victims. In all pathfinder forces, victim confidence and engagement largely suffer from the above working context within which investigation of rape and other sexual offences takes place.

49. Victims’ confidence in the police and the criminal justice system (see Pillar Three appendix 9) is low.

50. Operation Soteria Bluestone includes the voices of victims and their supporters (such as Independent Sexual Violence Advisers) at the heart of its approach. There is an overwhelming consensus of victims and their supporters involved in Operation Soteria Bluestone, that victim voices must be at the heart of any force improvement plan alongside transparency and feedback.

There is a direct link between officer burnout, a lack of learning and development for officers and the confidence of officers in using the right investigative strategies when conducting investigations.

51. Burnout takes a significant personal toll on investigators (both knowledgeable and inexperienced) and there is a lack of recognition, demonstrated through corporate support for managing workloads and well-being, to enable investigators’ work to be appropriately responsive to victims and challenging to suspects. During the Year 1 it was found that;

a) Burnout symptoms for emotional exhaustion were shown from a bespoke survey (see Pillar 4) to be higher in rape and sexual offences investigators than amongst NHS staff during the first year of the Covid-19 pandemic.

b) There has been a de-professionalisation of the rape and other sexual offences investigator role. The lack of appropriate and suitable learning and development is worrying and undermines the ability of any force to upskill officers. The deployment of very inexperienced officers places additional burdens on the few experienced investigators contributing to stress and exhaustion.

c) There are structural and systemic barriers for organisational health in each force (the way it is structured, managed in silos and fails to acknowledge the problem) and it is the responsibility of the corporate centre of each police force to fix.^{[footnote 33]}

52. The level of exhaustion and stress as measured by proven psychological research instruments were very worrying. Officers and other police staff who were passionate about providing a better service, some of whom have been doing well despite the severe limitations of available continuous development and proper supportive supervision of their work, felt hampered and unsupported by their force. The toll on investigating officers and those around them were distressing even to the researchers.

53. Overwhelmingly, officers were committed to grasp the opportunity presented by the Operation Soteria Bluestone to help transform their rape investigations and victim engagement despite the very high levels of burn out, over-demanding workloads and under-resourcing.

Strategic analysis of recorded rape is fundamental to understanding the offending contexts in investigation strategies and in monitoring performance.

54. In-depth analyses of over 80,000 recorded rapes in the past four years across the five pathfinder forces highlight the following features of reported rape that are critical to understanding the offending contexts, informing investigation strategies and understanding the complexities of sexual victimisation.

55. Consistently across the pathfinder forces, around one-third of police recorded rapes were flagged as domestic abuse (DA) related, as were around one in ten other sex offences. In most forces, there was consistency between the DA variable and the victim/suspect relationship variable(s)^{[footnote 34]}. In addition, around one in ten rapes were recorded as being intrafamilial (in forces A, B and C the proportion ranged from 8.5 to 13.6 per cent.

Figure 2: Proportion of rape and other serious sex offences recorded as being domestic abuse (DA) related or involving a current or former partner

Source: RASSO data from Soteria pathfinder forces, 2018-20 (Force A) or 2018-21 (Forces B-D)

Notes: (1) Force D relationship data were very incomplete and so not a reliable indicator.

(2) Averages of Forces A to C only.

56. In forces A (shown below for illustrative purposes) and B, the proportion of rape offences that were DA flagged has been increasing in recent years, while in Force C the proportion has fallen and in Force D remained at the same level.

Figure 3: Proportion of RASSO allegations recorded by Force A that were domestic abuse (DA) flagged by reported year.

Source: Force A RASSO dataset, 2018-20

57. At present domestic abuse and serious sexual offences are considered separate areas of policing in many forces, as well as in some aspects of victim support. The substantial overlap between domestic abuse and sexual offending is significant and requires consideration in investigative strategies, safeguarding and how officers engage victims.

58. Analysis of pathfinder force data reveals the charge rate for offences of rape differed greatly depending on the relationship between the victim and the suspect. The profile of relationship types also varies by victim ethnicity. The exact pattern and charge rates vary somewhat between the four pathfinder forces (complicated by differences in the completeness of relationship data).

59. In Force A the largest grouping was current and former intimate partners (33%), which were associated with the lowest charge rate (1.2%) accounting for less than one in five charges (18.2%). While much less common, rapes where there was no prior contact between the suspect and victim^{[footnote 35]} (5% of all recorded rapes in Force A) were associated with a much higher charge rate (8.2%) and accounted for almost one in five charges (19.1%).

Figure 4: Rape charges in Force A (2018-20) by suspect/victim relationship

Rape charge/summons rate by relationship type

Total rape allegations by relationship type

Percentage of all charges

Source: Force A RASSO dataset, 2018-20

60. Questions about the fairness of police decisions and trust in investigative decisions are important in demonstrating equitable procedural justice to different groups who have different levels of trust in policing^{[footnote 36]}. The data on recorded rape by victim ethnicity from Force A is shown below. Furthermore, in Force A the data shows that outcomes (especially charge rates) vary systematically by suspect-victim relationship type, and the relationship profile of police recorded rapes varies by victim ethnicity. This goes some way to explaining why charge rates vary by victim ethnicity and suggests that to answer any question about whether victims from different groups or ethnicities are treated equally, forces need to more robustly record both the ethnicity of the victim the relationship between the victim and suspect.

Figure 5: Rapes recorded by Force A, by victim ethnicity and victim/suspect relationship

Source: Force A RASSO dataset, 2018-20

Victim ethnicity

Figure 6: Rape charge rates in Force A, by victim ethnicity and victim/suspect relationship

Source: Force A RASSO dataset, 2018-20

Charge rates

61. Lower charges for cases involving intimate and former intimate partners, may be linked to cases resulting from incidental disclosure. Case file analysis of 741 rape cases closed with outcomes 14, 15 and 16^{[footnote 37]} highlighted these cases are far more likely than other suspect/victim relationship types to have resulted from an incidental disclosure rather than a victim actively reporting a rape to the police with a view to seeking a justice outcome. In the case file sample, 33% of rapes involved intimate or former intimate partners, but they comprised 50% of rapes coded as ’disclosure not allegation’. Incidental disclosures may happen during a domestic abuse risk assessment (such as the DASH^{[footnote 38]}) following an incident domestic abuse incident which is unrelated to the police recorded rape.

62. Rape charge rates vary by local policing areas within the pathfinder forces (even when controlling for relationship type). Forces should be curious about why these differences exist and aim to provide equal service to all victims as well as sensitive to different communities’ different levels of trust in policing. In the case of Force A overall three-year rape charge rates ranged from 1.8% to 3.0% by area, but charge rates also varied when looking at particular relationship types. Stranger 1 and intimate/former intimate are shown to illustrate, ranging from 3.2 to 13.1% (a four-fold difference) and 0.7 to 2.3% (a 3-fold difference) respectively. In force B, the local policing area with the highest charge rate was 1.3 times as high as the area with the lowest charge rate. In Force C the highest charge rate was 2.4 times as high as the area with the lowest charge rate.^{[footnote 39]} These statistics should spark a dialogue as to why these differences exist. Regular publication of such statistics provides transparency, and for communities, a means for holding criminal justice agencies to account. As such their availability is central for policing in a democracy, and police-community dialogue.

Figure 7: Rape charge rates in Force A, by area and victim/suspect relationship

Source: Force A RASSO dataset, 2018-20

Data for this chart was not available to publish accessibly. Figure 7 is a bar chart showing that rape charge rates in Force A vary across local policing areas within the pathfinder forces both when considering overall charge rates as well as looking at particular relationships. For example, total charges ranged from 1.8% to 3.0% by area, and Stranger 1 ranged from 3.2 to 13.1%. The findings from this figure are described in paragraph 62.

63. Outcome timescales, and therefore the average length of investigations, can differ widely by outcome and force. Across the four pathfinder forces investigative timescales showed both signs of consistency, notably in respect of Outcome 15, and also wide variations, notable for Outcome 1 (charge/summons) in force A, for which timescales were especially long, and Outcomes 14, 16 and 18^{[footnote 40]} in force C, which were especially short. Force A had the longest outcome timescales across all main outcomes. Overall, the profile of outcome codes varies between forces.

Figure 8: Average time (days) from rapes being recorded to outcomes, by force

Source: RASSO data from Soteria pathfinder forces, 2018-20 (Force A) or 2018-21 (Forces B-D)

64. In the pathfinder forces, there are systematic differences in the time elapsed from when an offence occurred to when it was recorded by police by suspect/victim relationship. Time taken to report is typically longest for abuse by family members and people in positions of authority, and shortest for stranger rapes. Force B, which had a moderate level of detail in its suspect/victim relationship variable, illustrates this association. For example, while the median time to report was only one day for stranger rapes, it was 12 years for rapes by close family, 14 years for other relatives, and 38 years for perpetrators who were in a position of trust at the time.

Figure 9: Average and median time (days) from rape offence to recorded as a crime, by detailed suspect/victim relationship, Force B

Source: Force B RASSO dataset, 2018-21

65. In the case of Force B, a large number of non-recent (male) institutional abuse cases were reflected in the fact that male rapes were much more likely to be historic than female rapes. A similar but somewhat less marked pattern is evident in Force A, where 93% of police recorded rapes with female victims had occurred between 2001 and 2020, compared to 79% of rapes with male victims.

Figure 10: Composition of rape victims by sex and offence date (year), Force B

Source: Force B RASSO dataset, 2018-21

Data for this chart was not available to publish accessibly. Figure 10 shows the composition of rape victims by sex and offence date (year) in Force B as a bar chart. The chart shows that a larger proportion of cases with a male victim were recorded before 2002 than for female victims.

66. The pathfinder force age profile of victims differs between rape (as a discrete offence) and other sexual offences. Consistently across the pathfinder forces, the victims of police-recorded other sexual offences had a younger age profile than rape victims, but in both cases victims and suspects were typically of similar ages, which is to say that most offending is peer-on-peer. The chart below relates to Force C.

Figure 11: Profile of victims by offence type, age at the time of offence and victim sex. Victim ages 0 to 90.

Source: Force C RASSO dataset, 2018-21. Victim sex based on the Home Office crime code.

Data for this chart was not available to publish accessibly. Figure 11 is a line chart showing the profile of victims by offence type, age at the time of offence and victim sex. Victim ages 0 to 90. This is illustrated by Force C’s rape and serious sexual offences dataset from 2018-21 which illustrates that the victims of police-recorded other sexual offences had a younger age profile than rape victims. This figure is referenced in paragraph 66.

67. Younger victims (at the time of offence) typically take longer to report their victimisation to the police. Across all pathfinder forces, and for both rapes and other sexual offences, the younger victims were at the time of their victimisation, the longer they took on average to report that victimisation to the police; as above, these cases typically involved abuse by family members or people in positions of trust. Consistently around seven out of 10 rapes with victims aged 12 and under were recorded more than five years after they occurred (the figures for the available data in three forces were 68%, 69% and 72%). By contrast, older victims were much more likely to report rapes soon after the offence occurred, as this chart for Force A illustrates.

Figure 12: Rape victims age at time of offence and time between offence and report to police, Force A. Only known ages shown.

Source: Force A RASSO dataset, 2018-20

Data for this chart was not available to publish accessibly. Figure 12 is a bar chart illustrating rape victims age at time of offence and time between offence and report to police in Force A 2018-20. The chart demonstrates that older victims were much more likely to report rapes soon after the offence occurred. This chart is described in paragraph 67

None of the deep dive forces had sufficient data systems, analysts or analytic capability to support good strategic analysis to improve rape and other sexual offences investigations, which could contribute to any focused local crime prevention activity or any forward planning to manage the workloads of the investigative workforce.

68. Across all pathfinder forces, police records were missing or contained incorrectly entered data, for example on victim ethnicity, the victim-suspect relationship and incorrectly applied outcome codes, in a significant proportion of cases. Poor quality police data is a limitation to a solid grasp of any differences in justice outcomes that might impact some groups of victims. Data quality varied across the pathfinder police forces, including with respect to completeness, accuracy, reliability and consistency. In some cases, gaps could be partially filled using multiple variables in combination. In others, however, large gaps remained, undermining the reliability of findings. The following table summarises the size of data gaps relating to police recorded rapes.

Figure 13: Missing data relating to rape suspects and victims

Source: RASSO data from Soteria pathfinder forces, 2018-20 (Force A) or 2018-21 (Forces B-D)

Note: Force A: Data for 2018-20. Victim with a URN; numbered suspects; suspect/victim relationship combination of two variables. Force B: Data for 2018-21. Force C: Data for 2018-21. Suspect /victim relationship for 2018-20 only; combination of two variables (56% missing on main variable). Force D: Data for 2018-21. Suspect ethnicity excludes where there was no identified suspect

69. Evidence from the in-depth analyses of outcome code case files also showed the limitations of recorded data. The review of 741 rape case files closed with outcomes 14, 15 and 16 showed that very often where data has not been recorded in structured fields, it had nevertheless been recorded in case files, generally as free text such as in the investigation log or other parts of the file. This free text data can be used to supplement structured data, resulting in large increases in the completion rate of variables such as victim ethnicity, victim and suspect sex, and the suspect/victim relationship. However, big gaps in suspect age and ethnicity remain, which is thought to be at least partly related to the fact that some suspects are not identified. In other cases, it appears that if cases did not proceed very far, police did not find out this information or record it.

70. Police forces lack the analytic capacity and capabilities to understand the nature of demand and their ability to respond to it. Using and understanding the police force’s own data on rape and sexual offences is critical for a strategic approach to improving rape and other sexual offence investigation.

71. The need for police forces to understand their data on victims on recorded rape and other sexual offences is a fundamental principle in the emerging National Operating Model. No private corporate entity would operate its business without a keen grasp of its customers, its demands, its supply chain, its own workforce resources, while at the same time scanning the futures landscape to monitor risk and opportunities. For policing, this relates directly to crime prevention.

72. Force-level data and its analyses underpin any ability to measure improvement and to assure its sustainability locally. Analytic capacity and capability are not a bonus; it is critical to force-wide improvement in the investigation of rape and other sexual offences. The analysis of data should be shared with Police and Crime Commissioners and other local partners, especially support service providers.

73. As the chart below demonstrates, grasping the complexities between police force area, CPS charging decisions and conviction outcomes is not simple. It is strongly recommended that conversations about improvement should be informed by available data and careful analytics. It is only then can that debate on where and how to improve practice be grounded in a data driven, locally owned plan for improvement.

Figure 14: Examining the relationship between CPS charge decisions as a % of total police recorded rapes, and prosecution conviction rates – by Police Force Areas. England and Wales, 2019/20 to 2020/21 inclusive.

Sources: CPS VAWG Annual Data Tables, years ending March 2020 and March 2021. Home Office Police Recorded Crime and Outcomes Open Data (as at July 2021).

Data for this chart was not available to publish accessibly. Figure 14 illustrates a scatter chart examining the relationship between CPS charge decisions as a percentage of total police recorded rapes, and prosecution conviction rates by Police Force Areas in England and Wales. It shows there is much variation across forces in relation to charge decisions and prosecution outcomes with many forces experiencing higher charge rates and lower convictions rates and vice versa

74. Building workforce capability and capacity requires a police force to grasp the strategic issues presented by their rape and other sexual offences local profile. This requires making strategic use of police data, which in turn requires good enough police data, data systems and strategic analytic capability. Forces would then be able to work together with community sector and other criminal justice partners to consider wider crime prevention and victim support locally. Different agencies have different service users, including victims who may not or no longer wish to interact with police. Information about fairness, how different groups of people and different rape contexts are treated and fare through the justice system enables wider conversations about the harm of RAOSO locally. These questions include:

a) Who are the victims reporting RAOSO here?

b) What does the information about these victims tell us about the contexts reported rape which is important in investigation here?

c) What can we learn about victims’ needs, support and other factors that enable best partnership discussions on the kinds of resourcing needed for victim support here?

d) What does this data tell us about sexual offending locally?

e) Who are the suspects?

f) How many of the force’s suspects are repeat suspects?

g) How does repeat offending feature in investigative strategies for linked series sexual crime or the kinds of investigative resources allocated to stopping the harm caused by these repeat suspects here?

h) How does a survey of victims (in development by Operation Soteria Bluestone) provide local feedback from rape victims directly to forces about their experience of the process?

From deep dive learnings to force-wide change

75. What happens following the deep dive? Each force frames a strategic improvement plan through the bespoke force-specific feedback from the academic-led deep dive. As such, forces’ approaches to improvement plans are managed differently in the five forces. The project team offers advice and guidance, facilitates integration with academics to enable forces to develop improvement plans. It is important that the improvement plan addresses transformational not transactional change. It is highly recommended that the process is led by a force Senior Responsible Officer (SRO) supported by a project manager to enable better communication between pillar leads and those leading change within force.

76. Progress and change across the five pathfinder forces so far is mixed. Changeover of SROs and police pillar leads has slowed the creation of a roadmap for change in at least two of the forces. Avon and Somerset, having benefited early from the Project Bluestone, is much further ahead than other forces in working through the implications for required change within force. South Wales, using the insight generated throughout Year 1, integrated learning from the deep dives in other police forces and began improvement prior to their own deep dive.

77. Green shoots of change are evident in all pathfinder forces and feedback highlights just some of the positive responses to the programme and improvements. The following are brief summaries and observations from the forces who participated in Year 1. Throughout Year 2, academic pillar leads working in collaboration with in-force pillar leads will advise and monitor whether the improvement is having the intended impact.

78. Application of specialist knowledge in South Wales Police

The Operation Soteria Bluestone findings provided the evidence base for a significant uplift in resourcing, enabling South Wales Police to introduce specialist rape investigation teams across the force. The support from the Operation Soteria Bluestone team in the ongoing development of this specialist capability has been invaluable.

Regular continual professional development and participating in the pilot for the rape investigators development program, developed in partnership with the College of Policing, are seen as essential in the journey of our staff to become specialist rape investigators. Our work with our partners from New Pathways, Health, and the Crown Prosecution Service has been enhanced and we have seen specialist officers and CPS identifying opportunities via a suspect focused approach.

One such case resulted in a non-recent rape investigation, where the suspect was identified as having committed an offence elsewhere in the country. The cases were subsequently joined and heard at court at the same time, where he was convicted for both offences.

Working closely with our CPS Wales colleagues, we have seen gradual increases in the use of early advice, and since the introduction of the specialist teams, our charging rate has increased from 5.7% to 9.6% YTD. In addition, HMICFRS have provided some positive feedback in relation to the quality of our rape team investigations.

T/Detective Chief Superintendent, Phil Sparrow, South Wales Police

79. Strengthening police and CPS collaboration in Avon and Somerset Police

Over the past 18 months, our professional relationships have been transformed. This effective partnership has come from investigators and lawyers training together, sharing challenges, and engaging much earlier in the investigation to make balanced case-building decisions in reflective, mutual and professionally respecting conversations. This has been pivotal in the rapid improvement in victim trust, and overall positive outcomes.

We have established joint NFA (no further action) scrutiny panels, comprising of CPS and police, and crucially an ISVA and an academic, to identify any learning or best practice. We’re working collectively in a governance structure that allows the regular flow of honest conversations and ‘critical friend’ observations, which in turn drives our improvement activity and rewards good work with the ever-improving positive voice of the victim.

We are speaking in a whole new language; one of looking at the perpetrators behaviour first, hearing the victim throughout and challenging myths and stereotypes that have, for so many years been ‘the norms’. We’ve switched from focusing on how to overcome the challenges in cases, to a more holistic view which focuses on strengths and we are building cases today that we would not have done two years ago. We consider ourselves truly fortunate to be involved in what feels and looks like true progress – without being one bit complacent as to what still needs to be done; together.

Vicky Gleave, Senior District Crown Prosecutor, CPS South West and Avon and Somerset’s force lead for Rape and Serious Sexual Assault, Detective Superintendent Lisa Simpson

80. Transforming victim engagement in Avon and Somerset Police

I have been blown away by the patience, support and care that you have shown my client throughout this process. The time that you have spent to make her feel as comfortable as possible has not gone unnoticed. You have involved me in each step of the process, which in turn helped her to feel even more supported and for the whole process to run as smoothly as it can. I haven’t before seen such a fantastic example of joint working and I just wanted to say how grateful I am for your work.​

ISVA from ​Safelink

I felt sick at the thought of going anywhere near the police, but the experience was so different! You (the ISVA) and the officer have made me feel so safe, supported and truly listened to. Both your kindness has made a huge difference to my life whatever happens next.

Survivor Safelink

81. Developing specialist skills in the Metropolitan Police Service

Dr Patrick Tidmarsh, an internationally recognised authority on sexual offending, is collaborating with Operation Soteria Bluestone to deliver cutting-edge digital training in the form of five masterclasses. These have been accessed by the Metropolitan Police Service:

This different thinking has contributed to the team achieving monthly increases in rape detections since they watched the webinars in December 2021, reaching an 18% detection rate for rape in August and 13.5% FYTD.

Detective Superintendent Tim Mustoe, Metropolitan Police Service

82. Victim focused transformation in West Midlands Police

We are moving into the second phase of Operation Soteria Bluestone with a renewed energy and focus. We see this project as an opportunity to transform and change the way we respond to and investigate reports of rape and serious sexual offences, putting victims at the heart of this work and supporting them through every step of the journey. We have already implemented a number of measures after listening to victims to better understand their lived experience. All our investigators have been trained in Trauma Informed practice and we have developed some innovative go-to guides for all staff. We have a rape triage car with an Independent Sexual Violence Adviser (ISVA) on board to provide crucial immediate support. We also now have CPS lawyers in police buildings providing quicker guidance to staff. We know that some of this will be challenging but with the academic insight and support we are optimistic and in a much stronger position to create long term sustainable change.

Detective Superintendent, Clare Caddick, West Midlands Police

83. Improving well-being for investigators in Durham Constabulary

Well-being measures have been introduced for investigative teams including peer led trauma informed support, specialist trauma impact support training, well-being packs for late working and the provision of health and well-being clinics. A working group of frontline officers has been established to assist in improving service delivery through more effective consultation and feedback mechanisms to strategic managers. These measures, along with planned structural improvements being introduced such as shift pattern and supervisor ratios have been received well and will enhance the ability of investigative teams to deal with RASSO cases.

Detective Superintendent Nicola Lawrence, Durham Constabulary.

84. Offering new learning and development for investigators across England and Wales

Further to force-led improvements, Operation Soteria Bluestone established an approach of continuous Knowledge sharing through a National Learning Network (NLN) set up in October 2021 and the creation of peer-to-peer learning sets. The latter have been established for pathfinder force pillar leads, SROs, project managers and OPCCs involved in the deep dives. This enables each group to meet, share learning and provide one another with support as forces underwent deep dives and moved forward into planning force-wide change.

85. Following each force deep dive, the academic team also presented the learning from the deep dive and discussed the evolution of the cumulative insight provided by this research into the ‘as is’ state for the investigation of RAOSO in England and Wales. Practitioners and leaders from all UK police forces were invited to join, as well as CPS, OPCC and government colleagues. These regular virtual learning events provide opportunities to discuss the findings and preview the evidence based behind the emerging National Operating Model. Members of this network are also invited to access the findings from the force-level deep dives on a Soteria Bluestone Knowledge Hub group^{[footnote 42]}. This group is very active, with over 600 signed up members. The programme engaged and commissioned Dr Patrick Tidmarsh^{[footnote 43]} to bring his insight on how to understand sexual offending as the heart of strategic investigative strategies through a series of webinars made available via the NLN.

86. Officers attended the National Learning Network Events in their hundreds throughout this year (see appendix 6). They spoke of the way they learned how to improve and became advocates and champions for change. Year 1 demonstrates that there is a groundswell of officers who wish to participate in transformation from within.

87. The College of Policing have also collaborated with Operation Soteria Bluestone on the development of the RAOSO Investigative Skills Development Programme (RISDP), a new package is currently being piloted in the Metropolitan Police Service, West Midlands Police and South Wales Police. The pilot will be the subject of an evaluation in quick time, before being made available to other forces involved in the programme, and then more widely to other adopter police forces.

Transformational change in police forces and systemic change across the policing landscape

88. The conceptual framework underpinning Operation Soteria Bluestone harnesses an approach that consciously enhances the existing ways of working within a local police force. The logic for Operation Soteria Bluestone includes the understanding that to improve the RAOSO investigation and victim engagement; a force must start by interrogating how it investigates these offences, how it interacts with victims and how it supports its investigators now. Transformational change in the investigation processes requires an organisation to grip many parts of the corporate business, from workforce planning to community outreach. This support should be consistent to enable officers make the best decisions by giving them the continuous, refreshed expertise they need to do a good job.

89. Transformational change is only possible when the staff deeply care about improvement and are made part of the journey. Little can change unless there is personal leadership, corporate and personal commitment from people across a force to align in-force processes and agree to a vision of what good investigations and victim care looks like. This vision must align with the views of advocates and community sector agencies working around sexual violence.

90. Operation Soteria Bluestone has a view on what ‘good’ looks like, for investigations, good victim engagement and the kind of internal policing corporate change within policing that is necessary to achieve both. This insight is being brought into the draft of the National Operating Model. Improvement can be measured through justice outcomes and the experiences of investigating officers, victims, and victims’ supporters. Change solutions are organic to local force-level circumstances and community support. The National Operating Model will embrace the local, while setting out national expectations and standards of practice expected to deliver the best for victims and staff.

91. Individual police forces cannot, however, create sustainable improvement alone. The responsibility for transformation must organically flow from their senior executives in collaboration with Police and Crime Commissioners (PCCs). There are several enabling agencies that could – and should – support force wide change and assist the willing police officers to enable and to create sustainable improvements in their own working practices in response to rape and other sexual offences. Key enablers and dependencies include Home Office, NPCC, The College of Policing, the Crown Prosecution Service, the Inspectorates, and PCCs.

92. Operation Soteria Bluestone found a significant number of inexperienced investigators making decisions in RAOSO cases. As noted above, there is considerable work being done to upskill these joining officers as well as their supervisors who frequently have short tenures in post. This can only be done in close collaboration with the College of Policing.

93. The College of Policing has a critical role to play in supporting and accelerating force-level change. Operation Soteria Bluestone found that the support from innovative and iterative continuous learning supporting investigators’ understanding of RAOSO, and victim engagement, was largely under-resourced within forces (see Pillar Four, Appendix 10). Conversations have just started and are ongoing with the College of Policing about how the National Operating Model can sow the seeds for some form of Licence to Practice^{[footnote 44]} in the investigation of RAOSO.

94. Support for improvement is also critical from the HMICFRS. Two recent inspections by the HMICFRS/HMICPS set out the ‘as is’ state of the relationship between policing and the CPS, described as needing fundamental improvement. If transformational change is to be delivered and embedded, it is vital that the learning from Operation Soteria Bluestone is one of the aspects HMICFRS takes into account in determining the efficiency and effectiveness of forces. Engagement with HMICFRS confirms that they are willing and enthusiastic about providing insight and advice as the National Operating Model is developed.

95. PCCs have a crucial role to play in the transformation of the policing response to RAOSO. The PCC and their offices hold the police force to account, working to ensure the force delivers an efficient and effective service to the community it serves. The PCC’s role in signing off the force budget and setting strategy give them significant influence over the force’s delivery of RAOSO services. They also have a role in convening community safety and criminal justice partners, to make sure local priorities are joined up, potentially enabling better working between policing and CPS. The PCC’s commissioning of local ISVA services will also play a part in shaping victims’ experiences.

96. Operation Soteria Bluestone has been working directly with the PCC offices of the five pathfinder forces via a peer-to-peer network established by the programme. This has enabled the PCC leads in those force areas to understand the programme and its findings, with the aim of informing their ongoing oversight of forces’ response to RAOSO.

97. Police decision-making on investigations is influenced by officers’ experience of Crown Prosecution Service decision making in RAOSO cases. This is a key dependency. The CPS are part of the wider Operation Soteria programme, and they attend the Home Office led Operation Soteria Programme Board alongside Operation Soteria Bluestone colleagues. Though both agencies and change teams work independently focusing on their respective areas, there is close collaboration with findings being shared and open dialogue occurring. Discussions to align the National Operating Models that Operation Soteria Bluestone and the CPS are on-going.

Next steps and concluding remarks: Year 2 and developing the National Operating Model

98. Collaboration with the five pathfinder forces continues in the second year. Deeper analysis of the data already gathered on reported rape, case file analysis and case reviews, interviews, focus groups and other detailed work is aimed to refine the learning, the products and the shape of the National Operating Model (see pillar lead reports Appendices 7 – 12).

99. To support more forces with preparedness for the National Operating Model and ensure the pilot was sufficiently robust, Ministers announced an expansion to a further 14 forces in December 2021^{[footnote 45]}. An innovative and in-depth self-assessment approach, built on the learning from the existing deep dive research, has been developed for use in these forces. This collaborative approach with 14 forces, designed to emulate the process of the original deep dives, is helping forces gain granular insight and enhanced understanding of their practice, enabling them to identify strengths and areas for future development toward improving practice aligned to the National Operating Model.

100. Operation Soteria Bluestone has already sparked numerous conversations about policing, policing practice, the ingredients necessary to improve investigations of rape and sexual offences, victims’ voice in transformation of practice and more. These concluding comments are the opportunity to emphasise these and invoke the reader to participate in the consultations and numerous discussions that will take place through the research programme in Year 2.

101. First, conversations within policing, and about the nature of policing with civil society have been core in this work.

i) Throughout, the terms trust and confidence have been placed at the centre of Operation Soteria Bluestone’s work. Police engagement with victims, the research found, was not consistently supportive of, or consistent in the way victims felt treated by investigators or first responders. Trust and confidence are not just terms, but are lived experiences, and trust and confidence arise through police interaction with victims. They are enhanced through better engagement with victims and are strengthened when people feel they have been treated fairly, respected, and have a voice. This engagement with victims also takes place in a wider context where some groups of people have different levels of confidence in policing in general. Placing victims voice at the centre of this work, and at the centre of transformation, is critical going forward, and will be at the heart of the National Operating Model for the investigation of rape and other sexual offences (see Appendix 2 for draft contents of this model).

ii) Throughout, the research found that police processes and procedure often trumped empathetic engagement with victims. This had an impact on police officers as it did on victims.

iii) That police processes and procedures trumped empathy was found by the research to be embedded in traditional policing approaches to learning and development. The consensus of the researchers across all the pillars is that this failure of traditional policing approaches to morph at a faster pace is linked to its history of ‘training’^{[footnote 46]}. Today, it is important to move toward an iterative, more open, research-informed learning culture in policing. This will take much more work.

102. Second, the use of data on the recorded rape and other sexual offences in the five pathfinder forces (and suggested across policing) needs an uplift in analyst staff. Specialist knowledge on sexual offending, as Chief Constable Crew suggests in her foreword, is critical to investigative strategies, not only to understand the crime but to understand how offenders groom investigators. Investigators of complex crime need to know about the strategic picture of offending, victims and suspects locally. This was largely missing in the pathfinder forces.

103. Third, the one in three recorded rapes occurring in a domestic abuse context is an issue that needs to be addressed not only by the police. Community sector support often (but not solely) is organised through support offered to victims of kinds of sexual or physical violence. This discussion will help frame the police approaches to victims’ engagement and accountability.

104. Fourth, the programme is only beginning to discuss with forces, and others, how to maintain an investigative workforce that has specialist knowledge and can flex to account for different rape contexts, such as that necessary for investigations of sexual exploitation, coercive control, or trafficking. Many police forces have specialist domestic abuse officers; few police forces have the number of specialist rape and other sexual offences officers necessary to meet the recorded workload. This project is not necessarily advocating stand-alone specialist rape investigation units. Local forces must find the best local solution on how to integrate specialist knowledge within investigation teams for the best outcomes for diverse victims ‘experiences and in different rape contexts.

105. Fifth, the programme demonstrated the usefulness of the academic-police collaboration, in-time and in-situ. There are many lessons from this collaboration that will be shared with both the police service and the academic community.

106. Finally, the project team looks forward to another year of robust conversations and sharing learning. A big thank you to all those who participated in the research and will continue to support the development of the National Operating Model for the investigation of rape and other sexual assault, which is due to be published in June 2023. Some of the accompanying products will be available then, and others will be delivered by September 2023.

Appendices

Appendix 1: The Theoretical Framework set out in Hohl and Stanko (2022)

Hohl, K., Stanko, E.A. (2022). Five Pillars: A Framework for Transforming the Police Response to Rape and Sexual Assault. International Criminology, 2, 222–229. https://doi.org/10.1007/s43576-022-00057-y

Pillar leads

Suspect focused investigations: Prof Horvath

Repeat suspects: Dr Davies

Procedural Justice approach to victim engagement: Dr Johnson and Dr Smith

Learning, development and officer well-being: Dr Williams

Data and performance: J Lovett

Digital Forensics: T May

Academic leads: Prof Stanko OBE and Professor Hohl

Appendix 2: Iterative draft outline of the National Operating Model

This is a working draft ‘table of contents’ outline of the National Operating Model, which will be more fully developed ahead of publication in June 2023.

Operation Soteria Bluestone: Iterative DRAFT of the National Operating Model

Aims

a) Enable truly transformative change in police investigations of rape, resulting in radically improved outcomes (including, but not limited to, justice outcomes);

b) Procedural justice for victims and suspects of all backgrounds;

c) Fair and balanced rape investigations to focus on the alleged offence, including the entirety of the suspect’s relevant behaviour, without over-investigating the victim;

Ambitions of the NOM

a) Contribute to ending sexual violence;

b) Radically and sustainably improve procedural and outcome justice of police responses to rape and other sexual offences;

c) Adequately resourced and supported specialist rape investigators: experts who feel proud of their work and are valued by their force

How the NOM differs from reviews, inspections, or conventional research outputs:

• Co-production between local force officer/staff pillar leads and academics

• Exploration of the barriers to change locally

• Getting beyond ‘personality-led’ to ‘whole system supported’

• Honest, data led exploration of key issues

• Managing the uncomfortable truth, and winding our way around a corporate solution

• Commitment to co-production with the community sector on the evolving National Operating Model

Draft structure of the NOM

1. Aims and objectives

a) Increase charge rate

b) Reduce disparity in charge rates between different BCU/geographical areas within force

c) Reduce disparity in charge rates for different relationship types

d) Reduce the disparity of experience and outcomes among different victim groups

e) Increase procedural justice for victims, victim “satisfaction” and trust

f) Improve officer well-being, competence, and capacity using workforce strategy

g) Create a self-sustaining specialist, knowledgeable workforce through more relevant and academic-informed learning and development approach

h) Strengthen peer to peer support systems for sharing practitioner specialist knowledge in force, and across all forces nationally

i) Become a data-led, smart organisation

2. Principles

a) Specialist knowledge enhanced through police-academic collaborative co-creation and exchange of knowledge

b) Specialist knowledge driving investigative strategy and victim engagement

c) Routine monitoring of the capability of the workforce with specialist knowledge

d) Parity of resource and organizational priority of RAOSO in line with other forms of serious violent crime

e) Procedural and organisational justice at the heart of victim and staff care

f) Self-sustainability of continuous improvement in the investigation of RAOSO

g) Acknowledge and support a whole systems theory of change supporting continuous improvement in the investigation of RAOSO

3. Addressing the investigative opportunities for workforce flex for resourcing investigative teams

a) Addressing DA-RAOSO overlap evidence: corporate knowledge about sexual offending and disruption of offending

b) Functional equivalence: investigative and learning resourcing which is more responsive to police-force specific demand profile, including DA, CSA, etc.

4. Specialist knowledge is required: Officer understanding, knowledge, and skills needed to investigate RAOSO

a) Addressing tension between suspect-focused and victim-centric through specialist knowledge and competent use of professional judgement

b) New interim RAOSO learning programme

c) Review and revision of current learning approach to RAOSO linked to SSAIDIP and a continuous programme of learning (with College of Policing)

5. Tools and Processes to enable improvement

a) Self-assessment tool for police forces to assist in monitoring progress and need for focused improvement

b) Digitalised Investigation Map which includes:

i. Guidance on repeat suspect identification, investigation, and disruption

ii. Guidance on digital forensics proportionate use in RAOSO investigations

iii. Digitalised guidance embedded within the investigation map (with police digital service)

c) Systemic approach to victim engagement throughout the investigation and criminal justice process, including:

i. Guidance on specific consideration of DA-RAOSO overlap, strengthening use of coercive control as a preferred charge when appropriate

ii. Guidance on consideration of intersectionality, special needs of victim and how this might be addressed

iii. Guidance on IVSA-police collaborative working

iv. Guidance on the use of a victim feedback survey

v. Victim booklet

vi. Guidance on the best practice for continuous feedback from NFA-scrutiny panels

vii. Police-Victim-ISVA communication plan

d) Learning and development: Guidance on in situ learning and continuous development, using introduction of the interim learning programme designed by Operation Soteria Bluestone with the College of Policing

e) Guidance on the learning and development required for first responders and new joiners

f) Guidance for senior officers to support performance management and monitoring through integrated data with good analysis of the problem of RAOSO locally, with routine monitoring of the right measures, enabling better strategic and operational decision making and the identification of where improvement is needed in time

g) Guidance on what is best practice in assuring organizational justice for staff, incl. officer retention and reward

h) Guidance exploring best knowledge and advice on assuring officer well-being and resilience

i) Guidance on long-term, self-sustaining embedding of the Operation Soteria National Learning Network (NLN) and peer-support network to assure rapid and ongoing cross-force sharing of best practice arising in different forces.

6. Protocols

a) Suspect-focused: all protocols (guidance etc) arising from the investigation map (embedded into digitised Investigation Map)

b) Suspect-focused: appropriate interviewing techniques and guidance

c) Victim-focused/investigation outcomes: Guidance on reporting & crime recording protocols (e.g.) disclosure vs complaint and suggested approach to anonymised intelligence-only reporting;

d) Victim-focused: digital disclosure/independent legal advice to victims

e) Officer-focused: Protocol for force pledge to monitor force guidance on reflective practice, provision of specialist continuous learning, and active monitoring of the well-being of its staff

f) Guidance on improving crime recording to reflect victim characteristics, particularly protected characteristic monitoring of justice outcomes

7. KPIs for monitoring progress and success (aligned with aims and objectives): Key features on knowing how well forces are investigating RAOSO

a) Justice outcomes, at minimum: split by relationship type, split by disclosures and complaints, split by ethnicity, split by rape and SSO

b) Repeat suspect identification and disruption

c) Victim satisfaction survey

d) Time to outcome

e) Officer caseload

f) Supervisor to investigator ratio

g) Officer sick leave and churn

h) Officer achievement of learning and development specialist knowledge

i) Legacy plan for retaining a specialist knowledge/capability in force

8. Public confidence and accountability

a) CPS (including early advice and other legal facilitators)

b) Transparency and community sector involvement in monitoring progress

c) Local police force ISVA resourcing and good working relationship, including by BCU/geographical area to avoid a postcode lottery

d) Lived Experience scrutiny panel and mechanisms to incorporate feedback as an accountability loop

e) The availability of independent legal advice for victims

f) Mechanisms for efficient and effective cross agency oversight – NFA panels, feedback to public and public accountability

g) Publishing data on victims’ right to review – and a short yearly report on the insight and learning from these cases

h) Advice to PCCs on oversight and the kinds of questions to ask when exploring best service to victims

i) Aspiration that all the above will begin to build community support and sanctuary – services for healing and crime prevention

9. Whole measure of success – public sector benefits and public costs – which public body will do this in conjunction with Operation Soteria Bluestone

You can contact the Operation Soteria Bluestone central team at operationsoteriabluestone@mopac.london.gov.uk

National Police Chiefs’ Council: Chief Constable Sarah Crew

NPCC Staff Officers: Inspector Catherine Larsen and Detective Sergeant James Kent

Central Team

Robin Merrett – National Programme Director

Lizzie Peters – Head of Programme Management

Professor Betsy Stanko – Strategic Advisor & Academic Lead

Professor Katrin Hohl – Academic Lead

Alice Light – Head of Communications

Ali Winn – Learning & Development Lead

Becca Potton – Programme Officer

Delisa Trought – Programme Support Officer

Henry Herrera – Programme Support Officer

Dr Patrick Tidmarsh – Lead, ‘Whole Story’ learning approach

Police Force Senior Responsible Officers

Durham Constabulary – Chief Superintendent David Ashton

South Wales Police – Superintendent Philip Sparrow

West Midlands Police – Superintendent Annie Miller

Avon and Somerset Police – Superintendent Edward Yaxley

The Metropolitan Police Service – Commander Melanie Dales and Superintendent Mel Laremore

Police Force Project Managers

The Metropolitan Police Service – John S Steel

Avon and Somerset Police – Ruth Stevens

West Midlands Police – Christopher Jones

South Wales Police – Amy Thomas

Durham Constabulary – Andrew Woodward

Universities

City, London University

University of Suffolk

Bournemouth University

Loughborough University

Middlesex University

Glasgow University

Birkbeck, University of London

Centre for Police Research and Learning at the Open University

London Metropolitan University

Pillar One – Academic Team

Professor Miranda Horvath

Professor Emma Bond

Dr Maria Cross

Dr Mark Manning

Dr Ruth Spence

Aneela Khan

Thistle Dalton

Dr Anna Gekoski

Kristina Massey

Dr Joana Ferreira

Dr Katherine Allen

Dr Patrick Tidmarsh

Arianna Barbin

Pillar One – Police Pillar Leads

Avon and Somerset Police – Detective Inspector Andrew Branch

Metropolitan Police Service – Detective Chief Inspector Kamal Patel (lead), Detective Inspector Jeremy McDermott (deputy lead)

Durham – Superintendent Yvonne Dutson

West Midlands Police – Detective Inspector Gagy Bedi

South Wales Police – Detective Inspector Dan Sweeney

Pillar Two – Academic Team

Dr Kari Davies

Professor Jessica Woodhams

Asmaa Majid

Ioana Crivatu

Rosa Heimer

Anca Illiuta

Margaret Hardiman

Sophie Barrett

Louise Trott

Pillar Two – Police Pillar Leads

Avon and Somerset Police – Chief Inspector Martyn Cannon

Metropolitan Police Service – Detective Chief Inspector Tim Mustoe (lead), Detective Inspector Ross Morrell (deputy lead)

Durham Constabulary – Superintendent Yvonne Dutson

West Midlands Police – Sgt Paul Shannaghan

South Wales Police – Detective Inspector Chris Evans

Pillar Three – Academic Team

Dr Kelly Johnson

Dr Oona Brooks-Hay

Dr Olivia Smith

Dr Andy Myhill

Professor Katrin Hohl

Sophie Geoghegan-Fittall

Dr Rosa Walling-Wefelmeyer

Sarah Molisso

Dr Susan Hillyard

Dr Beth Jennings

Adrian Harris

Pillar Three – Police Pillar Leads

Avon and Somerset Police – Detective Chief Inspector David Lewis

Metropolitan Police Service – Detective Chief Inspector Tracey Cormack (lead), Detective Inspector Gail Steele (deputy lead)

Durham Constabulary – Detective Constable Nicola Lawrence

West Midlands Police – Detective Inspector Chris Jones

South Wales Police – Detective Inspector Gareth Davies

Pillar Four – Academic Team

Dr Emma Williams

Dr Linda Maguire

Dr Nicky Miller

Jennifer Norman

Dr Rachel Ward

Dr Arun Sondhi

Richard Harding

Daniela Abinashi

Ria Williams

Pillar Four – Police Pillar Leads

Avon and Somerset Police– Chief Inspector Jason Shears

Metropolitan Police Service – Detective Chief Inspector Nicola Franklin (lead), Detective Inspector Kleo Papachristou (deputy lead)

West Midlands Police – Detective Sergeant Barbara Martin

Durham Constabulary – Superintendent David Cuthbert and Detective Sergeant Kezi Smith

South Wales Police – Detective Chief Inspector Andrew Bartholomew & Detective Constable Jenna Hargraves

Pillar Five – Academic Team

Jo Lovett

Professor Liz Kelly

Dr Fiona Vera-Gray

Gavin Hales

Dr Gordana Uzelac

Dr David Buil-Gil

Dr Andy Myhill

Dr Maria Garner

Pillar Five – Police Pillar Leads

Avon and Somerset Police – Jon Dowey

Metropolitan Police Service – Detective Chief Inspector David Kemp (lead), Detective Inspector Allison Upchurch (deputy lead)

Durham Constabulary – Fiona McGinn

West Midlands Police – Detective Inspector Mark Cooper

South Wales Police – Detective Inspector James Howe

Pillar Six Lead – Academic Team

Tiggey May

Dr Rachel Skinner

Dr Catherine Talbot

Samantha Atkinson

Chantelle Butt

Nicola Campbell

Emilly Holtham

Tamzin Jeffs

Amrana Latif

Georgie Markham-Woods

Tiago Garrido Marques

STAR Digital Forensics Project – Academic Team

Dr Kari Davies

Tiggey May

Samantha Atkinson

Arianna Barbin

Chantelle Butt

Emilly Holtham

Tamzin Jeffs

Amrana Latif

Asmaa Majid

Georgie Markham-Woods

Tiago Garrido Marques

Elena Reid

Dr Rachel Skinner

Dr Catherine Talbot

Louise Trott

Police Pillar Leads for STAR Digital Forensics Project:

Bedfordshire Police – (STAR) Detective Chief Superintendent Julie Henderson

Norfolk – (STAR) – Detective Inspector Patrick Thompson

Pillar Six – Police Pillar Leads and leads for STAR Digital Forensics Project:

Avon and Somerset Police (STAR and Pillar Six) – Samantha Strong

Durham Constabulary – (STAR and Pillar Six) Detective Inspector Luke Terry

South Wales Police – Detective Inspector Dale Pope

Appendix 4: The National Picture: A wider context using Home Office published data^{[footnote 47]}

1. There has been a four-fold increase in rates of police recorded rape nationally since 2012/13, although the scale and timing of increases varies between police force areas. There are also large differences in rates of police recorded rapes between forces. Operation Soteria Bluestone is framed in the national picture of recorded rape and other sexual offences. In 2021/22, rates of police-recorded rapes ranged from 72 per 100,000 residents in Surrey to 171 per 100,000 in the West Midlands (2.4 times higher). The rate for England and Wales as a whole rose from 29 per 100,000 in 2012/13 to 118 per 100,000 in 2021/22.However, estimates from the Crime Survey for England and Wales indicate that “the prevalence of rape or assault by penetration (including attempts) has remained relatively stable over the last 15 years”[^48).

Figure 17: Rapes recorded by police forces in England and Wales per 100,000 population. Financial years 2012/13 to 2021/22.

Sources: Home Office Police recorded crime open data, Police Force Area tables from year ending March 2013 onwards. Mid-year population estimates to 2016 received from the Home Office (personal correspondence); 2016-20 from ONS, Crime in England and Wales: Police Force Area data tables, March 2022 edition.

Data for this chart was not available to publish accessibly. Figure 17 illustrates a line graph of Rapes recorded by police forces in England and Wales per 100,000 population. Financial years 2012/13 to 2021/22. It demonstrates that In 2021/22, rates of police-recorded rapes ranged from 72 per 100,000 residents in Surrey to 171 per 100,000 in the West Midlands (2.4 times higher). The rate for England and Wales as a whole rose from 29 per 100,000 in 2012/13 to 118 per 100,000 in 2021/22.

2. The overall composition of all sexual offences – and rape and other sexual offences specifically – recorded by the police varies nationally, as therefore does the nature of demand. In 2020/21, 43.2% of all sexual offences recorded by West Midlands Police were classified as rapes (highest nationally), compared with 37.6% for England and Wales as a whole and 29.1% in North Wales (lowest nationally, apart from 4.7% on the rail network (British Transport Police).

Figure 18: Rape vs all other sexual offences recorded in England and Wales, 2020/21.

Source: Home Office Crime Outcomes in England and Wales Open Data, Year Ending March 2021 (21 July 2022 update)

Data for this chart was not available to publish accessibly. Figure 18 shows a bar graph outlining the proportion of rape versus all other sexual offences recorded by forces in England and Wales, 2020/21. It shows that the overall composition of all sexual offences – and rape and other sexual offences specifically – recorded by the police varies nationally, as therefore does the nature of demand. In 2020/21, 43.2% of all sexual offences recorded by West Midlands Police were classified as rapes (highest nationally), compared with 37.6% for England and Wales as a whole and 29.1% in North Wales (lowest nationally, apart from 4.7% on the rail network (British Transport Police). This figure is described in paragraph 2 of appendix 4

3. Nationally, rape charge rates for a given year increase over time, given extended average timescales for rape investigations, especially where charges are authorised. Taking the example of rapes recorded in the year to March 2021, charge rates increased from 1.51% to 2.99% over the course of the year from July 2021 to July 2022, as more investigations were concluded and the proportion awaiting an outcome fell from 32.5% to 12.8%.

Figure 19: Charge and ‘awaits outcome’ rates for rapes recorded in England and Wales in 2020/21.

Source: Home Office Police Recorded Crime and Outcomes Open Data Tables: Outcomes open data year ending March 2021. Update versions: July 2021, November 2021, January 2022, April 2022 and July 2022.

Data for this chart was not available to publish accessibly. Figure 19 is a line graph that illustrates the charge and ‘awaits outcome’ rates for rapes recorded in England and Wales in 2020/21. It shows that nationally, rape charge rates for a given year increase over time. Taking the example of rapes recorded in the year to March 2021, charge rates increased from 1.51% to 2.99% over the course of the year from July 2021 to July 2022, as more investigations were concluded and the proportion awaiting an outcome fell from 32.5% to 12.8%.

This chart is described in paragraph 3 of appendix 4

4. Nationally, there are wide variations in rape and other serious^{[footnote 49]} sexual offence charge rates by police force, confounded somewhat by differences in the time taken to conclude investigations and therefore differences in the proportion of cases that remain open at a given point in time.^{[footnote 50]} For rapes recorded in 2020/21, as of July 2022 the average charge rate was 3.0%, but at the police force level it ranged from 0.0% in City of London and 1.0% in Gloucestershire, to 7.1% in Lancashire and 8.5% in Durham.

Figure 20: Rapes recorded in England and Wales in 2020/21: charge and ‘awaits outcome’ rates as at July 2022

Source: Home Office Police Recorded Crime and Outcomes Open Data Tables: Outcomes open data year ending March 2021. Update version: 21 July 2022.

Data for this chart was not available to publish accessibly. Figure 20 shows a bar graph of rapes recorded in England and Wales in 2020/21 and the ‘awaits outcome’ rates at July 2022 broken down by police force. It shows that the average charge rate was 3.0%, but at the police force level it ranged from 0.0% in City of London and 1.0% in Gloucestershire, to 7.1% in Lancashire and 8.5% in Durham.

This chart is described in paragraph 4 of appendix 4

Examining rape and other serious sexual offence charge rates at force level, they are only weakly correlated (r2=0.24). In general, serious sexual offence charge rates are higher – in most cases a lot higher – than rape charge rates, with the single exception (in 2020/21) of Durham. Comparing Durham and the Metropolitan Police, for example, while there is a large difference in rape charge rates (8.5% vs 2.5%), the gap is much smaller for other serious sexual offences (8.1% vs 6.9%).

Overall, 9.0% of other serious sexual offences recorded in 2020/21 were awaiting an outcome as of March 2022 (not shown in the chart), compared to 12.8% of rapes recorded that year, at the same point in time.

Figure 21: Rape and other serious sexual offences recorded in England and Wales in 2020/21, charge rates as at July 2022

Source: Home Office Police Recorded Crime and Outcomes Open Data Tables: Outcomes open data year ending March 2021. Update version: 21 July 2022.

Data for this chart was not available to publish accessibly. Figure 21 shows a bar graph broken down by police force area demonstrating rape and other serious sexual offences recorded in England and Wales in 2020/21, charge rates as at July 2022. It shows that in general serious sexual offence charge rates are higher than rape charge rates.

Figure 22: Rape and other serious sex offence charge rates by police force, rapes recorded in 2020/21 as at July 2022.

Source: Home Office Police Recorded Crime and Outcomes Open Data Tables: Outcomes open data year ending March 2021. Update version: 21 July 2022.

Data for this chart was not available to publish accessibly. Figure 22 illustrates a scatter chart for rape and other serious sex offence charge rates by police force, rapes recorded in 2020/21 as at July 2022.

5. National data show interaction of charge and prosecution conviction rates: high charge rates can be eroded by low prosecution conviction rates, and vice-versa. Combining police and CPS data for 2019/20 to 2020/21 it is possible to calculate an overall conviction rate, combining an indicative charge rate (number of CPS decisions to charge across the three years divided by the number of police recorded rapes) and the percentage of CPS prosecutions that resulted in convictions. Nationally, a 3.34% indicative charge rate and 69.6% prosecution conviction rate give an overall conviction rate of 2.32% (3.34*0.696=2.32). Charting the data for all forces, it can be seen that Northumbria and Sussex (along with the likes of Warwickshire and Northamptonshire) ended up with very similar indicative overall conviction rates for rape, despite the former have a charge rate (4.8%) that was 1.7 times as high as the latter (2.9%). This was because a much smaller proportion of Northumbria’s rape prosecutions across the three years resulted in a conviction (52.7%) than was the case for Sussex (80%).^{[footnote 51]}

Figure 23: Examining the relationship between CPS charge decisions as a % of total police recorded rapes, and prosecution conviction rates – by Police Force Areas. England and Wales, 2019/20 to 2020/21 inclusive.

Sources: CPS VAWG Annual Data Tables, years ending March 2020 and March 2021. Home Office Police Recorded Crime and Outcomes Open Data (as at July 2021).

Data for this chart was not available to publish accessibly. Figure 23 illustrates a scatter chart examining the relationship between CPS charge decisions as a percentage of total police recorded rapes, and prosecution conviction rates by Police Force Areas in England and Wales. It shows there is much variation across forces in relation to charge decisions and prosecution outcomes with many forces experiencing higher charge rates and lower convictions rates and vice versa

Appendix 5: Project Bluestone key observations (Avon and Somerset Star Funded Pilot, Spring 2021)

• An approach to investigation that led to suspicion of victims, which was embodied in the decision-making process and language used to describe them in data recording systems, outcomes frameworks and among officers themselves;

• A lack of focus on the behaviours of suspects, poor understanding of sexual offending and limited investigative curiosity;

• Poor prioritization of investigative actions and a lack of critically reflective decision-making within investigations;

• Inconsistent and poor use of intelligence and information to establish whether named suspects had a previous criminal history, a poor understanding of what constituted a repeat suspect and a lack of influence of the aforementioned investigative strategy;

• Large delays in investigations, both in terms of speaking to suspects and in the collection of time-sensitive evidence;

• A lack of an institutional, holistic approach to disrupting repeat offending, that went beyond the immediate case being investigated, for example, little use of long-term measures to disrupt repeat offending such as referrals to offender management units;

• Poor safeguarding and risk assessment and management that left victims at risk of further harm and did not consider the potential for reoffending;

• Poor coordination with the CPS that led to further delays in investigation to charge decisions, as well as limited joint working with other police units;

• Poor use of NICHE (the crime reporting data system), including a lack of effective data recording of key demographics to explore fair treatment across different categories of victims;

• No mechanisms for identifying either good or poor practice because of limited case review and no or few opportunities for reflective practice;

• Limited opportunities to apply and embed learning from rape and other sexual offences into practice;

• Lack of experienced mentors (a common problem across policing in England and Wales);

• Lack of an iterative professional learning process to enable officers to gain competence and expertise about sexual offending and sexual offences and how sexual offending might be adapting to the digital world. Continuous Professional Development was poor and any integration of insight from academic research and bespoke evidence derived from local data analysis was lacking;

• Poor and inconsistent supervision both in relation to the case progress, officer well-being checks and caseload management;

• A lack of a ‘critical analysis’ approach and prioritization of actions in on-going investigations which resulted in a process-action, strategic evidential approach, i.e., actions being worked through with little consideration of their necessity or urgency in terms of building a whole case to document the rape or sexual assault;

• Lack of tactical intelligence and criminal intelligence analysts to support investigation, diminishing the efficiency gains from enabling other non-warranted police staff to support rape investigations;

• Significant missing data and quality issues around key variables to understand as broadly as possible the ‘problem’ of rape as reported by victims;

• Lack of strategic/performance analysts to understand case progression through the experiences of victims and the broad range of contexts within which rape and sexual offences occur, in a manner that allows the police service to embed on-going review of police investigation practices with an eye to continuous improvement;

• Limited and irregular production of strategic problem profiles on the kinds of rape and other sexual offences reported to the force, failure to provide a strategic overview of reported rape and sexual offences, and failure to support the better understanding of the nature and progression of RAOSO offences in the force area. This information is crucial for partnership working and routine reviews of resourcing.

Appendix 6: National Learning Netword attendance figures

The attendance numbers for the National Learning Network (NLN) Deep Dive Findings Events were as follows:

• Metropolitan Police Service 327 (plus 125 downloads)

• Durham Constabulary 323 (plus 50 downloads)

• West Midlands Police 434 (plus 114 downloads)

• South Wales 383 (downloads too early to measure at time of publishing)

The attendance for the NLN webinars was as follows:

• Digital Forensics and Social Media 216 (plus 81 downloads)

• Transforming Forensics 198 (plus 39 downloads)

• Digital Material and Technology in RASSO 338 (plus 58 downloads)

The number of practitioners who are members of the Operation Soteria Bluestone Knowledge Hub site are:

• 600. However, this may be an underestimate.

The number of downloads of Dr Patrick Tidmarsh’s masterclass series is:

• 7,642. There were 934 live attendees at masterclass 1 but attendee numbers at masterclasses 2-5 are unknown.

• Download figures do not reflect actual viewings as the recording can be saved in force shared folders – and shared, and/or published on force intranet sites.

Appendix 7: Pillar One – suspect focused investigations: End of Year One Report

Pillar team

Pillar leads: Professor Miranda Horvath and Dr. Kari Davies.

Pillar team (in alphabetical order): Dr. Katherine Allen, Arianna Barbin, Sophie Barrett, Professor Emma Bond, Ioana Crivatu, Dr. Maria Cross, Thistle Dalton, Joana Ferreira, Dr. Anna Gekoski, Rosa Heimer, Anca Iliuta, Aneela Khan, Margaret Hardiman, Asmaa Majid, Dr. Mark Manning, Kristina Massey, Dr. Ruth Spence, Louise Trott, and Professor Jessica Woodhams

Background

The extent of sex offending and sexual victimisation

Sex offending and sexual victimisation are widespread^{[footnote 52]}. According to the Crime Survey for England and Wales for the year ending March 2020 an estimated 773,000 adults were victims of sexual assault in the previous year, ^{[footnote 53]} and the prevalence of this type of offending can be seen worldwide^{[footnote 54]}. The prevalence of sex offending is likely to be an underestimate of the true extent of this issue as a result of issues relating to underreporting and the normalisation of problematic sexual behaviours^{[footnote 55]}. The importance of the victim experience has – rightly – been the subject of much of the academic literature focus in this area, borne of the fact that police investigations continually focus on issues of victim credibility and contain biases about victims of sex offences. While exploring victim experience is important, focusing our efforts on sex offenders – who they are, what types of offences they commit, and why they commit them, must remain a priority in the research agenda and within police investigations. This includes furthering our understanding of sex offender behaviour and how they use grooming, manipulation, and coercion to both commit their offences and evade police scrutiny^{[footnote 56]}.

Sex offenders – who are they and what do they do?

While the issue of sex offending is clearly widespread and of international concern, there are some fundamental misconceptions about sex offenders that exist amongst both the general public and people working in the criminal justice system. Davies et al.,^{[footnote 57]} summarise this problem as “…misconceptions include unfounded rape myths and stereotypes which serve to wrongly narrow the definition of what constitutes rape into a ‘real rape’ stereotype^{[footnote 58]} and place unfair blame on the victims themselves^{[footnote 59]}. These myths perpetuate the belief that a ‘true’ sexual assault can only involve a stranger offender, carrying a weapon, attacking victims in dark alleyways^{[footnote 60]}.” Most sex offending is perpetrated by someone known to the victim^{[footnote 61]}, very often a partner or ex-partner^{[footnote 62]}, and sex offences are most commonly committed inside the victim or perpetrator’s home^{[footnote 63]}. Sex offending is gendered; the majority of sex offence perpetrators are male, and the majority of victims are female^{[footnote 64]}. Whilst the majority of sex offences against both men and women are perpetrated by men, men can be victims^{[footnote 65]} and women can be perpetrators^{[footnote 66]}.

As touched on above, there are also some common misconceptions about the behavioural aspects of sex offending. The prevalence of violence, for instance, is often likely to be overestimated, as is a victim’s propensity to physically resist an offender^{[footnote 67]}. In reality, most sex offenders do not use violence^{[footnote 68]} and victims are more likely to cooperate or freeze rather than fight off an offender^{[footnote 69]}. The available evidence speaks to a worrying misunderstanding of the nature of sex offending which only serves to perpetuate the idea that an offence is ‘real’ under certain circumstances, thus perpetuating issues with offence disclosure, reporting, and victims being believed or even understanding that what happened to them was a crime.

Tidmarsh^{[footnote 70]} has led the field in pulling together understanding of the strategies and tactics used by sex offenders, drawing on the evidence from grooming^{[footnote 71]} (traditionally only applied to understand adult sex offenders’ tactics to access children to offend against) and domestic abuse (concepts like coercive control^{[footnote 72]}), and explains that^{[footnote 73]} ”although different grooming methods are used in adult and child cases, the same principles apply: [they] create a relationship of power, control and/or authority over the intended victim, then move the relationship into a sexualise frame. Grooming is present in every offence, however quickly it occurs”. This approach has been adopted by Victoria Police (Australia) who now recognise that grooming can occur with both adults and children and use a broad definition which includes two stages 1) non-sexual and 2) sexual: “the process by which a person prepares a victim for, and strengthens, a pattern of sexual abuse”^{[footnote 74]}.

Offenders target potential victims using both individual vulnerabilities (e.g. being young, having disabilities or mental health problems, having consumed alcohol or drugs) and social contexts to their advantage. Offenders also create vulnerability; Tidmarsh^{[footnote 75]} gives examples such as using information about someone to create situations offenders can take advantage of or exacerbating someone’s existing feelings of lack of confidence. Policing has learned, for example in its work on child sexual exploitation (CSE), that offenders often choose victims to target based on their lack of resilience, and as such it is vital that it is the suspect’s behaviour that is questioned and challenged robustly. This same strategy is used by sex offenders against adult victims, not just in CSE contexts. Often, victims provide police with the names of suspects, but this sometimes leads to investigators exploring only the legal evidence of ‘consent’ rather than exploring what the suspect did and what a reasonable person can observe which undermines the ability of the victim to consent.

As alluded to above, research on sex offenders is limited compared to that on victims. The existing evidence base on sex offenders it is primarily from research on criminal justice records, and often only contains data on convicted sex offenders^{[footnote 76]}. We know, however, that given the societal issues outlined above that make it difficult for victims to report these offences, or for justice to be accessed by them when they do which includes widespread investigative failings, only a few offenders are charged or convicted. Further, research often focuses on samples of offences containing one victim type (e.g. stranger adult females^{[footnote 77]}). While this is understandable from a methodological viewpoint, with small sample sizes of the less common offences and offenders making study is in this area difficult, it further serves to present a one-dimensional and incomplete picture of the scope and breadth of sex offending (see Pillar Two for more detail and discussion).

Psychological theories of sex offending

We can look to psychological theory to understand why sex offenders behave the way they do. A recent review of the most prominent theories of sex offending by Cording and Ward^{[footnote 78]} highlighted that theories are vital for developing “effective primary and secondary prevention and harm reduction efforts” (p.124). A useful framework^{[footnote 79]} for understanding the variety of sexual offending theories distinguishes between different levels of theory. As Cording and Ward^{[footnote 80]} summarise “No level of theory is to be understood as more important than another; instead, theories across different levels work synergistically to develop a comprehensive understanding of the factors and processes that cause and maintain sexually harmful behaviour across time and context.” Sex offending behaviour is influenced by internal (e.g., fantasy, offender age and gender, their experience and expertise and mental disorder) and external factors (number of offenders, cultural aspects of behaviours, one off versus serial offending, situational variation)^{[footnote 81]}. Understanding these underpinning elements of sex offending is essential for police officers.

The effects of ingrained cultural attitudes on investigative decision making

The effects of rape myths and stereotypes, and of these inaccurate and ingrained cultural attitudes, can be seen beyond their effect on victims. There are numerous documented examples of failures of police investigations e.g., John Worboys and Kirk Reid, where a lack of awareness and understanding of sex offender behaviour coupled with ingrained attitudes that victims lie (early reports were ‘no crime’^{[footnote 82]}) and have led to internal reviews by the police^{[footnote 83]}. Historically and currently police investigations of rape and other sexual offending (RAOSO) focus on victims – their veracity and their credibility – and rely on myths about what constitutes real rape^{[footnote 84]}. This is despite evidence, for example Waterhouse et al.^{[footnote 85]}, which used police data and found that out of 400 cases of rape reported none had all the characteristics of the ‘real rape’ myth but the majority (70.7%) were committed by people known to the victim, occurring inside a home, with most victims having no physical injuries. False allegations of RAOSO are perceived as occurring at high levels despite evidence which suggests they are no higher than in any other crime^{[footnote 86]}, with a recent meta-analysis of studies from western countries estimating about 5% of rape allegations are false^{[footnote 87]}. The often-sole focus on victim credibility in RAOSO investigations draws on lines of enquiry criticised by academic research over decades^{[footnote 88]} creates ‘conditions of virtual impunity for predatory men’^{[footnote 89]} and is one of the drivers of poor justice outcomes. Numerous government reports and studies^{[footnote 90]} have highlighted there is a vital need for specialists at all stages of the criminal justice process; that investigations require improvement and better monitoring of cases is needed. Walker et al.^{[footnote 91]} recently highlighted that the system is in a perpetual cycle, as Jan Jordan^{[footnote 92]} put it, of accepting ‘the rhetoric of reform while the underlying realities remain little changed’ (p.234). RAOSO have not been prioritised or taken seriously, and when they are reported they are met with poor responses. In summary, the evidence available all points to a system that is failing and needs radical overhaul.

Refocusing attention on suspects

There is clear evidence to suggest that our society’s current views relating to women, to sex offence victims, and to the nature of sex offending itself, are adversely affecting the ability for the criminal justice system to work in an efficacious way. Given the importance of this task, both in tackling offending to safeguard the public, and in terms of providing access to justice for victims, law enforcement needs to be ‘groundbreakers’ in the way that they investigate these offences. Investigations, in essence, need to move away from the problematic societal attitudes that have thus far shaped the way that sex offences are perceived and dealt with. In 1990 Diana Scully^{[footnote 93]} wrote about interviewing convicted rapists to understand ‘what do men gain from rape?’, she concluded then that rape is ‘a low risk, high reward crime’. If suspects do not become the focus of police investigations the answer may still be the same in another 30 years. It is clear that where investigations are currently lacking is that officers are not being equipped by their organisations to carry out their role with the specialist knowledge and skills which are required to move beyond these problematic societal attitudes to be able to effectively investigate this type of crime. Initial evidence has identified that specialised knowledge and skills, including understanding sex offenders’ behaviours, are essential for improving investigations of RAOSO^{[footnote 94]}, and this requires urgent review and attention if policing practice is to improve in this regard.

We have conceptualised a suspect focused investigation in line with Hohl and Stanko^{[footnote 95]} who stated “the investigation must begin by examining the suspect’s offending behaviour early in of the investigation, rather than focussing on the victim as the first and primary site of the investigation.” The key elements of this approach include foregrounding:

• The actions of the suspect before, during and after the offence and their explanations for those actions;

• The circumstances and context of the offence with specific emphasis on whether the suspect targeted, groomed, exploited, manipulated, or exerted control or coercion over the victim;

• The relationship between the suspect and victim; and

• Awareness that the suspect may use or attempt to use grooming techniques such as manipulation, coercion, and control on officers.

The purpose of Pillar One

To investigate how investigations are (or are not) currently taking a suspect focused approach, with a mind to providing recommendations and outputs to improve the investigative process.

Methodology

Data collection for Pillar One was shared with Pillar Two, due to the multiple points of convergence in terms of the information required to answer the research questions. Exploratory research was conducted in four police forces using a mixed methods approach as a way of obtaining depth and breadth of information and compensating for any weaknesses in a single research design^{[footnote 96]}. Using a mixed methods approach draws on the strengths (and minimises limitations) of both quantitative and qualitative approaches and can provide innovative opportunities for understanding knotty issues such as the investigation of RAOSO^{[footnote 97]}. Taking a mixed methods approach a gave us a more rounded view of RAOSO investigations and RAOSO investigators; considering data from different sources (triangulation^{[footnote 98]}) collected using different methods allowed us to compare and contrast the theory of what is supposed to happen in an investigation and how investigators should behave and the reality of what is happening, which was essential as there are very few internal accounts of why so many RAOSO investigations go wrong.

Exploratory research questions were:

• Understanding the efficacy of RAOSO investigations including whether the focus on the investigation was predominately on the victim or the suspect and how this affected the course of the investigation.

• How much specialist knowledge officers and police staff have around targeting suspects using a suspect focused approach.

• What are officers’ perceptions of investigative efficacy and the utility of a suspect focused approach?

• What guidance is provided to officers and how this relates to conducting (or not) a suspect focused approach?

• What are officers’ perceptions of the changes implemented within the pathfinder Force E and have the changes led to more suspect focused investigations?

The outline of the methodology provided here highlights how each data strand was used to answer specific research questions relating to Pillar One.

Strand 1 – Case reviews used to address research questions 1, 2, 3 and 4

Non-probability quota sampling was used to identify cases for review. 72 closed cases dated between 2019 and 2022 were requested from each force. Officers identified the cases and aimed to include a sample comprised of an equal number of adult rapes involving a stranger, acquaintance, or domestic relationship, and that either had a charge, the victim does not support prosecution , or no further action outcome (in line with current Home Office outcome codes, see Pillar Five for more discussion of these). Officers conducted the case reviews, giving them opportunity to obtain insights on how investigations were being handled in their force as well as developing the critical reflection skills (see Pillar Four). Officers received training on how to conduct case reviews from the researchers and the researchers reviewed the case reviews focusing on consistency and accuracy as well as content. First reviews were conducted by DCs (also DSs in Forces A, B) and second reviews were conducted by DIs, DSs (Forces C, D), DCIs, Supts, Chief Supts, and the ACC. A total of 215 case reviews were completed ranging from 33 to 59 per force. An inductive approach was used as the coding framework, with content analysis used for the qualitative section of the reviews. Descriptive statistics were derived from the quantitative data on timeliness of investigative milestones.

Strand 1a – Development of case reviews for use by researchers in pathfinder force E used to address research question 5

Pathfinder Force E were already implementing changes based on our recommendations. The case review tool described in Strand 1 was developed for use by researchers so we could assess the effect of the changes on investigations. 13 case reviews were completed by researchers who had access to Force E’s crime recording system on force issued laptops. 7 of those were selected based on victim-perpetrator relationship and outcomes from an excel list of cases provided by Force E. The other 6 cases were selected from those provided by each Force E Bluestone team Detective Inspector who selected two cases, one which represented good practice and the other where improvement was needed (these included some cases requested based on victim-perpetrator’s protected characteristics, e.g. sexuality, religion etc.. The same analysis process was used as described in Strand 1.

Strand 2 – interviews used to address research questions 1, 2, 3, and 4

Qualitative, or conversational, interviews were used to explore research questions two and three. A semi-structured interview schedule was produced, to cover both Pillars One and Two, using open-ended questions, with prompts where applicable. The schedule was used flexibly, as a guide, in order to both address pre-defined areas of interest and also to allow topics to arise inductively during the conversations. Key areas covered included: officers’ experience and training; investigative challenges; intelligence used in rape investigations; named suspects; repeat suspects; how the police work with the Crown Prosecution Service (CPS) on rape investigations; and conducting suspect-focused investigations.

In reporting the results of the interviews, numerical data was not used. The use of numbers/quantification in qualitative research is a controversial issue. Some researchers argue for the use of numerical or quasi-numerical data – or quantification or semi-quantification – arguing that this can: improve transparency; add precision; enable themes to emerge with greater clarity; and increase the meaning of crucial findings^{[footnote 99]}. However, Maxwell^{[footnote 100]} notes that since the ‘paradigm wars’ of the 1970s and 80s, “many qualitative researchers have rejected the use of numerical data in their studies” for both philosophical and political reasons. Philosophically or conceptually, it is argued that using numbers in qualitative research is not compatible with a constructivist stance. Thus, to evaluate qualitative research in a scientific framework in which there is seen to be a single reality, that can be objectively measured and quantified in order to reach valid and generalisable conclusions, is not appropriate. Using numerical data in qualitative research can superficially appear to make a report more ‘scientific’ or ‘rigorous’ but fails to meaningfully contribute to its logic; evidence should not be reduced to the amount of evidence. Thus, as argued by Pyett^{[footnote 101]}, “counting responses misses the point of qualitative research”, as frequency does not determine value. More practically, or on an applied level, if all participants in qualitative research have not been asked exactly the same questions in the same way, reporting frequencies of a particular theme or response risks misrepresenting the data, even within the sample studied. Moreover, quantification may detract from the nuance and detail that are key in qualitative work^{[footnote 102]}.

With these arguments in mind, reflexive thematic analysis (RTA) was used in this research, which is a purely qualitative approach that draws upon appropriate qualitative paradigms such as constructivism^{[footnote 103]}. RTA embraces, rather than rejects, subjectivity, flexibility, and creativity, taking an organic and reflexive approach. Rather than striving for replicability and consensus, it encourages the researcher’s role in active knowledge production and rejects the idea of ‘accurate’ or ‘reliable’ coding and interrater reliability. The six phases of conducting RTA were followed, which include: (1) the researcher familiarising themselves with the data; (2) the generation of codes; (3) the construction of themes; (4) the reviewing of potential themes; (5) defining and naming themes; and (6) writing up and presenting the themes.

Strand 2a – Follow up interviews in pathfinder force E used to address research question 5

Between February and June 2022, the same approach was used as outlined in Strand 2 with 8 officers from pathfinder Force E. A semi-structured interview schedule was produced, to cover both Pillars One and Two, using open-ended questions, with prompts where applicable. The interviews focused on looking at how the recommendations from the pathfinder pilot had been implemented and officers’ perceptions and experiences of the new approach. Interviewees were Engagement Officers and Disruption Officers, including a Detective Sergeant of each and one Detective Inspector (all from Bluestone pilot areas in Force E). The interviews were analysed the same way as described in Strand 2.

Data insights and limitations

There are some limitations to the data which need to be acknowledged:

• The case reviews were reliant on participants accurately and fully completing the data spreadsheet provided to them. Where data are missing, it is possible it is because the data were not completed properly.

• Equally, the data represented in the case reviews is a reflection only of the data recorded in the forces’ crime recording systems. While officers should note down all actions taken in an investigation, there is no guarantee that this is done (see Pillar Five report for evidence data are not properly recorded), and therefore analyses may include incidents that have been completed, but not recorded.

• There was some variability in the quality of the case reviews, some officers clearly found the approach easier than others.

• The interviews were arranged by the Police Pillar Leads in each of the forces. As a result, they may have been subject to selection bias, over and above the selection bias normally associated with a snowball or mass approach to participant engagement.

• Likewise, the information requests were sent out and documents for review collected by the Police Pillar Leads, which again may have led to bias.

The approaches to data collection we used combined traditional well tested methods such as interviews but also methods which were being trialled for the first time, case reviews. Case reviews were a unique source of data. Information about case progression itself is usually prohibitively difficult to obtain due to the need to manually extract it from crime recording systems and the time and security constraints associated with this. The novelty in the case reviews came from having officers conduct them and having two officers of different ranks and experience review the same case providing insights into their perceptions of investigations. Using this approach enabled us (and the officers who conducted the reviews) to see and diagnose what is going wrong in investigations and were an integral part of assisting forces in assessing the efficacy of their own investigations. Furthermore, many officers reported that whilst doing the case reviews was time consuming, the insights they provided were enlightening.

Findings

A lack of specialist knowledge

Across all forces – albeit to differing degrees – the workforce investigating RAOSO tended to be young and often inexperienced. It was reported that officers could be thrown into investigating rape – sometimes with no investigative background broadly, or no background in investigating rape specifically. For example, direct entry detectives, response officers or officers straight from uniform, were filling gaps on rape investigation teams. This was felt, particularly by more senior detectives, to be inappropriate at best and dangerous at worst.

[Force A] is fast tracking detectives at the moment, and it unsettles me greatly … there are offices that are on the fast track detective scheme that are investigating serious sexual assault … The thought of something happening to me or someone close to me, and then they’re being investigated by somebody with six months’ service. I’m just like, that’s incredible … it just it just makes me cringe a little bit

(FORCE A7)

I don’t think the makeup of our teams is effective. As an organization, we have lost an awful lot of experienced staff and the force is made up of very young staff now. I personally don’t think it is appropriate that we have staff put onto complex teams with very little experience of any other investigation type and then expect them to be able to deal effectively with rape and sexual offenses with minimal supervision

(FORCE C3)

In terms of length of service and experience, some of the officers are very, very young in service and haven’t got a lot of experience. They are coming into these sort of teams without PIP accreditation, without necessarily being SOIT^{[footnote 104]} trained initially, and they’re sort of catching up with that quite quickly and getting thrown into the deep end really without an opportunity to get better, and have a proper mentor, and see things through a little bit before just being thrown in the deep end

(FORCE D5)

A lack of specialist training was also found across all forces. In Force A, for example, while SOITs are given a 4-week training course, detectives investigating rape are not given any specialist sex offences training:

When you land as a detective sergeant in [SPECIALIST UNIT] … I was very surprised that I wasn’t going to get any training, because it is so different. And trying to adapt to that, I know for me and a lot of my colleagues … was very challenging. So why the detectives don’t get [training] is something of a mystery to me (Force A9)

In Force B there was also no specialist training for sex offences:

Some of the officers, particularly in safeguarding, have had awareness training around trauma – like ‘flight, fight or freeze’ – but certainly not everyone who deals with RASSO has, and that is a gap for us

(Force B8)

Training would be useful before you investigate RASSO, as it’s assumed you know what help is out there for victims. So things like what SARCs^{[footnote 105]} and ISVAs^{[footnote 106]} are and what they can do. Or DA support services, which can help with housing, and women’s refuges

(Force B10)

In Force C officers received no specialist sex offences training and reported having no time for CPD due to the sheer volume of work, lack of staff, and shift structure.

Where’s the training? Where’s the investment in the future detectives? The quality is going down because the upskilling is not there

(Force C7)

While in Force D, specialist sex offences training is not always available, or access is restricted due to learning and development capacity and individual officers’ workloads. CPD is also not formally provided – but seen as personal responsibility – and the days are in place of the older Specialist Sexual Assault Investigators Development Programme (SSAIDP) course. Officers also reported a lot of ‘on the job’ learning and a lack of formal mentoring.

There’s probably always going to be gaps in your learning, and that’s probably where your actual specialist training could come into play … The onus is on you really to just get yourself up to speed with that – speaking to colleagues in similar roles and doing your own due diligence really

(Force D6).

Our findings in this area are consistent with Pillar Four which should be referred to for further detail on this subject. Fundamentally, specialist knowledge and skills were lacking across all forces, and even where specialist training was available, there was often no time to take advantage of it or no offer to send staff on these courses, all of which resulted in poor investigations. Officers were ill-equipped to deal with sex offence investigations. RAOSO investigations can be complex, hence more appropriate learning and development which takes a trauma-informed approach and combats poor attitudes is essential. Without these, officers are simply not prepared to conduct these investigations which has a knock-on effect in several key areas below.

A reliance on ‘victim credibility’ tests which undermine the appropriate formation of effective investigative strategies

The lack of officer expertise is most notably demonstrated in the reliance on issues of victim credibility to decide what to do with an investigation. This finding corroborates the most recent HMICFRS reports^{[footnote 107]} and many other reports and investigations^{[footnote 108]}. Officers across forces talked of ‘victim credibility’ being used to determine whether cases were likely to proceed further. Victims who were potentially felt not to be credible included those who made repeat/multiple allegations, sex workers, those with mental health or substance abuse issues, intoxicated victims, victims who give ‘inconsistent’ or ‘incomplete’ accounts, and victims who may have lied in the past (in almost any area of their life). This was particularly the case in Force A.

A large proportion of our cases are people suffering with mental health issues and also kind of repeat allegers

(Force A1)

It’s often one of the very first questions that the CPS lawyer asks when you ring them up for charging advice. They will ask about the character of the victim. Quite rightly, they want to know whether or not the victim is known to the police for anything like, you know, and robbery, burglary, dishonesty type offence, which can show a propensity towards, you know, lying. But the more slightly contentious issue often comes when they also want to know about previous allegations from the victim of a similar nature. Similar fact

(Force A10)

Victims that are sex workers … I suppose that is always going to have an impact on the way you think … you know, they’re in these situations because of the job that they do … you gotta be mindful that are the allegations being made kind of because of money payments not being made

(Force A3)

The credibility of victims seemed particularly important in cases which revolved around consent.

When two people know each other – which is most of the complaints we receive – and it boils down to consent, it’s very hard to prove to the level you need. So we look at which party is the most truthful, and it’s hard to prove which party can be relied on more to tell the truth

(Force B7)

However, the notion of most cases boiling down to ‘he said/she said’ is particularly concerning when considering the number of missed opportunities found within the case reviews to progress the investigations. The use of ‘victim credibility’ within the context of an investigation was often used as an absolute factor not to proceed an investigation, rather than critically assessing whether it had any bearing on the veracity of the allegation and conducting a full and appropriate investigation of all the facts.

Complainant’s behaviour, demeanour, alcohol intoxication and possible MH issues, make allegation unlikely to have happened and no further police actions taken. Report submitted for closure by IIO on the same day.

(Force A – Case 40, Reviewer 1)

This was reflected in the case reviews, 82% of which demonstrated there were missed opportunities to progress the investigation (see section 3.3 for more on missed opportunities). This judgement of ‘victim credibility’ factors were, in fact, often issues of victim vulnerability, although it was clear from the number of missed opportunities that these factors were seen as a way of lessening the amount of investigation required, rather than critically considering how the offender may have taken advantage of the victim as a result of these vulnerabilities. In several cases that involved mental health issues, for instance, officers were quick to judge the report as a false allegation, in some cases acting against standard procedure and failing to complete basic investigative actions. For example, in one case, a victim situated in a mental health unit was interviewed by officers in front of the member of staff they alleged had assaulted them. This issue was also highlighted by reviewing officers themselves.

I feel that the OIC, DS and DI overview of this crime are biased by the mention of the victim suffering from mental health issues. Some medical/mental health records were requested but there is no log recording the decision around these or what they contained.

(Force B – Case 17676, Reviewer 1)

Whilst there are instances where officers do not rely on inaccurate perceptions of victim credibility and who do not miss investigative opportunities, the overwhelming direction of travel is still reliant on inaccurate understandings of victims and offenders. This blanket judgement of what is a ‘victim credibility’ issue means that victim vulnerability is not appropriately considered and as a consequence, victims are not adequately safeguarded or provided with appropriate support (see Pillar Three for further details):

There is nothing recorded on the investigation or victim care log to show that the victim was offered any kind of meaningful support.

(Force B – Case 15837, Reviewer 2)

It is clear these victim credibility issues are influencing officers’ decisions on how to proceed within an investigation, even where it was inappropriate to do so. At worst, officers demonstrated explicit victim blaming and lack of belief in the victim, which impacted on the subsequent investigation. For example, victim credibility was often focused on and used to either close or not investigate cases within some forces.

The OIC^{[footnote 109]} uses third party material and previous police log to undermine her account. The victim’s account appears clear and concise and there is no evidence to suggest that she is not telling the truth.

(Force C – Case 15, Reviewer 1)

Several case reviews in force B highlighted instances where the investigators did not believe the victim and used victim blaming language. Once officers decided no crime had occurred based upon victim credibility, they demonstrated a lack of investigative curiosity, missing potential lines of enquiry. In Force C, decisions were not always made in the best interest of the victim’s case but rather steered by limited resources.

I wouldn’t say that the credibility of the victim is questioned, however, what does concern me is that the victim tells the OIC in VRI^{[footnote 110]} that she believes she was the victim of a sexual assault. However, the officer forms his own view that this was not a sexual assault, and this then impacts on how the suspect is dealt with.

(Force B – Case 14112, Reviewer 2)

The decision that this was not a criminal matter was made very early within the investigation. This was evident in the attempt to have this removed from the staging database (and not record a crime) 16 hours after it was reported to Police.

(Force B – Case 11297, Reviewer 2)

The DI has recorded on the investigation log that the swabs should be submitted, and two hours later the OIC has recorded that they have informed the victim that the DI has not authorised the submission.

(Force C – Case 35, Reviewer 2)

This footage and photos of the victim’s injuries were stored on the evidence.com website but had been marked for deletion after 6 years rather than 100 years.

(Force C – Case 35, Reviewer 2)

As a result of the fundamental misunderstanding of the nature of sex offending and the focus on victim credibility, there is little scope for change. The problems highlighted in the introduction are very much still present in policing today. Myths and stereotypes are protecting offenders and negating the need for officers to consider victim vulnerability under the guise of considering ‘credibility’. We did see some emerging awareness of the problems from some officers and this was more evident in some forces than others, however, overall officers are not nearly knowledgeable enough to know how to deal with these issues.

I think it’s extremely victim based and kind of proving their integrity, if you like. I think it’s wrong … it’s so difficult though because sometimes there have been cases that actually they’ve just been lying about, perhaps lying’s not the word, but they haven’t been telling the whole truth about what happened. And it’s come out via a text message, for example, that’s been found on the victim’s phone, or it’s come through a conversation that somebody else and somebody else has told

(Force A11)

Officers need to be trailblazers, to resist myths and stereotypes about victims and perpetrators being held to a higher standard than the general population, if investigations are to be improved.

A formulaic approach to investigations with little critical reflection and lots of missed opportunities

Police officers are working largely in an environment that is driven by today which is at odds with investigating RAOSO where there are often complexities and cases may take time. Coupled with officers inadequate learning, development, experience, and resources it is unsurprising that a considered investigative strategy is often absent, and when some form of investigative strategy is present it is focused on set series of actions without consideration of whether they are the most appropriate actions to be taken.

There were several missed opportunities that led to fundamental flaws within the investigation process. In force B, for example, apart from suspect interview (conducted in 59% of cases), all other investigative milestones were present in less than half of cases, with offender risk management present in only 11% of cases. In forces B and C there was a lack of consideration to the wider risk of repeat suspects and how forensics may assist with suspect identification and case linkage. In these cases, forensics were sometimes not obtained where there was potential for them to be collected, and when they were collected, samples were often not submitted for testing at all.

No intimate nor non-intimate samples taken from suspect. Both samples were considered but reason for not taking them was given as the suspect needed an interpreter. This would in fact not have prevented police using powers under S63 PACE to take non-intimate samples, and in any event an interpreter did attend the station later. In interview suspect denied sexual contact.

(Force C – Case 35, Reviewer 2)

The complainant consents to EEK and forensic samples… It may have assisted as part of the investigating to include the rational why forensic samples were not sent off – the only time these samples are mentioned are part of the DI Rape Review.

(Force B – Case 10595, Reviewer 2)

Background checks (Police national database [PND] / Police national computer [PNC]) on named suspects were carried out in less than 50% of case reviews across most forces. As a result, there was a lack of appropriate risk management surrounding suspects. For example, background checks on identified suspects were conducted in only 38.88% of cases in force D, in 50.00% for force B, and in 75.86% for force C.

Not a great deal of evidence of suspect consideration on investigation log. There is no PND/PNC action recorded. As such it’s not clear what risk management has been conducted.

(Force B – Case 15147, Reviewer 2)

This also links to findings in Pillar Two around the appropriate identification, management, and disruption of repeat suspects.

Collection of digital material was minimal across all forces, with digital material taken from victims in 32/168 cases and from suspects in 46/168 cases. Digital material which could have corroborated the victim or suspect’s account was often collected after a delayed period of time or not at all. When digital material was collected, it was often victim-focused, ignoring evidence the suspect may hold.

The phone of the victim has been examined, but the suspect phone doesn’t appear to have been examined. The digital strategy is very one-sided on the victim and not both parties.

(Force B – Case 13224, Reviewer 2)

DS review highlights that victim phone download will be necessary as the victim and suspect have communicated, but necessity to seize suspect’s phone is not highlighted and decision is taken that a voluntary interview is appropriate. Second supervisor … highlights need to seize suspect’s phone, however, the suspect’s phone is never seized… CPS charging advice highlights that phone evidence will be crucial and CPS refer the case back to police with an action plan to re-interview the suspect about various matters including the phone messages.

(Force B – Case 6, Reviewer 2)

Some case reviews in force D highlighted instances where officers did not collect relevant digital materials or did not have the resources required for phone downloads or digital material analysis (Digital media investigators). See Pillars Five and Six for more detail.

Requesting the CCTV^{[footnote 111]} 21 days into the investigation can be a risk and may have been imperative evidence to this case which could have been lost due to the turnover timeframe in the recordings of some CCTV software. This perhaps could have been tasked to another Officer to obtain within the first few days of the investigation or requested the assistance of a CCTV retrieval team who are able to download CCTV when there are no experienced staff on site to do it.

(Force D – Case 46, Reviewer 2)

In summary, the data suggest that investigative actions were either a) not being done when appropriate, b) being done but too late, or c) being done but not meaningfully, e.g. collecting forensic evidence and then not testing it. Perhaps unsurprisingly, when senior officers were conducting the second reviews of the cases, they were often able to identify these missed opportunities and see the fundamental flaws in investigations.

The inability to appropriately manage or close cases linked to a lack of capacity and expertise

Across all forces we saw some evidence of strong starts and long drifts in investigations. From the case reviews, the average length of investigation across all forces was 4 months, with the longest case lasting 2 years and 7 months (there are issues with defining length of investigation see Pillar Five). Immediate investigative actions were often completed within the first week of the investigation and then left to drift due to delays. For example, in Force A, long-term actions (investigative actions from the first week to the end of the investigation) were present in only 40% of cases reviewed, with 20% containing neither immediate nor long-term investigative actions. In force B, there were several delays throughout each step of the investigation process that ultimately led to victim withdrawal in some cases (see Pillar Three for more on the impacts on victims).

It is quite evident that it was not felt that a crime had occurred here, and it appears that investigation paused for several weeks whilst a decision was made as to whether or not the crime should be cancelled. There was a delay in the submission of forensic samples (only done after the DI review). By the time the results of this forensic work came back the victim was in the throes of a mental health episode and unable to provide the required DNA elimination sample.

(Force B – Case 11297, Reviewer 2)

There were examples across all the forces of cases being inappropriately managed or failure to close cases. For example, in Force A officers reported some cases not being closed soon enough; instead, they were simply put on the back burner, with nobody taking the responsibility for making a decision (also found by Pillar Five). This results in poor victim service and could contribute to the high rates of victim attrition (see Pillar Three):

What happens at the moment is cases is get left open for years that never going to be charged, because no one will take responsibility for it closing it. Which is bad enough anyway, it’s just not good service for that victim by any means. It’s just we pretend we’re investigating it, and then two years later saying it’s closed, which is not great. But also stops us from investigating other stuff that actually might be able to go somewhere

(Force A8).

I’m a big believer in this case is not going to go, if it’s not going to get to court for whatever reason, then we need to we need to close it as soon as possible for the sake of the victim

(Force A9)

We have cases that are two and three years old and they’re nowhere near a prosecutor or resolution. I suppose because of the nature of the crime, supervisors find it difficult to make hard decisions … So, we keep a victim hanging on and hanging on at times

(Force C1)

We acknowledge, as officers did both in the interviews and case reviews, that a lack of capacity to deal with the volume of demand is also driving poor timing in investigations, e.g. not having time to collect time-sensitive data. This is also leading to failure demand (see Pillar Four), i.e. creating more work because officers did not have time to do it when they should have, and constantly needing to catch up.

Poor liaison and communication with the CPS

Recent HMICFRS reports^{[footnote 112]} describe the relationship between the police and CPS as needing fundamental improvement. In our deep dives across all forces, poor liaison and communication with the CPS was noted. Officers reported not being able to communicate with the CPS directly – e.g. on the phone or in face-to-face meetings – instead relying on electronic systems.

Normally it’s just electronic contact. You send it electronically and then CPS send it back electronically, with an action plan of things that they review in the case and they’ll say: ‘X, Y, Z, I need this, this, this and this’

(Force C9).

This seemed particularly the case in Force D, where prosecutors’ details – email and phone numbers – used to be displayed at the bottom of all communications. However, after moving to a new system, issues with limited word counts have meant that these have been removed, leaving officers with no way of directly communicating with the CPS.

Originally we used to have discussion over the phone with CPS and then the (NAME OF SYSTEM) system came into place and now we communicate with CPS via memos electronically

(Force D2).

This issue was also particularly acute for lower ranked officers, who rarely had prosecutor’s direct contact details.

It would be useful to be able to pick up a phone and have a discussion … It’s easier at the higher ranks but not as easy for the officers

(Force B5).

Officers across forces reported this lack of direct human contact and reliance on systems as leading to frustrations, misunderstandings and things ‘getting lost in translation’:

It’s the barrier of electronic memos … we don’t have those face-to-face discussions or over the phone discussions, it’s all in writing where things can be missed, or misunderstood

(Force D2).

The CPS were engaged with in just over a quarter of the cases that were reviewed (45 out of 168). This occurred on average 76.87 days (range: 0 – 467) after the start of the investigation, suggesting early advice was rarely sought. In cases where the CPS were engaged, significant delays occurred due to errors on behalf of the police, lack of compliance and capacity, and lines of enquiry not being pursued prior to advice. An emphasis on victim credibility rather than suspect-focused points of action were also highlighted in cases submitted to CPS.

The charging advice was delayed by a period of about 8 months as the CPS set action plans for the police which were not fully completed before the police sent the case back to CPS.

(Force C – Case 5, Reviewer 2)

No Early Advice sought, there were significant delays in the case being sent for a charging decision and EIA could have directed the forensic enquiries in an effort to streamline these enquiries for a more prompt charging decision.

(Force D – Case 64, Reviewer 1)

One of the strongest, most replicated, finding across every force, was the fond recollection of prosecutors being imbedded in police stations (also see Pillar Five). This face-to-face, single point of contact was felt to facilitate officers and prosecutors building rapport and communicating/liaising effectively; fostering good professional working relationships; and obtaining timely advice – as opposed to playing ‘memo tennis’.

I’m going to hark back to old days now: we used to have a CPS lawyer based at the police station … you used to go up and have conversations with them, easily … Things could just be ironed out by a two minute, corridor face to face question and answer session over a cup of coffee and an informal setting like in someone’s lunch break

(Force A10).

I remember the days we used to have a CPS lawyer in the police station. You could take any crime report and say: Is it worth it or is it not? Would you look to take this case if I did X, Y and Z and the outcome was X, Y and Z? Well, no, we wouldn’t. Yes, we would

(Force C6).

Other officers said they would like to see prosecutors not necessarily embedded in police stations, but co-locating with the police in the same building, in order to achieve the same ends of better communication and outcomes.

I’ve been saying for a long time that we should cohabit with prosecutors in the same building so that we could have conversations … while you’re cooking your dinner and your mate is at the microwave. If we had the human contact and a better professional working relationship, I think that charge rates would increase

(Force C1).

The belief that improved communication and relationships between the police and CPS would lead to an increase in the number of charges is supported by evidence from Force C, who recently ran a successful pilot of face-to-face early advice clinics. Of the nine cases reviewed over a five-week period – that would otherwise probably have been closed as outcome 15 (due to evidential difficulties) or outcome 14/16 (because the victim did not support police action) – it is thought that most will result in a charge.

Poor supervision

All of the factors outlined so far are reflected in the poor supervision that both supervisors themselves and other officers reported to us across all forces. This seemed particularly acute in Force C, with officers reporting having a lack of experienced supervisors, many of whom had never been investigators. Other issues included supervisors moving on quickly, higher ranks stepping down to fulfil supervisory roles, and supervisors not making crucial decisions, all of which impact on the quality of the investigations that are being conducted.

I think I’ve probably had about 12-13 supervisors in a two-and-a-half year period, which is ridiculous. And those supervisors, a lot of them have never been accredited detectives or come from a background where they’ve ever done any kind of investigations

(Force C11).

I think we have a supervisory issue again because of the volumes of workload. Our central team has, on paper, 34 staff and it is carrying 600 plus crimes. Now it is impossible, especially when we haven’t got a full complement of sergeants, for folks to effectively supervise those

(Force C3)

Some supervisors also reported not being trained properly and being overwhelmed by their caseloads (also see Pillar Five), with some carrying over 100 cases. One Force D supervisor, for example, said:

I checked the scores on the doors this morning, and I’ve got over a 100 cases. It’s ridiculous, to speak frankly. How am I expected to manage and supervise that amount? When it gets anywhere over 70, you’re really struggling to keep up to date with what’s what … People in my office are coming up to me and saying: ‘This is what’s happening, this is an update, I just got off the phone with X, Y, Z.’ And I’m looking at them blankly because I don’t know what they’re talking about. And I personally don’t like that as a supervisor. It gets me because I like my finger on the pulse, I like to do a good job, but I feel like I’m getting lost in the volume … I stress and lose sleep about it a lot … We’re having to pick and choose cases based on timelines, constraints with bail clocks, people in custody, PACE clocks. There’s a lot that has to wait, even though it shouldn’t

(Force D8)

Delays in investigations were also connected to the poor quality and inconsistency of reviews by senior officers. Irregular reviewing of cases by supervisors left investigations to drift. In several cases from Force A, investigations continued for over a year, yet only received a minimal number of reviews from supervisors, with some of the first reviews occurring months after the initial report.

Investigation runs for approx. 3 years and there is no footprint from a DCI.

(Force A – Case 11, Reviewer 2)

Where the supervisor review process was undertaken more regularly, it was often to satisfy required performance measures. In such cases the supervisor input was more regular, but consisted of minimal information with no set action plan. In these cases, senior officers failed to review all of the information at hand and left detectives to lead their own investigations with minimal guidance. This, again, directly impacts on the quality of the investigations and has consequences for staff wellbeing (see Pillar Four).

Structural issues causing tension between staff

Sometimes tensions in teams were found, with officers in different roles not always working as well as they could together, sometimes the cause of this appeared to be as a result of differing levels of experience between officers. For example, in Force A, first responders, SOITs, and detectives of different ranks/experience did not always work in a cohesive way, which could – at worst – lead to mistakes being made which might compromise the investigation, as in the examples below.

The uniform officers who are first on the scene, they don’t receive any specialised training either. Which obviously creates its own challenges in terms of the necessary support that the victims will receive upon arrival or even obtaining incorrect evidence … I’ll give you an example. Recently, there was a time when an officer, a uniform officer has turned up at a call and has provided the victim a swab to swab herself vaginally

(Force A5)

The other day, I was interviewing with someone who hasn’t even done this basic interviewing course or the detective’s course, he’s not passed the exam and you’re supposed to do that. That you’re supposed to pass before you’re considered to be a trainee detective. So he was kind of pre-training detective, and he’s supposedly the lead interview on the rape investigation suspect’s interview, which is appalling

(Force A8)

You’ve got to remember that some of the officers that are now investigating these cases have two or three years’ experience of policing and, without being rude, they’re young people, young in life experience themselves. And have probably often never even come in contact with the socioeconomic group of the victim that they’re now actually supporting. And their understanding is minimal really

(Force A10)

Change is possible

The findings presented in Sections 3.1 to 3.8 have clearly identified numerous problems in investigations. As a counter to these, officers in Force C spoke of how they have not always got it right in the past but thought that a fundamental change in how rape and sexual offences were thought about was both possible and had already begun.

I think we are a lot more aware of some of the myths and issues that we’ve had in the past … It’s a cultural shift of actually being, ‘We’re here to help you, we want to investigate.’ As opposed to – in years gone by – it was like people would have their mind made up before they went to the offence: maybe that their boyfriend’s found out, so that’s the only reason they bring it up now, or they are drunk. As I say, I think it’s certainly moving in the right direction with the specialist teams and trying to create that better understanding

(Force C6)

Further evidence of this is emerging in Force E where they have already made significant adjustments to their approach to RAOSO investigations in pilot areas, including taking a team approach and prioritising specialist knowledge. Our follow on (post-deep dive) case reviews in Force E (conducted one year after they began introducing changes) show some evidence that investigation strategies are becoming more suspect-focused. However, in some instances, there continued to be delays in completing actions around the suspect, with actions around the victim being completed first. In three cases, there was evidence of improved suspect focus regarding a proactive approach towards suspect arrest. The suspect was arrested immediately following the report, with a suspect strategy set from the outset, including consideration of bail conditions to prevent contact between the suspect and victim. This was in line with findings from our follow-on interviews, where most participants expressed this shift in attitude and investigative strategy. In some instances, lack of early suspect focus in cases, or failure to fulfil initial suspect strategy, were due to delays in assigning an OIC/Disruption Officer due to staffing issues (particularly at the start of the pilot) or the OIC/Disruption Officer being preoccupied with other cases due to high workload. More recent cases, which have not been impacted by this, were more suspect focused, and actions around the suspect were completed in a more timely way.

Sometimes investigative actions/crime recording logs varied in suspect focus based on individual officers, demonstrating the importance of having multiple officers working on each investigation. For example, there was a domestic rape, victim does not support prosecution case which generally lacked suspect focus; however, one officer and the supervisor were more suspect focused and considered the need to conduct crime recording system, PNC, and PND checks on the suspect, as well as the need to establish whether the victim was aware of any other safeguarding risks in relation to the suspect (e.g. risk posed to others). This supervisor was the first Bluestone officer to work on this case, according to crime recording system logs.

Differing from some of the first researcher conducted case reviews, the new and more recent cases reviewed (4) seemed to be more focused on the strengths of the investigation. Various lines of inquiry and available evidence were explored to build the case. This was rated as ‘good’ for three of these cases and ‘some improvement needed’ for one – where victim credibility was raised, though not used to dismiss the case, and triggered further lines of enquiry. As well as focusing on strengths, one case showed evidence of a shift in the language used, where officers logged awareness of ‘challenges to overcome’ rather than ‘weaknesses’. Some of this language was evident in earlier cases in the deep dive, but not consistently and varying depending on the officer. Logs showed awareness of some rape myths and stereotypes (e.g. awareness that individuals react to trauma in different ways and the report that the victim ‘seemed fine’ following the incident was not indicative of whether or not the incident had occurred) and conflicting evidence was not used to question the victim’s credibility.

Summary and conclusion

We know that people make decisions to commit RAOSO offences and become sex offenders. The suspect’s behaviour toward the victim, the choices they make, and the way they account for these choices, must be the central focus of the investigation. Considering the suspect behaviour allows for the more appropriate prioritisation of the investigative strategy, better support for the victim, and can help facilitate stronger case building.

At present, RAOSO investigations are not suspect focused, and significant changes are required to enable this to happen. In part this is a result of a lack of specialist knowledge about sexual violence and sex offending. Many basic investigative functions are currently not being completed, with officers relying on ‘victim credibility’ tests which undermine formulation of effective strategies and actions to investigate the suspect. The approach to RAOSO investigations is procedural to the point of box checking with little or no critical reflection or thinking, all of which is exacerbated by inexperienced officers and a lack of supervision and guidance. Many cases are left to drift through a combination of lack of capacity, lack of specialist knowledge and skills, an inability to appropriately manage or close cases, poor liaison and communication with the CPS, and structural issues causing tensions between staff.

We advocate for the development of a specialist sex offence investigator role, with specialist knowledge, skills and capability to handle all functions within an investigation and not become assigned to certain aspects, such as victim care. A significant finding for Pillar One is the importance of recognising the specialist knowledge and skills required to investigate RAOSO if positive change is to occur. This also requires recognition of:

• The niche specialisms within these investigations, e.g. investigative interviewing;

• Where specialisms may overlap, e.g. the trauma informed approach required to deal with victims of several different interpersonal offences such as child exploitation and domestic abuse.

The findings from the deep dives and existing evidence base provided in the background section justify a specialist approach to RAOSO investigations. This suggests that structural changes are required, and from a functional equivalence perspective how this is actioned may look different in different forces, e.g. the use of specialist RAOSO units in some forces, compared to the use of broader, ‘interpersonal offences’ units in others. Recognition of the requirement for specialism should at a minimum, however, provide opportunities to secure and justify ring-fenced funding for resources where RAOSO is considered a priority, and to ensure that all officers working on these types of offences have the appropriate expertise, skills, and support in place to effectively do so. Alongside the need for specialism with Pillar Four we are calling for critical evaluation and reflection to become a key part of investigations and officers’ roles more broadly. Considering our findings from the case reviews, that are often in direct contradiction to officers’ accounts in interviews, there are many missed opportunities in RAOSO investigations as a result to tackle this we propose a team approach moving forward. The basis for a team approach is constructive challenge and support between colleagues. A team approach also needs to be coupled with regular meaningful supervisory oversight and guidance of investigations to provide opportunities for learning, development and improvement around investigative strategy as well as supporting officer wellbeing (see Pillar Four for more on this). Given the many failures in investigations, the lack of ability to produce and implement effective investigative strategies, and many missed opportunities, we have produced the Operation Soteria Investigations Map to help guide officers and other decision makers on the key elements of effective RAOSO investigations.

The overall deliverable is the contribution to the National Operating Model for the investigation of rape, which will act as a bridge between relevant overarching strategies and day-to-day policing operations. Our contribution will ensure that the National Operating Model embeds specialism, a team approach, and the investigative map, which includes a suspect-focused approach to investigations alongside both the immediate and long-term disruption of repeat suspects.

Appendix 8: Pillar Two – Targeting repeat suspects: End of Year One Report

Pillar team

Pillar leads: Dr. Kari Davies and Professor Miranda Horvath. Pillar team (in alphabetical order): Dr. Katherine Allen, Arianna Barbin, Sophie Barrett, Professor Emma Bond, Ioana Crivatu, Dr. Maria Cross, Thistle Dalton, Joana Ferreira, Dr. Anna Gekoski, Rosa Heimer, Anca Iliuta, Aneela Khan, Margaret Hardiman, Asmaa Majid, Dr. Mark Manning, Kristina Massey, Dr. Ruth Spence, Louise Trott, and Professor Jessica Woodhams

Background

Repeat offending – a critical issue in the investigation of rape and other sexual offences

Repeat offending is a critical issue within the policing of sex offending. A 2013 report by the Ministry of Justice, the Home Office, and the Office for National Statistics, for instance, highlighted that of the 2,900 defendants prosecuted for rape in 2011, they were prosecuted for an average of 2.3 rape offences each. Further, around four percent of offenders convicted of a sex offence had five or more previous convictions or cautions for a sex offence^{[footnote 113]}. While these official figures demonstrate some of the extent of repeat offending, there are several challenges surrounding the reporting, investigation, and prosecution of sex offences in general which result in low conviction (and reconviction) rates (see Pillar One for a discussion of these issues). These problems make it likely that official figures on the number of repeat offenders are just the tip of the iceberg. Indeed, when we look to other sources of information on repeat offending, such as studies using self-report methodologies, they can provide a more comprehensive account of the nature and scope of repeat offending. For instance, one piece of research showed that, for every one violent sex offence (e.g. rape, child molestation) committed by an offender, they reported committing around another 30 offences^{[footnote 114]}. Research has also demonstrated that offenders admit to committing further non-sexual as well as sex offences^{[footnote 115]}. In another study, for instance, a sample of 120 ‘undetected rapists’ (those whose offences met the legal definition of rape or attempted rape, but who had never been prosecuted) were responsible for 1,225 interpersonal violence offences, including violent offences, rape, and child sexual and physical abuse^{[footnote 116]}.

While there are issues around the substantiation of such figures in these studies because the veracity of the claims made by offenders aren’t verifiable, they do provide us with an understanding of the extent to which offenders repeatedly commit crimes and paint a picture of at least some offenders who demonstrate a propensity to repeatedly offend. This is particularly significant in policing as, while there are important ethical considerations around the inclusion of unconvicted suspects in samples for research purposes, on the basis of ensuring that guilt is not assumed in these circumstances, the premise of police work is based on the collation of intelligence, which is predicated on allegation as opposed to conviction data. The police, by virtue of recording criminal allegations, have a wealth of information which can be used to explore repeat offending and repeat suspects^{[footnote 117]}. The police, therefore, have an opportunity to draw on the intelligence contained within their own records to better understand the nature of repeat offending and repeat suspects and to make more informed decisions about how to tackle this type of offending.

Psychological theories of repeat offending

Psychological theories of offending can offer a framework to determine whether and how offenders may (or may not) commit several offences over the course of their criminal career. One such theory suggests that patterns of antisocial behaviour can be characterised as either adolescent- limited or life-course-persistent^{[footnote 118]}. This demonstrates that there are at least a proportion of the offending population that are likely to repeatedly offend across the course of their lifetime. Recent work by Dr. Patrick Tidmarsh^{[footnote 119]} suggests that adolescents commit a large number of all sex offences, and while adolescent-limited offending may be constrained to a particular period in time, the theory does not preclude these offenders from repeatedly offending during this period.

There are also a number of factors which may contribute to an individual’s propensity to continue to offend. Research, for instance, has explored the factors associated with reincarceration, suggesting vulnerabilities such as drug addiction may play a part in repeat patterns of criminal behaviour^{[footnote 120]}. Domestic violence interventions and treatment programmes are framed through the Risk, Needs, Responsivity principles which tackle the reasons why offenders may commit, and continue to commit, certain types of offences^{[footnote 121]}. From a sex offending perspective specifically, the motivation to commit sex offences such as anger or sexual gratification^{[footnote 122]} may drive repeat offending.

The versatility of repeat offending

Repeat offending can look diverse. The way offences are repeated, including their frequency, the type of victim targeted, and the behaviour used, can vary. The types of offences committed across a criminal career may span several different offence types. When considering repeat offending, the scope of how patterns of offending behaviour are conceptualised is broad, and the terminology used to describe the complexity and variety of repeat offending is often not explicit. We present here some of the key considerations around the potential versatility and breadth of what can be considered within the context of repeat offending.

One of the most important aspects to note, when discussing ‘repeat’ offending, is that this is actually an umbrella term which can be used to encompass several different patterns of repeated behaviour that can occur within several different contexts:

• ‘Multiple’ offending – offending several times against one victim in the same incident;

• ‘Spree’ offending – offending against multiple victims in the same incident;

• ‘Persistent’ offending – offending against the same victim on several different occasions;

• ‘Serial’ offending – offending against different victims at different points in time^{[footnote 123]}.

Suspects can repeatedly offend in one, some, or all these ways. The definition of a persistent suspect would fit, for instance, a domestic abuse suspect who repeatedly targets their partner over a period of time. They may also be a serial suspect if they have offended against multiple, different partners over the course of their criminal career. We acknowledge that this definition of repeat suspects is very broad, and in many contexts only certain types of the repeat suspects outlined here may be relevant to an investigation. It is essential, however, for policing to understand that repeat offending is not homogeneous, and the different contexts in which repeat offending can manifest, to make the most informed decisions throughout an investigation and how to contribute to the disruption of repeat offending.

The notion of repeat offending is also not limited to the idea of offenders repeatedly committing sex offences. There is a body of literature which suggests that sex offenders do not tend to ‘specialise’ to commit sex offences only, and in fact are versatile in the types of offences they commit^{[footnote 124]}. The reason why this is relevant within the context of policing is because it means that, alongside the wealth of information the police may have about a repeat suspect’s sex offending behaviour, there is the opportunity to draw on information about non-sexual criminal history. Research suggests, for example, that there are links between burglary, sex offending, and offender characteristics, such as the length of their criminal career^{[footnote 125]}. It is important, therefore, that even when previous offending may not be deemed to be directly related to a current offence being investigated, that it is still considered within the context of gathering a holistic view of a suspect.

As well as committing several different types of offences, a suspect may also have offended against several different types of victims. There is a body of literature on sex offending which demonstrates how some suspects offend against both male and female victims, victims of different relationships to the suspect, and victims of different ages, across a series of offences, with around a quarter demonstrating crossover behaviour in at least one of these categories^{[footnote 126]}. The proportion of offenders that have targeted different categories of victims rises when offenders are asked these questions under polygraph conditions^{[footnote 127]}.

In summary, understanding the scope and breadth of repeat offending is important as it has critical ramifications for policing. It is important for recognising patterns of behaviour which may assist in linking offences together as a series committed by the same suspect. It has the capacity to feed into more appropriate risk and safeguarding assessments, as well as to create more comprehensive investigative strategies including the investigation of several crimes as a series where appropriate. The collation of multiple crime scenes based on a series of offences could allow for more effective collation of evidence. Arbitrarily placing limitations on what should and should not be considered relevant offending, therefore, means relevant intelligence about suspects may not be appropriately factored into investigative strategies and can affect the efficacy of police investigations into rape and other sex offences at a number of levels.

Investigative options for disrupting repeat offending

Short-term investigation of repeat offending

There are several areas within the immediate investigation of a sex offence where information on repeat offending becomes relevant. As part of the investigative strategy, repeat offending intelligence can also be used to inform the risk assessment of the suspect conducted by officers. It should factor into how the initial investigative strategy is formulated, and repeat offending intelligence can have an impact on this strategy in a number of ways. It can assist in the collation and combination of evidence from multiple offence scenes, e.g. behavioural and ANPR data, forensic evidence from different crime scenes, or witness accounts and items found at separate scenes. It can also assist in decision making around investigative actions taken, such as the decision to voluntary interview a suspect, or to arrest and interview them under caution. Behavioural analysis can be conducted on a series of offences to gather more intelligence on a suspect’s course of conduct, as well as to search for potential links to other, as of yet unknown, offending the suspect may have committed (e.g. finding further victims of a suspect who routinely uses Snapchat as a method of targeting victims). Previously closed cases could be considered in the context of reopening them to investigate a suspect for a series of offences.

It is important to note that considering how repeat offending intelligence factors into an investigation can be considered both when the suspect is known, as well as when they are unknown. Simply because an offender is unknown, does not mean that attempts cannot be made to recognise series of offending and investigate these accordingly; for instance, linking of forensic evidence can be used to establish whether a set of offences is a series likely to have been committed by the same offender. In this respect, behavioural crime linkage can also be used to establish whether offences are likely to have been committed by the same suspect^{[footnote 128]}.

Last, understanding series of offending is not only important from an investigative and evidential perspective, but also from a logistical and resourcing perspective. Recognising series of offending – of both known and unknown suspects – allows for police resources to be more appropriately allocated, for officers to have a more holistic and detailed understanding of a suspect’s behaviour overall, and to ensure less duplication of effort on the part of law enforcement in dealing with repeat suspects.

Long-term disruption of repeat offending

One of the key aspects of recognising repeat offending is the potential for law enforcement to take a more strategic, long-term approach to disrupting repeat suspects. The long-term disruption of (sex) offending is particularly important, not just to obtain better justice for victims and make better use of police resources, but also to prevent harm rather than reacting to crime; a key Peelian principle of policing^{[footnote 129]}. Risk, here, also needs to be assessed, which goes beyond the risk posed by an offender in a specific investigation of an offence or offences but can start to be conceptualised in terms of the wider risk – or harm – that person may pose. This speaks to the more long-term disruption of offending and the type of crime prevention work that law enforcement should (and sometimes do) work on. While there are important ethical issues to consider around the targeting of suspects, as opposed to offenders, within the parameters of policing this is not a new concept; problematic and persistent offenders are often prioritised for targeting to disrupt harmful patterns of offending, such as in organised or gun crime. This could include options such as providing bad character evidence in court or applying for civil orders.

Rationale for Pillar Two – better targeting of repeat suspects

There is a substantial amount of information which suggests sex offending is not understood well by law enforcement. Pillar One outlines the poor cultural attitudes which permeate the police as well as more generally within society, which see victims deemed responsible for sex offences perpetrated against them and offending and its ramifications minimised. Pillar One also outlines how the suspects involved in these offences are not appropriately considered or focused on, particularly with regard to their intent to commit sex offences and their ability to take advantage of victim vulnerability and use techniques such as grooming and coercive control to offend in this manner. These issues relate equally to Pillar Two; the lack of suspect focus and understanding of sex offending and suspects extends to repeat offending and the potential for sex offenders to repeatedly commit sex and other offences. The lack of understanding of repeat offending is not singular to sex offending; we see evidence of law enforcement struggling to conceptualise or disrupt offences that are, by their very nature, repeat offences, as evidenced in the recent HMICFRS report into stalking and harassment^{[footnote 130]}.

The inability of law enforcement to appropriately recognise repeat suspects and use intelligence to factor into investigations is evident in several high-profile cases which demonstrate the scope of failings in this regard (e.g. John Worboys and Kirk Reid). More recently, the case of Stephen Port demonstrated significant failings to conduct routine intelligence checks which would have resulted in the generation of relevant intelligence during the investigation^{[footnote 131]}. The fact that law enforcement is not equipped to see offending as a pattern of behaviour which can manifest throughout several different offences and different types of offences is a serious failing in policing. Added to the already poor way sex offending is dealt with at a general level, because of the lack of suspect focus, urgent attention is required as to how law enforcement better conceptualise, identify, and disrupt repeat (sex) offending on both a short- and long-term basis.

In short, repeat suspects pose a significant risk to the public and the police are currently ill-equipped and lack expertise to deal with this type of sex offending. Repeat offending is diverse in nature and needs to be properly understood for it to be investigated efficiently and fully, including a having specialist understanding and appreciation of the broad range of options for targeting repeat offending on both a short- and long-term basis, and the capacity for officers to do so.

The purpose of Pillar Two, therefore, is to investigate the current practice of forces in identifying and disrupting repeat suspects, with a mind to providing recommendations and outputs to clearly articulate what repeat suspects look like, how many there are, and how policing can improve their disruption of repeat offending in both the short- and long-term.

Methodology

Data collection for Pillar Two was shared with Pillar One, due to the multiple points of convergence in terms of the information required to answer the research questions. For further details of the underlying methodological approach and of the data collected as part of both Pillars, please see the Pillar One report. Strands 1, 3, and 5 were specific to Pillar Two only and are therefore detailed here. The methodological detail below highlights how each data strand was used to answer specific research questions relating to Pillar Two.

The exploratory research questions for Pillar Two were:

• What is the extent of repeat offending in the forces?

• What does repeat offending look like?

• How (effectively) do officers conceptualise repeat suspects?

• How (effectively) do officers identify repeat suspects?

• How (effectively) does the knowledge of repeat offending factor into investigative strategies?

• What long-term options are used to disrupt repeat offending?

• What are officers’ perceptions of the changes implemented within the pathfinder Force E and have the changes led to more focus on repeat offending?

Strand 1 – quantitative data (Pillar Two only; used to address research questions 1 and 2)

Two quantitative datasets were collected from the five pathfinder forces. The first (coordinated with Pillar Five) was an anonymised, four-year sample of all sex offences recorded by the forces between January 2018 and December 2021 including the following information:

• Offence ID, type (offence group and offence description), Home Office code, date occurred, date reported, domestic violence flag, locality, outcome date, and policing outcome

• Suspect ID, age at the time of offence occurrence and reporting, ethnicity, and relationship to the victim

• Victim ID, age at the time of offence occurrence and reporting, and ethnicity

The second dataset contained details of all previous criminal history data for all named offenders in the first dataset, containing some suspect details and offence details – HO code, description, date of offence, and outcome. Details of suspects’ criminal histories was obtained as far back as forces were able to provide them, which tended to be 1995, apart from Force A which went back to 1993 and Force C which went back to 1996. Details of all offending not limited to sex offences were included.

The total number of data points obtained from all forces (A, B, C, D and E) are as follows:

Strand 2 – case reviews (used to address research questions 1-6); Strand 2a – development of case reviews for use by researchers in pathfinder Force E (used to address research question 7)

See Pillar One methodology 2.1 for details.

Strand 3 – survey (Pillar Two only; used to address research questions 4-6)

The survey comprised of 10 open ended questions asking about initiatives and projects targeting sex offence suspects and offenders, the use of intelligence systems, details of offender management teams/frameworks, the use of civil orders, the use of third-party disclosures, details of work with external organisations to target high harm offenders, and any other examples of good practice associated with investigating or targeting sex offence suspects and offenders and any challenges faced. As RAOSO investigations and management largely falls within public protection, the survey was sent to all DCIs in public protection within the five forces, as well as to any other officers the Police Pillar Leads thought would be relevant. One force also sent out the information request nationally, which led to additional information being provided from forces more widely, and which are included in the total. In total, 20 information requests were received from forces, ranging from zero in one force to nine in another. Content analysis was used for analysis to draw out specific examples of police practice relevant to the project. The survey was targeted primarily at Public Protection Officers to gain further insight into how officers recognise and tackled repeat offending, including any strategies and initiatives aimed at conducting long-term disruption.

Strand 4 – interviews (used to address research questions 3-6)

See Pillar One methodology Strand 2.2 for details.

Strand 5 – document review (Pillar Two only; used to address research questions 3-6)

Police Pillar Leads in each force provided all rape and sexual assault investigation policy and procedural documentation they deemed relevant, including (but not limited to) national and local policy documents, local training and guidance. A national call was also put out for relevant documents. 177 documents were provided in total; these were reviewed by the research team, rated for relevance to rape and sexual assault investigations (high, medium, or low priority – based on very relevant, some relevance, no relevance), and analysed. Time constraints meant that in some forces only high and medium priority documents were analysed. Thematic analysis broadly corresponding to Braun and Clarke’s^{[footnote 134]} six stages was used, i.e. initial familiarisation with the dataset, coding each document, review of potential ‘themes’ or central concepts in relation to the whole dataset, and defining boundaries/identifying distinctiveness of each ‘theme’ or concept. Themes/concepts from the document review were also discussed by the team.

Data insights and limitations

There are some limitations to the data which need to be acknowledged:

• Some forces do not use unique identifiers for individuals in their crime recording systems. Therefore, to search for named suspects’ previous criminal histories, in some forces this had to be done by matching records using suspects’ first names, last names, and dates of birth. This is not a fool-proof method and will almost certainly have led to over-identification (e.g. different people with the same name being matched) and under-identification (e.g. the same person not being linked because of miss-spelling or missing data) of repeat suspects^{[footnote 135]}. It does, however, represent the method officers themselves would have to use to obtain intelligence on repeat offending.

• The case reviews were reliant on participants accurately and fully completing the data spreadsheet provided to them. Where data are missing, it is possible it is because the spreadsheet was not completed properly.

• Equally, the data represented in the case reviews is a reflection only of the data recorded in the forces’ crime recording systems. While officers should note down all actions taken in an investigation, there is no guarantee that this is done (indeed, please see Section 3.6.3 of this report for evidence data are not properly recorded), and therefore analyses may include actions that were completed, but not recorded.

• The interviews were arranged by the Police Pillar Leads in each of the forces. As a result, they may have been subject to selection bias, over and above the selection bias normally associated with a snowball or mass approach to participant engagement.

• Likewise, the information requests were sent out and documents for review collected by the Police Pillar Leads, which again may have led to sampling bias.

Equally, however, the data collected as part of Year 1 of Operation Soteria Bluestone represent some of the most detailed and diverse information of this nature to be collected:

• The sample of quantitative data from each of the five pathfinder forces , represents a sample of offences, suspects, and victims, unique not only in its scope and size but also because of the inclusion of offences that had variable policing outcomes (as opposed to focusing solely on charge or conviction data). The ability to match suspects and victims to individual incidents and to link criminal history data to suspects in this sample will allow us to provide novel insights into offending patterns and trajectories in Year 2.

• The case reviews conducted represent another unique source of data collected as part of Pillars One and Two. The information about the case progression itself is usually prohibitively difficult to obtain, due to the need to manually extract it from crime recording systems and the time and security constraints associated with this. This information provides us with a unique insight into the ways investigations progress. Further, the review data are also unique in that they provide a novel insight into both OICs (Officers In Charge) and senior officers’ perceptions of the efficacy of investigations. These case review data were also an integral part of assisting forces in assessing the efficacy of their own investigations.

Findings

Identifying different types of repeat suspects

As noted in the introduction, there are multiple ways that repeat offending can be conceptualised. It is important for officers to understand the different ways in which repeat offending can manifest because of the ways in which this information should factor into an investigation. For instance, while ‘multiple’ offending may represent several different types of offences committed simultaneously, such as rape and robbery, particularly within the context of repeatedly committing the same offence, e.g. multiple rapes, it is likely to be associated with having access to the same victim for a prolonged period of time. This in turn has ramifications for victim safety, with sexual murderers shown to spend more time with victims than rapists^{[footnote 136]}. Persistent offending may equally suggest that a suspect has prolonged access to a victim, perhaps within the context of domestic abuse, which has implications for the continued victimisation of that victim and consequently how officers think about appropriate victim safeguarding. The risk associated with ‘spree’ offenders should be carefully considered, given the additional risks that could be associated with maintaining control over multiple victims at the same time, which may demonstrate the increased level of risk that an offender is willing to take. Finally, serial offenders demonstrate that there are suspects that will continue to offend, whether within the same offence type or by committing lots of different types of offences, and there is a need to consider whether and how these patterns of offending will continue without law enforcement (or other) intervention.

For the purposes of the analyses here, we defined a repeat suspect as anyone involved in more than one offence, e.g. two rape offences. With this information, we were able to explore the number of different types of repeat suspects encompassed by this umbrella term;

• ‘spree’ offending, or two rapes committed by the same suspect in the same incident against different victims,

• ‘persistent’ offending, or two rapes committed against the same victim in different incidents, and

• ‘serial’ offending, or two rapes committed in two different incidents against two different victims.

The analyses included all repeat offending history and was not limited by offence type, e.g. a suspect could be classed as ‘persistent’ if they had raped a victim and subsequently robbed them at a different point in time. We note here that our data do not include ‘multiple’ offending, where one victim is offended against multiple times within the same incident, e.g. a victim being kidnapped and raped four times over a period of 12 hours. We were unable to assess this type of offending for the logistical reason that in these circumstances only one offence per incident is recorded by the police, so it is not possible to get an accurate figure of this type of offending without taking the data directly from the victim’s account.

From the data sampled, the vast majority of repeat suspects are serial suspects. For instance, in Force D, there were only 1.1% of spree-only suspects, and 6.9% persistent-only suspects (including all offence types); 8.0% in total. The rest of the suspects (92.0%^{[footnote 137]}) were serial suspects, which could include serial-only, serial-spree, serial-persistent, or serial-spree-persistent suspects.

Number of repeat suspects

Based on the named suspect data, which allows us to track these suspects’ criminal history, Table 1 demonstrates the number of suspects who are linked in the forces’ databases to more than one sex offence, to more than one offence of any type, and who have never come to police attention before. Note that the majority of these suspects represent those that have committed more than one crime against more than one victim (as per the figures in Section 3.1).

Table 1. The number of named suspects with criminal histories

• These columns add up to 100%.

As Table 1 demonstrates, the number of named suspects linked to more than one sex offence is high, ranging from 19.5% to 45.6%. The largest number of sex offences committed by one suspect was 173. When considering all offences, the number of repeat suspects increases, ranging from 41.4% to 63.9%. In all forces other than Force B, less than half of the named suspects had never come to police attention before, demonstrating the scope and breadth of the repeat suspects within the sample obtained here. These figures are particularly impactful when considering that our data were time and force limited (we cannot, for instance, say anything about repeat offending which crosses force boundaries). We note that not all this previous criminal history will necessarily be relevant to the investigation, but it demonstrates just how frequently potentially relevant intelligence will be available to officers and the necessity for them to be equipped with the relevant knowledge and time to be able to appropriately assess this intelligence in terms of how it factors into both a short- and long-term investigative strategy.

While analyses of these data are still ongoing, initial analyses showed that only 7.3% of repeat suspects in Force C and 12.7% of repeat suspects in Force D only ever committed sex offences, demonstrating the diversity of offending type included in the sample. Further analysis of these data will allow us to look for specific associations between repeat offending and whether suspects demonstrate particular patterns of offending, such as committing several types of domestic abuse offences, including sex offences, whether the suspects in our sample are generalists or specialists, and how variable they are in the types of victims they target. It is also worth considering that our initial analyses demonstrate the extent of some suspects’ criminal histories, e.g. one suspect in Force D was linked to 315 offences. This has important implications for how we conceptualise repeat offending in terms of the prolific nature of some suspects, and what these offending trajectories look like.

Issues around repeat suspect identification

The first step to targeting repeat suspects is their effective identification. The analyses uncovered a number of issues around repeat suspect identification which has the potential to impact both the subsequent short-term investigation and the consideration of more long-term disruption options.

A lack of understanding of what constitutes a repeat suspect

As noted above, the scope of what constitutes repeat offending can be broad, but officers should be prepared to consider what repeat offending may look like at a number of levels so they can appropriately identify this offending and use this knowledge to contribute to an effective strategy for investigation and offence disruption. When discussing repeat offending with officers, generally their understanding of what constituted a repeat suspect centred around the majority type identified here; serial suspects who have offended against more than one victim at different points in time. At a general level, previous criminal history was considered relevant, but where some officers were inclined to consider all types of offending in their assessment of how this intelligence should factor into an investigation, there were parameters that some officers imposed upon the definition of the repeat suspects which limited their conception of repeat offending. These differences were seen at both an inter- and intra-force level. For instance, the definition of repeat offending was often centred around what constituted ‘relevant’ criminal history. As noted above, it is unlikely that all criminal history will be relevant to the progress of an investigation or the manner in which officers approach the more long-term targeting of said suspect. There were, however, instances where officers imposed blanket restrictions on what constituted relevant offending. Most officers considered a past history of sexual and violent offending as being directly relevant, although there were some disagreements as to whether ‘dishonesty’ offences would be considered relevant:

If they’re a violent individual and they’ve got previous for carrying knives or particularly if they’re involved in gangs … the risk would be heightened … And that would certainly cause room for concern … So it would form part of our vision

(Force A5)

Violence is always relevant because it can be a scale in which they build. Stalking is another one, harassment – they can all be precursors to sexual offences

(Force D4)

If it’s a dishonesty offence, then of course that’s relevant because as we said before, its one word against the other, and if the other has a huge number of convictions for dishonesty offences … they may not be telling the truth in this case … I would have thought in rape investigation, dishonesty would be a key one

(Force A13)

I think most offences are relevant, although I wouldn’t necessarily look at a simple shoplift as being a precursor to sexual offences

(Force D4)

Limiting what constitutes a repeat suspect at an aggregate level, without consideration of the individual circumstances of the offence, is an issue because officers are potentially missing relevant information, either through lack of searching for it or lack of critical analysis, that could factor into the investigation for the more effective investigation and targeting of the suspect. Further, while some repeat offending may not be relevant to the course of the immediate investigation, it may become relevant when considering how a particularly prolific offender could be disrupted on a more long-term basis. It is important, therefore, that officers understand these nuances and how they factor into investigations at different stages.

Poor identification of repeat suspects

As well as the challenges around recognising the scope of repeat offending, there was also the issue of identifying the intelligence required to make judgements about whether any named suspect was a repeat. Officers recognised that checks should be undertaken using different databases to search for information about previous criminal history when any suspect is named. It was clear from both the interviews and the case reviews, however, that this was not always done and this was a consistent finding across all forces. For example, in Force B, suspects, especially those who were foreign nationals, often did not have their criminal history checked at all – therefore their status as a repeat suspect was unchecked (or not logged). Officers across forces reported PNC/PND (the Police National Computer/Police National Database) checks for repeat suspects sometimes not being done (or being done but not recorded).

I’d love to say we checked everyone nationally … I think we would definitely check on PND if we knew that they had lived in a different force … then we would ensure that was checked … But not everyone has access … PND is a separate course you would have to do. So we would have to put a request into our local intelligence teams for them to do that … So I wouldn’t say that every single person is checked on PND

(Force A3)

Does it always happen [checking the PND]? Probably no, I would say if I’m honest, no, it doesn’t always happen … it is an always action set but would I say confidently they always are 100 percent done? No, I wouldn’t say so

(Force A5)

Officers in Force C reported that their new ‘difficult’ and ‘clunky’ operating system hampered effective reporting and recording of intelligence checks:

On the old system an officer would pick up a job and do the checks and they would then collate all of those results. And they go on to our investigative record really neat and tidy. So you could see what the history was between the two parties. What happens now is I think those checks are being done, they’re just not being recorded, and that’s probably because [Force system] is more difficult. It feels like it’s clunky. And I think there’s a risk there

(Force C3)

These findings show that poor identification of repeat suspects, or at the very least the poor recording of repeat suspect checks, affects investigative strategy and decision making throughout investigations as officers are not equipped with the relevant information to make the most appropriate investigative decisions.

Ignoring information about repeat suspects

At worst, instances were found where officers flagged named suspects as being repeat offenders, and then subsequently noted that they were not repeat suspects, despite having found clear evidence to the contrary. This was seen in the case reviews, where across forces A to D, 19% of suspects were identified as repeat suspects by officers. However, when the researchers went through the information provided to them by the reviewers, they identified 38% of suspects as repeat suspects. This means that, even though the information about a suspect being a repeat was readily available, some suspects were not being recognised as repeat suspects. While some of this decision making around who constituted a repeat suspect may be attributable to officers applying a narrower definition of offending, there were examples where details of previous criminal convictions specifically for sex offences were documented and subsequently ignored.

There is, however, a disclosure of an assault by the victim in 2018. I can see no PNC or PND checks on victim or suspect and I am unaware if the suspect has any previous convictions for similar offences. There is an entry from the ** officer that initially spoke with the victim that the suspect is known for assault. I cannot see any further information in relation to this.

(Force C – Case 15, Reviewer 1)

Lack of suspect focus. Was a PND completed? There are 158 Niche occurrences linked to suspect 1 – was this fully researched? There are 38 occurrences for suspect 2 – again was this researched? He is flagged as a serial perpetrator of domestic abuse, I cannot see this logged anywhere on the OEL. S1 is also flagged as a high-risk perpetrator of DA.

(Force D – Case 4, Reviewer 2)

Highlighting the direct impact on investigations and investigative strategy (which is developed in Section 3.6), we found in Force C that repeated patterns of concerning behaviour, including previous similar offences/related crimes such as kidnap, assault, and domestic abuse were often ignored and as a result these histories were not fed into risk assessments or decision making which could have altered the course of investigations.

The use of repeat suspect intelligence during an investigation

When intelligence on repeat offending is uncovered, its relevance should then be considered to the investigation. This should be conducted at several levels, including how it is used to conduct risk assessment and how it factors into an investigative strategy.

Inconsistent risk assessment

Considering repeat offending is a key aspect of risk assessment, as it should factor into officers’ perceptions of how they manage that risk, and consequently safeguard the victim in question and the general public more widely. There was some indication of repeat suspect intelligence being used to judge a suspect as higher risk:

So I think you investigate someone and you see that they’ve been arrested, certainly convicted of an offence, but even arrested for an offence, your risk level goes up. And it might affect how more quickly you’re able, you prioritise that I suppose, how much quickly you invite that person in for an interview or you go out and arrest them. So intelligence is quite key in terms of risk assessment

(Force A13)

The offender was a repeat offender in terms of DV and this is referenced throughout and informs the decision to remand and the decision for the MARAC [Multi-Agency Risk Assessment Conference].

(Force D – Case 71, Reviewer 1)

While one of the above examples comes from the case reviews, in these data it was also clear that repeat offending intelligence was often ignored when considering suspect risk, suggesting that this may be something that is considered in principle, rather than undertaken in practice:

I note the DP was flagged as a serial perpetrator of domestic abuse against partners, and his previous partners have been high risk of DV and subject to MARAC. He is clearly a risk to females and I cannot see this documented anywhere. He was remanded until the court case, however, was astonishingly found not guilty at court. Considerations could have been made to refer him to the Wisdom project due to his clear risk to females.

(Force D – Case 60, Reviewer 2)

It is not clear if the SOM [Sex Offender Manager] has discussed the offending/behaviour with the suspect or considered if this is a pattern behaviour. It is therefore not clear if any protective orders were considered.

(Force C – Case 11, Reviewer 2)

There was also evidence of risk being incorrectly assessed:

Lack of wider considerations around referrals / ISVAs – the risk grading on the DARA [Domestic Abuse Risk Assessment] was medium – this should have been high, given multiple rapes – he was linked to rapes of family members, had previous for sexual offending, was using violence etc. Also force policy is that all rapes are classified as high risk.

(Force C – Case 4, Reviewer 2)

These findings also fit with those from Pillar One which suggest that, in general, risk assessments were often simply not undertaken, and highlight that when they are, they are often incorrect or do not make good use of intelligence to factor into the decisions made.

From some of the quotations above from officers discussing relevant repeat offending behaviour, there is some evidence officers have an awareness of an offender’s potential to escalate, with some indication they felt certain types of offending were relevant from a risk assessment perspective. There were, for instance, some interesting observations on how particular behavioural information may factor into perception of risk, e.g. an offender’s willingness to weapon carry and how this may relate to other offending. Officers across forces, however, frequently noted that repeat stranger offenders were in general seen to be ‘riskier’ than those who offended against the same individual (usually domestic offences). Officers often cited risk to the public and the ‘unknown’ element to stranger offences, not knowing where they might strike next, against whom, or whether their offences might escalate, with these offences being worthy of additional urgency, time, and resources. In contrast, the domestic offences were believed to be less risky on the basis of the ‘known suspect’ element to them, meaning officers could impose conditions to help protect the victim.

When it’s a stranger attack, there’s the chance that the stranger might commit further offences and we don’t know who he is. So it’s important that we track that person down as quick as we can, try and try and find out who they are and locate them, and arrest them as soon as possible. So there’s no significant harm done to any other persons or victims

(Force C9)

Whenever I think of a stranger rape, I think about it in terms of threat, risk, harm, community impact, press … With a stranger rape there’s a certain element of unknown risk that could pose a risk to other individuals out in the community, that could lead to community tensions. It means that the investigative process to identify that individual has got to be quick. So, getting things fast tracked through our forensic service providers, for example. So a domestic rape case, I’m not minimising that at all, but we don’t necessarily need to fast track it – if the partner is saying my husband has raped me last Tuesday – because there’s a known element to it

(Force D3)

If it’s a stranger, all three teams should be all hands to the pump. Everyone should be downing tools and working on that … having briefings between the three teams, seeing what they can do in terms of CCTV, fast-tracking forensics, any witnesses

(Force C7)

From the perspective of prioritising a suspect that is ‘at large’, there is a logic to judging this as riskier to other members of the public or to dedicating resources to identifying that offender, as well as considering the wider implications of community expectation. There is also the possibility that officers can provide better safeguarding to an existing victim of a domestic abuse offence, the circumstances of which permit that victim to be given additional support. There is, however, some confusion demonstrated here by officers in terms of the risk of recidivism that stranger versus a domestic abuse perpetrator may pose to future victims, with the former being presumed to be much more likely to reoffend. Further analyses in Year 2 will be undertaking this work, but it is important to note that the blanket assumption that stranger suspects are more likely to reoffend, without consideration of suspects who repeatedly target the same victim (and who may have continued, repeated access to that victim), or to the idea that suspects that commit domestic abuse offences may not also commit offences against other types of victims, is not a valid assumption to make based on the available evidence. Again, the notion of dedicating resources to an ‘at large’ suspect makes sense from a risk perspective, but officers do not currently acknowledge (or seem to understand) that domestic abuse or acquaintance rape suspects who are also ‘at large’ may still – given many suspects’ propensity to commit offences against victims of different relationships to them – pose a level of risk to victims outside of the one victim that reported the offence.

Inconsistent use of this intelligence factoring into the investigative strategy

When a suspect is identified as a repeat suspect, there should be consideration of how this information factors into the investigation and whether it will influence the investigative strategy. In the interviews, officers across forces often said that identifying a suspect as a repeat would affect the investigation on different levels. For example, officers said that they would often seek arrest rather than invite a repeat suspect in for interview, or try to have them remanded in custody rather than bailed, and that they would consider the use of civil and restraining orders in domestic abuse cases:

If somebody has previously been accused of a similar thing, you want to get a tight grip on them. Whether you’ve got enough to get them charged and remanded in custody that day or in terms of bail conditions … are they a threat to anyone else, how we manage them in the community

(Force B6)

Sometimes we’ll arrest someone for an offence, discover another linked offence and then not only will we do more analysis of that person’s phone, but we would ask our Local Intelligence Team to do some crime mapping analysis to see if there’s other offences that that person could possibly have committed at we could link them to. And that wouldn’t typically happen for a first time offender who we have no reasonable grounds to suspect of any other offences as well

(Force A10)

If he is deemed to be a sexual predator … if we deem that they are a risk to society, let alone the victim, then we’ll be making the decision to arrest him at earliest opportunity

(Force A5)

Evidence from the case reviews, however, suggests that the repeat offending intelligence was often not used to make decisions about the investigation which, as noted by Reviewer 2 in Case 65, could have safeguarding implications for the victim:

Nothing mentioned on the crime regarding the suspect’s previous offending history. Looking at [crime recording database], the suspect is subject to a harassment order against his ex and was then accused of rape by a different female 6 months later.

(Force B – Case 19466, Reviewer 1)

The accused has had a history of domestic incidents and had previously been charged with rape. This did not form part of the investigation and was not presented to the CPS in the CM01. This therefore may have ultimately led to him being bailed from court and could have resulted in further incident to the victim.

(Force D – Case 65, Reviewer 2)

Again, the difference between the interview and case review data suggests that there is a discrepancy between what officers want to (or feel they should be) doing, and what is happening in practice. The lack of repeat offending intelligence factoring into the investigative strategy has the potential to impact the investigations on several levels, such as the generation of appropriate arrest plans and interview strategies.

Finally, as part of the generation of an investigation strategy is the concept of reopening closed cases to investigate a set of cases as a series, or combining several ongoing cases. This was largely undiscussed, but when it was the attitude towards it was negative:

Sometimes we go back and speak to old victims but there’s not much appetite for that from police and CPS due to capacity

(Force A19)

This negates the idea of a more efficient investigation due to the collation of numerous reports, intelligence, and evidence, which could in the long run assist with capacity because suspects that are coming to the attention of police again and again are being dealt with on a more holistic or long-term basis.

Logging the information is not consistent

The poor use of crime recording systems was evidenced at times where repeat suspect intelligence checks were not logged, even when the case reviews suggested they had been conducted:

Previous allegation of a sexual offence against the accused which is not documented. No PND for accused.

(Force D – Case 71, Reviewer 2)

The ramifications of poor data recording are discussed in Pillar Five in more detail, but we note here that, from a Pillar Two perspective, improved recording would result in a clearer understanding of any repeat offending, both for any other officers reading through the case materials, and for the OIC who, as noted in Pillar One, may have to revisit the case notes after a long period of time.

Very little evidence of long-term disruption to target repeat and prolific offending

While officers across forces talked about the possibility of using of civil orders to disrupt offending on a more long-term basis, there was little evidence that was actually done in practice, due to large workloads and/or a lack of understanding. Officers in Force A said:

Officers often have the opportunity within certain investigations that they’re holding to apply for either criminal or civil orders like Sexual Harm Prevention Orders or Sexual Risk Orders. And they don’t understand, they don’t even know what they are, or how to go about doing that. And then when they do, they then have the challenge of managing their current caseload with managing this particular application that they have to do. And that creates a problem, and I don’t think that we obtain enough of them, frankly, as a preventive measure

(Force A10)

This notion of crime prevention was not discussed in detail by any of the officers, neither in the interviews nor the case reviews. We note here that targeting repeat offending moves beyond the immediate need to investigate a suspect and should – from a crime prevention, a victim safeguarding, and a resourcing perspective – consider longer term options for offence disruption. Policing must be more proactive in its approach to disrupting suspects they have intelligence on who are repeatedly committing offences. In this way, the importance of Pillar Two is highlighted through both officers’ need to consider short-term ways of investigating repeat suspects, but critically, understanding how long-term disruption plays a role in combating repeat offending.

A lack of engagement with other law enforcement and police agencies to target repeat suspects

There were several instances where a lack of engagement with either other law enforcement or police agencies was demonstrated by officers. For instance, and as noted above, foreign national intelligence checks were often not completed, liaison with Offender Management Units (OMUs) was not considered, and other, existing police agencies were also not considered:

It does not appear that this case was raised for MARAC which is worrying as it is clear that the suspect is a potentially dangerous individual and it will be a matter of time before he reoffends or worse.

(Force B – Case 14109, Reviewer 2)

We see a similar lack of coordination with the CPS to target repeat suspects. Early Advice was only used in 6.78% of cases where the police identified a repeat suspect (6/76; or 3.95% of cases if using the researchers’ identification or repeat suspects; 6/152), and this was reflected in a prosecutor’s comments about the lack of information given to them around repeat offending and the issues this caused:

I tell you what, we’re not notified necessarily that these are repeat offenders … I think we don’t do enough between ourselves and the police, that there isn’t a connection for those types of cases … I don’t have a link with anybody on repeat offenders or … the sort of a person I would call a dangerous offender that we need to do something about. And actually, that might be something that we need to have at a much more senior level, a high level conversation about where we’re going and what our strategy is for those cases

(Force A, CPS1)

There is limited evidence of any joint working, including to enhance investigative expertise through the use of colleagues, and all of this feeds into a picture of a lack of understanding around the need to identify, investigate, and strategically target repeat suspects.

A lack of specialist knowledge and capacity that hinders the ability to identify, recognise, and investigate repeat suspects and conduct long-term disruption

The above lack of recognition of what constitutes a repeat suspect, the lack of ability to effectively identify this intelligence, or the understanding of how to use this intelligence both in the immediate investigation and as part of a long-term plan to disrupt offending is evidence of officers who are ill-equipped, both in terms of their understanding and their capacity, to appropriately target repeat suspects. This links directly to findings in both Pillars One and Four around the crucial need for a specialist approach to investigating sex offences. It demonstrates that currently, officers in general do not understand sex offending, including repeat offending, generating a reliance on outdated cultural attitudes, and resulting in missed opportunities and poor investigative practice in both the short- and long-term. While there is some evidence of specialist knowledge in relation to targeting repeat suspects that could be drawn on from other areas of policing, there is little evidence this knowledge is currently being shared.

It is important to note here that for investigators, the idea of long-term disruption was clearly not built into their roles or considered at a capacity level. As denoted in Pillar One, many missed opportunities were found in the case reviews, and those with repeat suspects were no exception, with 67% of cases where suspects were identified as repeats containing these types of investigative oversights. There is no time or space for officers to consider long-term options for offending disruption, and no clear pathways for officers to be involved in this work once an investigation has closed. This includes a lack of analyst support to assist in identification and consequent long-term disruption of offending.

Promising practices

From the surveys and document review, several promising practices were identified that some of the forces had implemented to target suspects of repeat sex offending. This included evidence of more proactive approaches to policing such as:

• Operation Vigilant: policing public areas of the night-time economy to safeguard potential victims;

• Operation Bassona: Proactive operation to interrupt, disrupt, or proactively police those that have been deemed as potential high-risk offenders using the Recency, Frequency, Gravity model;

• Predatory Offender Unit; an offender-focused unit designed to tackle repeat sex offending.

It is important to note that, while avenues for long-term disruption were often mentioned within the context of this type of data collection, such as the use of civil orders, evidence from the interviews and case reviews suggests that, in practice, these options are neither well understood nor routinely considered or used. While informed and ethical policing was also mentioned as part of this data collection, in terms of noting where officers were provided with insight and knowledge to equip them to deal with sex offence investigations, there was also compelling evidence in the interviews and case reviews that knowledge in this regard is also lacking in practice (Pillar Four covers this in more detail). Further, some issues in practice were raised with some of the initiatives, such as the Predatory Offender Unit focusing largely on stranger offences, and often not being provided with the requisite information to assist in the investigation of these offences.

Change is possible

The case reviews conducted as part of the pilot work in Force E demonstrated some general level of improvement regarding the identification and disruption of repeat suspects. For instance, most checks seem to have been done/requested on suspects in all cases reviewed, although these checks were still sometimes delayed, and in other cases there was no evidence on the crime recording system showing that checks were completed after being set as an action. In six of 12 cases available at the time of writing, intelligence checks were requested on victims, and in two cases the victim check was conducted before the suspect check, suggesting there remains a level of suspicion and scrutiny of victims even as investigations are now focusing more on suspects.

Some of the more recent cases reviewed made adequate use of short-term disruption tactics, such as the use of bail conditions and witness intimidation warnings, and there seemed to be an increased awareness of the use of civil orders. There were two examples in domestic cases of good practice in this regard. In one case, a disruption management plan was immediately set up by the Detective Sergeant supervisor, which included suspect arrest and the use of bail conditions. Despite one of these cases being closed as outcome 15 (due to evidential difficulties) and the other as outcome 14/16 (because the victim did not support police action), there was good collaboration with Independent Domestic Violence Adviser/Independent Sexual Violence Adviser to ensure the long-term disruption of the suspect and safeguarding of the victim by assisting the victim in securing a non-molestation order. A comprehensive ‘suspect exit strategy’ was also put in place, including a risk management plan for long-term disruption, with Probation continuing to manage the suspect (who was recorded to have previously breached a non-molestation order against the victim and was therefore identified as a repeat suspect).

There was also some evidence of the more appropriate allocation of resources to investigate a set of offences as a series, rather than as separate, discrete investigations. The reviewed case was immediately linked to two other cases (and later three as another linked incident occurred) involving the same suspect (one with the same victim, one in which the victim was the current victim’s partner and a witness in the current case), with the crime recording logs being linked to the other relevant crime logs. Rather than being investigated separately, these cases were assigned the same OIC, and the suspect and victim completed interviews for all three cases. There were some slight issues with recording over the three separate crime recording logs; for instance, the victim’s safeguarding referral was recorded on one crime log but not the others, though this was flagged as not being present on one of the logs by their data processing software and subsequently amended by the OIC. This demonstrates some attempt on the part of Force E to consider repeat offending more holistically and make different investigative decisions as a consequence.

Summary and conclusion

The challenges with identifying repeat suspects, using this intelligence during an investigation, and considering the long-term disruption of offending, were similar across the five pathfinder forces. The data collected and the issues identified highlight just how important the topic of repeat offending is and the urgency with which the challenges need to be addressed. The common challenges found across all forces are demonstrative of a general lack of understanding of how repeat offending can manifest, how this offending can be identified, and how the information gained on repeat offending can factor into the investigation on several levels. It is also apparent that there is little understanding of how to disrupt repeat offending on a more long-term basis. Where the knowledge does exist, there is often a lack of capacity from officers to be able to conduct a full and thorough investigation and to take ownership for tackling repeat offending beyond the confines of the immediate investigation.

As noted above, the long-term disruption of repeat offending is particularly important if law enforcement is to have a strategic, meaningful impact on offending from the perspective of prevention. In some other policing contexts this type of disruption is being done in some capacity by Offender Management Units (e.g., High Harm Perpetrator Units22), but there remains a disjunct from officers who have been involved in the relevant case being able to retain ownership – and to have the expertise to do so – of a repeat suspect that requires more strategic targeting. Again, there are opportunities to draw on other offence types to look at more strategic options for disrupting offending, such as the use of Multi-Agency Risk Assessment Conferences (MARACs) to target domestic offending, but again, evidence here suggests available options are not currently being considered and there is no widely established, sex offence-specific equivalent for officers to draw on that operates beyond managing convicted offenders.

The outputs produced in Year 1 of Pillar Two address some of the key challenges identified during the deep dives. The new Investigations Map outlines how officers can incorporate consideration of repeat offending into all points of the investigation. The proposed specialist, team approach should allow officers the time and space to be able to consider both short-term investigation and long-term disruption and provide officers with the support and expertise to critically reflect on decisions made, with appropriate support from both supervisors and analysts to make sure that tackling repeat offending becomes a central part of policing strategy to combat sex offending.

The overall deliverable for Pillar Two is the contribution to the National Operating Model for rape and other sexual offence investigations, which will act as a bridge between relevant overarching strategies and day-to-day policing operations. Our contribution will ensure that the model embeds specialism, a team approach, and the Investigations Map, which includes a suspect-focused approach to investigations alongside both the immediate and long-term disruption of repeat suspects.

Appendix 9: Pillar Three – Embedding procedural justice and engaging victims

Pillar team

Pillar leads: Dr Olivia Smith and Dr Kelly Johnson. Pillar team (in alphabetical order): Dr Oona Brooks-Hay (co-Investigator), Sophie Geoghegan-Fittall, Dr Beth Jennings, Adrian Harris (Project Co-ordinator), Dr Susan Hillyard, Professor Katrin Hohl (co-Investigator), Sarah Molisso, Dr Andy Myhill (co-Investigator), Dr Rosa Walling-Wefelmeyer.

What are the foundations of Pillar Three?

Research demonstrates that the criminal justice process can retraumatise victim-survivors of sexual violence. The Rape Review recognised this, noting that “the trauma of the crime and their subsequent experience leads many victims to disengage from the criminal justice process… We know that victims and the organisations that support them have felt badly let down”^{[footnote 140]}. One victim-survivor we spoke to summarised this problem as: “you’re put in an intrusive situation when you get raped and then the investigation’s just intrusive as well from start to finish” (Victim-Survivor, National Panel 3).

The well-established failings of the criminal justice system in sexual violence have contributed to low confidence in policing. In a survey by the Victims’ Commissioner^{[footnote 141]}, only 14% of 491 victim-survivors agreed that justice can be obtained by reporting to police. The Rape Review sought to rebuild confidence by increasing charge rates and introducing performance delivery dashboards. However, this alone does not meaningfully or comprehensively reflect the needs of victim-survivors. As one victim-survivor put it:

My case ended in a guilty verdict with a ten-year sentence, but when people ask me about my experience it isn’t ‘yay he got put in prison’, it’s ‘well I was let down at this point, and this point, and at this point, and this point’

(Victim-Survivor, Force B).

The leading theoretical framework for increasing confidence in police is Procedural Justice Theory (PJT). PJT is well outlined by Hohl, Johnson and Molisso^{[footnote 142]}, but the core idea is that by recognising the symbolic power of how they treat people, police can offer a sense of justice beyond substantive outcomes. In other words, how police treat victim-survivors, suspects and other members of the public sends a message about their value in society more widely. Victim-Survivors are aware of this symbolic power:

If police are not taking it seriously enough and it is a criminal offence, why shouldn’t society do the same?

(Victim-Survivor, Force C)

The components of PJT have changed over time, but are mostly described as: voice, dignity and respect, neutrality, and trustworthy motives. However, PJT was developed in the context of stop and search, usually drawing on quantitative surveys with members of the public^{[footnote 143]}. Relatively little is known about how PJT applies to victim-survivors, and even less has been explored in the context of sexual violence^{[footnote 144]}. Pillar Three therefore explores how police can embed procedural justice when engaging victim-survivors of rape and other sexual offences.

Pillar Three’s Methodology

We adopted a range of methods in Pillar Three. Nationally, we consulted 42 victim-survivors across five expert-by-experience panels about their views of procedural justice theory. These panels included consultations with neurodivergent^{[footnote 145]}, learning-disabled, male, and racially minoritised victim-survivors.

We also completed four ‘Deep Dives’. The exact data sources varied between each Deep Dive depending on access and availability, but our sample across the year comprised:

• 324 hours of ethnographic observations with 14 investigation units.

• 34 hours of video recorded interviews (VRIs) with victim-survivors in 27 cases.

• 17 x focus groups with police from dispatch, response, and investigation teams.

• 15 x interviews or focus groups with support workers from third sector providers, mostly Independent Sexual Violence Advisers (ISVAs).

• 14 hours of body worn video (BWV) from first contact with 16 victim-survivors.

• 8 x consultations with victim-survivors from local expert-by-experience panels.

• 8 x interviews with senior police managers holding responsibility for strategic decisions about resourcing and KPIs.

• Mapping 108 internal policy and guidance documents relating to victim-survivors.

Police engagement with victim-survivors is inconsistent

In all deep dives and the national expert-by-experience panels, we heard about variation in the quality of police responses to sexual violence. Victim-survivors expected that most experiences would be negative and felt ‘lucky’ if they avoided this:

It’s a postcode lottery and it depends on the officer you get.

(Victim-Survivor, Force C)

[Name]’s had a rubbish experience with the police, what [Name] said is a great experience with the police. It’s like a ‘lucky dip’. If you go on Thursday, you’ll get the lottery. If you go on Friday, you’ll lose.

(Victim-Survivor, National Panel 2)

Most police officers we spoke to and observed wanted to provide a good service for victim-survivors. They highlighted the importance of treating victim-survivors well, listening to them and making them feel heard. Doing a ‘good job’ for victim-survivors was often described as a rewarding part of their role, and officers often expressed frustration at having limited time to engage with victim-survivors and offer them support.

Forces were at different stages of considering the various components of procedural justice, but many individual officers recognised the importance of voice, trust, fairness, dignity, and respect. For example, several officers in Force B recognised that victim-survivors have complex lives outside of the investigation and that safety might be most important:

In many cases you might not get a criminal conviction, but you’ll get a safeguarding result which is great from their perspective. You might also get something through the family courts or even a Domestic Violence Protection Order.

(Observations, Force B)

At the same time, the force was recognised as being “massively statistic-driven” (Police Officer, Force B), with no formal way to acknowledge the importance of victim-survivor treatment.

Wherever we identified good practice, this tended to be driven by individual leaders at the expense to their own wellbeing and despite a perception of limited buy-in from Chief Constables or other senior managers. For example, Force C featured a highly ‘burnt out’ workforce, with managers who were working unsustainable hours to set up nationally leading innovations. These included pilots that we will take forward into year 2, such as: ISVA-police co-location, dedicated courtrooms, community engagement consultants, and Victims Voice forums.

ISVAs and victim-survivors recognised the efforts of many police officers to provide a good service. However, we also heard about deeply traumatic experiences of reporting to police:

I did everything I could and everything I was asked by the police, but dealing with them has been the worst and most traumatic experience. They are inhumane, misogynistic, thoughtless and cruel… I have been violated and scrutinised, my blood was taken, swabs from all parts of my body, people looking at my genitals, touching me, my phone and belongings taken and never returned. My history taken and reviewed and judged for any scraps of information they could use against me.

(Victim-Survivor, Force A).

Therefore, while some victim-survivors are treated well, others are not, and this inconsistency must now be addressed:

It shouldn’t be one out of ten police officers treat a victim with respect, it should be ten out of ten police officers treat a, treat a victim with respect. Ten out of ten victims would call the police again. It shouldn’t be one out of ten victims.

(Victim-Survivor, Force B)

Failure to meet victim-survivors’ Victims Code rights

All forces recognised the importance of the Victims’ Code of Practice (Victims’ Code). As part of the Victim’s Code commitment to information about the investigation, Force D had an internal target of updates every seven days. Force B also checked whether there had been updates at least every 28 days as part of routine case reviews by senior officers. Similarly, they had forms to encourage officers to consider when and how to engage with victim-survivors, for example, whether certain days or times should be avoided because of their religious affiliation, domestic situation, medication, or other circumstances. Awareness and use of the form were patchy, but it showed a push for tailored and considerate engagement of victim-survivors.

Elsewhere, the Victims’ Code right to updates at least every 28-days was considered a luxury:

The Victims’ Code. We never adhere to it.

(Police Officer, Force C)

The communication has been rubbish and even highlighting on the Victim Code the things that they’re supposed to do, nothing’s changed. And they apologise or they, I feel like they minimise and make excuses really, rather than addressing the concerns. But I just don’t feel like anybody cares.

(Victim-Survivor, Force C)

The Victims’ Code also outlines the right for victim-survivors to be supported to understand the justice process and communicate well, however, all forces found it difficult to quickly secure high-quality interpreters and intermediaries.

Privacy rights may be ignored or undermined (often inadvertently):

Our observations and police focus groups across four pathfinder forces suggested that victim-survivors’ phones are routinely downloaded during their video-recorded interview. Officers were doing this to support victim-survivors by ensuring they were not left without a phone, recognising that this would be hugely disruptive:

The officer tells me that the process for booking phones in works quite well and victims are only without their phones for two hours [while being interviewed]. She tells me that they have phones that they can give victims if they are going to be without one, but they have become a bit of a joke in the office as they don’t have chargers or there will be no credit on the SIM. When the manager gave her a phone to give to a victim and she said there’s no charger, they said ‘well she can just buy a charger’. 

(Observations, Force D)

However, this inadvertently means that most victim-survivors were being asked for their phone before either their own or the suspect’s interviews had been completed. This seems to contradict the ‘thinking approach’ set out in R v Bater-James (2020), as it is unlikely that there is a ‘clearly identifiable foundation’ for belief that the phone contains relevant evidence at this early stage. Force D’s internal guidance even encouraged phones to be taken at the point of first response, before the investigation began. Therefore, officers were seeking to put victim-survivors at the heart of their decisions about digital evidence, but this was having unanticipated negative consequences.

The impact of disproportionate phone downloads is well-explained elsewhere (for example, the Information Commissioner’s 2020 report’^{[footnote 146]}). Our findings add further support to this body of work, highlighting the need to carefully consider digital evidence^{[footnote 147]}, particularly where the same scrutiny was not put on the suspect:

Mobile phones were not even invented when I was abused but they had my phone for seven months and messages were used against me in court. I felt as if I was under investigation.

(Victim-Survivor, Force A)

What about his mobile phone, what about his social media, what about digging into stuff that he may have text messages he might have sent, emails he might have written at the time that this happened. They’re not interested and, ‘Well it’s not a reasonable line of enquiry, it’s not reasonable for us to actually look at his social media records, it’s not reasonable to look at his text messages because that was so long ago it wouldn’t be reasonable for us to do’. Well, it’s reasonable to look at mine. You’ve looked at my msn chat messages, you’ve looked at my mobile phone, my sim cards, why not his?

(Victim-Survivor, National Panel 5)

There was also little awareness amongst officers about the need for data access to be ‘strictly necessary’ and not accessible via other practicable means. No forces first used less invasive techniques, such as screenshots. Similarly, there was very limited evidence of routine use of the National Police Chiefs’ Council Digital Processing Notices (DPNs), and some suggestion that DPNs might only be completed at the point of referring a case to the CPS, rather than at the point of access to digital devices.

Chapter 3 of part 2 of the Police, Crime, Sentencing and Courts (PCSC) Act^{[footnote 148]} establishes, for the first time, a clear statutory basis for the extraction of information from electronic devices with the agreement of the device user, and introduces safeguards to protect the privacy of victims, witnesses, and others. The Act makes the use of a written notice such as Digital Processing Notice a lawful requirement and introduces a new code of practice to give police guidance on the use of the powers^{[footnote 149]}. The Government held a public consultation on the code of practice which concluded in July 2022. The final code of practice was laid in Parliament on 17 October 2022 and both the powers, and the code of practice came into force on 08 November 2022.

There remained a sense among police in all forces that some requests for digital evidence and third-party material remain wide-ranging:

For victims we will request everything, school records, social services records, everything, but for the suspect we don’t. We will look at if he’s been accused of a sexual offence before, but we will look at the credibility of the victim, ‘I definitely think we scrutinize the victim more than suspects’. 

(Observations, Force D)

The officer gave an example of a DA [domestic abuse] case she had where they [CPS] had requested details of a previous abusive partner from ten years previously. She said she could understand if it was the victim who had a history of violence as that’s potentially undermining, but being a victim of another abuser is irrelevant. She questioned whether some of the requests were even GDPR compliant. She said the CPS say they are not investigators, but then proceed to dictate the investigation; she said they are too risk-averse and don’t want any surprises or to be ‘caught with their pants down’. She asked the Sergeant if interaction with the CPS had become worse. He said the CPS don’t really want to engage.

(Observations, Force C)

This later quote reflects the comments we heard in all forces that police officers felt the CPS would request sensitive data about the victim-survivor. There was frustration about these perceived requests from the CPS because of the potential negative impacts on both investigation times and the victim-survivor. Force B had undergone recent training that was perceived as highly successful in empowering officers to challenge CPS requests for disproportionate evidence:

I had one for that court case last week, where CPS was saying ‘oh can we get all third party’, and I’m saying, ‘well, what for what? What information have you got that I don’t have, that means that we have to get this third party?’ and they didn’t have anything. And the defence again had said, oh, we’d like to review all third party, and again there was no reason for it. It was just a fishing trip. Um, so I think we are quite good at pushing back on that and saying ‘it’s not, it’s not required’ and like you said, would only go and get the third party if we absolutely need to.

(Police Officer, Force B)

Both digital evidence and requests for third-party material led to delays in the investigation. For example, an ISVA recalled:

We had a client who was raped as an adult reported it to the police, and they requested all her social care records from when she was a child. Now, there was that many social care records it massively delayed the case, because there was boxes and boxes of social care records and educational records, for the police officer to go through. Well, what’s that got to do with the rape that she experienced as an adult? But they insisted that they needed to go through it, and like I say, it caused massive delays in the case. So, they need to be saying, we need to look at this, because we think this, or we think there might be this in that, you know, just give a specific reason.

(ISVA, Force C)

We asked for the consent forms used around third-party materials and were only ever provided forms for the disclosure of forensic medical examinations. We only received the DPN forms for digital evidence from one force upon requesting all internal guidance and forms relating to sexual violence complainants.

Poor victim-survivor treatment was linked to a lack of specialism, resources, and burnout

Echoing the findings of Pillars One, Two and Four, all forces struggled to recognise the additional time and specialism needed to engage victim-survivors of sexual offences, leaving senior managers unable to accurately assess caseloads. This lack of sufficient resources meant that engaging victim-survivors was deprioritised:

Officers say initial contact with victims is usually good, but it then tails off because they don’t have enough time to dedicate to contacting victims.

(Observations, Force D)

The guilty party here, that’s central government, which doesn’t properly fund the police and the CPS. So actually then all the problems that already exist in the police get magnified because they’re under pressure, because they haven’t got enough resource, they’re overtired and overstretched and etc, etc.

(Victim-Survivor, National Panel 4)

Officers also tended to be relatively inexperienced:

We do have a very young workforce now on [unit], I think my shift alone consists of about 80% of people with less than 2 years’ service in. And when a sexual offence job comes in, there’s almost like this panic of like ‘oh my god, what do I do’.

(Police Officer, Force B)

[Officer says] because all the DSs don’t have RASSO experience, they’re teaching each other what they don’t actually know, and the DCs are being supervised by people who don’t understand RASSO cases. It’s like the blind leading the blind. All the experienced officers have left. There is also not enough RASSO training. They don’t have enough officers trained in doing VRIs and they only do them when they think a case is ‘going somewhere’.

(Observations, Force C)

Vacancies in the team were hard to fill, for example several officers in Force B talked about challenges recruiting, and Force D recently advertised posts but received no applications. Officers shared negative views of rape investigation teams held by themselves and colleagues:

An officer says that he was previously in CID and ‘to be honest’ used to consider RASSO the ‘pink and fluffy’ cases as they were victim focused, and that he avoided them in favour of burglary and robbery. He tells me it’s a really different way of working. He says that they are built to want to catch the baddies so in RASSO motivation can be harder.

(Observations, Force D)

When I joined [unit], I was warned off it by a number of officers from across the forces who said, what you doing that for? That’s a punishment posting… And sometimes you say, oh, well, you know, if I’m going to kick and scream about doing X, Y and Z, I’m already at [unit] what more can they do to you?

(Police Officer, Force C)

Compassion fatigue, burnout, and secondary trauma appeared to create further barriers to the sensitive and dignified treatment of victim-survivors:

I’ve decided to care less, as it’s the only way to get through

(Police Officer, Force A)

It got to the point where she came to my house and she actually was nearly crying because she said she couldn’t apologise enough and she was so overworked, and, and then I ended up feeling sorry for her, and I understood her. But, you know, it’s, it’s not the way. 

(Victim-Survivor, Force C)

A high turnover of staff also means that victim-survivors feel passed around:

I think there must be 10 different officers I’ve now spoken to. [laughs] Nobody really knows about the case. Nobody cares

(Victim-Survivor, Force C)

I have, four-year investigation where the client was going through hers, and for eighteen months of that her file was just sitting because there was so much swapping and changing of staff and DIs and DSs and whatever. Her file hadn’t been picked up for over eighteen months, hadn’t even been looked at.

(ISVA, Force D)

Additionally, ‘live’ cases take precedence over non-recent cases in every force, creating delays and a sense of hierarchy for cases perceived as ‘live’:

The answer I got from [asking for updates] was because it’s a historical case, ‘yours is not the only case’. This is from word to word, ‘yours is not the only case we’ve got to deal with. There are 20 other cases sitting in front of you’. 

(Victim-Survivor, Force C)

They are constantly having to respond to live cases (not just RASSO) and that puts all existing commitments on the back burner. They had to put a hundred historic cases into a closed inbox folder for 12 months and not touch them because they did not have the staff to cover it.

(Observations, Force C)

I would get my hopes up to speaking to an officer, that officer tells me they’re going to be here at three o’clock. I build my hopes up. Then there’s a no show from the officer and a couple of days later the officer gets in touch and says, ‘Sorry, but a live case came in.’

(Victim-Survivor, National Panel 5)

Indeed, victim-survivors talked about feeling they were “just a number” (Victim-Survivor, Force C). Others noted being made to feel like a burden, or losing trust as a result of being passed between officers:

I was made to feel that I was an inconvenience, and that there was something wrong with me, for being so anxious.  As opposed to this being completely normal in an abnormal situation.

(Victim-Survivor, Force A)

You’ve got to build on trust with somebody, like, and you’ve had your trust broken with the perpetrator and stuff like that. So, what I think that you’re on about there, like, different officers and stuff, you might have built up your trust in one person, then they get a different officer. I think that’s not right.

(Victim-Survivor, National Panel 3)

Good practice observed against the odds

Despite the lack of experience, high staff turnover, and limited resourcing, most officers that we spoke to were motivated to ‘do a good job’ for victim-survivors. In the observations, we saw examples of officers building rapport with victim-survivors and the police focus groups highlighted the importance of building trust.

In the body-worn video (BWV) and video recorded interviews (VRIs), we noted further attempts to build rapport, although these were not always successful. For example, one BWV showed the police officer ensuring she sat down so that she was at the same level as the victim-survivor, she introduced herself and the process, then offered space to ask questions (BWV Review, Force B). In other cases, the BWV showed awkwardness and a focus on completing evidential requirements, which we will unpack further in year 2.

Force D had particularly good practice around their VRIs. While there were still areas for improvement, the interviewing officers offered a chance to talk about the ‘whole story’ and gave space for victim-survivors to talk about the wider relationship contexts that explained the interpersonal dynamics of the alleged abuse. They also asked victim-survivors what they were feeling at each point of the abuse, providing further space for voice. Officers also tended to ask victim-survivors what they wanted out of the investigation, which may be difficult to answer but shows an attempt to engage with victim-survivors needs beyond simply a conviction outcome.

There remains some evidence of myths about rape and victim-survivor credibility

Echoing the findings of Pillar One, many officers highlighted the importance of challenging rape myths, understanding different victim contexts, and looking for vulnerability rather than making ‘credibility’ judgements about a victim-survivor. This was particularly notable in Force B, where rape myths were strongly acknowledged and challenged. However, a significant minority of officers elsewhere openly made statements rooted in myths:

[Unit] should investigate real rapes, S1 and 2s [‘stranger’ rapes], and DA [domestic abuse] rapes, which clog up the system, should be investigated by the DA team.

(Police officer, Force A)

Others said that while they knew the realities of rape, juries are influenced by them and so they still influence case outcomes:

The more experience you’ve got in, you know, in court processes and things, you know, the more you know exactly what the defence are gonna jump on… If there’s any, kind of, anything, that victim’s personality or their lifestyle, that they can jump on to undermine them, then they will.

(Police Officer, Force D)

This contradicts official policy, which states that referral to CPS should be based on what a reasonable jury might think, not a typical one.

Force C created a ‘triage’ system to manage caseloads, but it had the potential to reinforce rape myths because of a focus on ‘case viability’ linked to perceived victim-survivor credibility. Police officers in the investigation teams acknowledged there was a risk with this approach:

I asked how the classifications were determined and she suggested it was subjective. She said it should really be the Sergeant, but in reality it is them. She said any investigator that doesn’t have a ‘gut feeling’ about whether a case will proceed or not is a liar.

(Observations, Force C)

Observations within this force demonstrated other problematic attitudes among some of the investigation teams, including about cases that ended in a conviction:

The officer said there are two people involved and the man could be falsely accused and have his life ruined too. He said that men just think they’re having a good night, whereas the women might feel violated. He said that there are other women who cheat on their boyfriends, but claim it was rape after when their boyfriends ask questions about what they were doing. He showed me several videos of women drunk and acting ‘flirtatiously’ with men. He said they could never get convictions because they were ‘leading the men on’ before the sexual act. The officer said that most of his work was tracking women’s nights that they couldn’t remember. 

(Observations, Force C)

The officer described how in messages / the chat, the suspect had shared a picture of the victim sleeping and she was sending him messages saying, ‘I love you, I love you’ and then ‘suddenly “you raped me”’. It was this difference in tone that seemed to be what made the officer doubt the case… I ask what might have persuaded the jury in the end and it seems that the victim ‘came over better’ than he did, she ‘cried in the right places’ and also ‘came from a good family’ whereas he was ‘a bit greasy’ and wore ‘trackie bottoms.’

(Observations, Force C)

There was a belief among senior management in Force C that “the system is clogged up with false allegations”. This may be why they (as well as Force D) began several victim-survivor interviews with a discussion about truth and lies. This discussion should only occur where the witness is under 18 or has communication / learning disabilities that could impact their ability to distinguish between fact and fiction:

It is important that we all tell the truth​

(VRI, Force D)

I am not going to insult your intelligence here, but obviously you understand the difference between telling the truth and telling a lie and the consequences of those actions.” The officer asks the victim-survivor to give a simple example of the consequences of telling a lie “Obviously … [I know the] difference between bullshit and telling the truth… I get what you are saying, I know lies, I know the truth.

(VRI, Force C)

Support services are sometimes valued and sometimes not

All forces had ambivalent relationships with their local third sector support services, although we found examples of excellent partnership working (and the subsequent benefits for victim-survivors^{[footnote 150]}). For example, Force D used crisis support workers to provide emotional help for victim-survivors during the video recorded interviews. Specialist support, particularly ISVAs, were noted by some officers as extremely helpful for engaging victim-survivors, providing “tremendous support” (Police Officer, Force D). However, communication was variable and depended on individual officers, with updates being described as either “really thorough, very basic, or not at all” (ISVA, Force D).

Some officers had limited understanding about support services or the distinctions between different support roles. For example, officers from Force B talked about Independent Domestic Violence Advisors when they meant ISVAs, and some victims were briefed about ISVAs as providing counselling:

I then overhear a phone call to a RASSO victim in which she seems to be offering a referral to counselling. I hear her saying that there is a ‘lack of evidence in this case’ and then later ‘I am sorry to upset you about that’. When she later returns to the subject of counselling, she mentions [service] and I wonder how clear it is for the victim whether it is an ISVA or a counsellor she is referring the victim to. (Observations, Force C)

To address this, ISVAs tended to offer police training, but the high turnover of officers meant that this did not solve problems of awareness. High turnover also reduced partnership working because it meant that personal relationships could not be cultivated:

I wouldn’t even know where to find an ISVA’s phone number.

(Police Officer, Force B)

A lot of the existing officers are no longer there and there is just always a high turnover of new officers, some of them don’t understand our roles, some of them don’t know who we are. 

(ISVA, Force C)

In some locations, partnership working with ISVA services had completely broken down, as police officers felt that challenges to their practice overstepped the ISVA role. Such advocacy was viewed as personal criticism of the officers, rather than being an ‘essential element’ of the ISVA role^{[footnote 151]}.

Limited funding and high demand also created extensive waiting lists across the force areas. Appropriate and sustainable funding is needed across the third sector. There was an almost total exclusion of independent ‘by and for’ services, and when asked about local services, police officers focused on SARCs and then ISVAs. ‘By and for’ services are more likely to be accessed by minoritised survivors, so it is important to widen partnership working to better engage the whole community. There are also gaps in specialist support, for example our consultation panel with neurodivergent and learning-disabled victim-survivors highlighted a lack of provision tailored to these people.

Victim-survivors’ thoughts on existing conceptualisations of procedural justice

Away from the deep dives, we consulted victim-survivors in national ‘expert-by-experience’ panels about the traditional concepts in procedural justice theory (PJT): Voice, neutrality, respect and trustworthy motives. Their insights have raised many questions amongst our team, and we will spend a further three months to fully unpack the insights from these experts, as well as similar discussions held with police officers.

What emerged clearly from the consultations was that most victim-survivors had upsetting and/or frustrating experiences with the police. Many felt that the police had failed to treat them with any justice or to engage with them on a human level:

You’re going through a process where your, your decisions, and your rights, and your dignity is taken away. You feel, that’s what you feel like, you’re re-traumatising somebody over, and over again.

(Victim-Survivor, National Panel 3)

I don’t mean to be horrible to the police, but I don’t think they take the person as a whole person.

(Victim-Survivor, National Panel 1)

Victim-survivors’ thoughts on ‘voice’

Perhaps the most widely accepted element of procedural justice in the existing literature is ‘voice’, conceptualised as providing space for victim-survivors’ to express their viewpoint and give feedback^{[footnote 152]}. Our consultation panels suggested that voice is indeed important:

Choice is so important for all survivors because when we were sexually assaulted, abused, raped, molested, however you want to phrase it, whatever type of sexual assault crime was done against us, our choice over our body, our personal choice was taken away from us. If anything, to feel fully heard and to feel fully understood and to feel justice, whether we report or not, it’s give us the choice back. Give us a choice back.

(Victim-Survivor, National Panel 5)

Most discussions were rooted in examples of when victim-survivors felt their voice was ignored:

[If they’d listened, police] wouldn’t of showed up at my door. They would have gone through my ISVA as told and directed, because that’s one safety buffer I really needed because I don’t get on well with the police.

(Victim-Survivor, National Panel 5)

Most, some people, it takes them time to process information. Don’t assume that everybody here processes information at the time. It takes some people some time. You have to repeat yourself. So you treat everybody individually, know how they process information.

(Victim-Survivor, Panel 2)

This meant that ‘voice’ alone was insufficient, because it had to result in action and “the other person has to listen”

(Victim-Survivor, National Panel 5).

There is also a need to recognise different communication needs, particularly for neurodivergent or learning-disabled victim-survivors. One ISVA also highlighted language barriers that might not always be recognised:

Certain survivors have specific needs around language, and I’ve seen in one case… they wouldn’t use the interpreter when they’d call her. And it’s ‘no, but she understands’, and I would say, ‘well, she doesn’t. I do sessions with her with the interpreter present, and there is a communication gap there that you’re not paying attention to’. So that’s something we really had to push back on … just getting the police to pay attention to someone’s specific needs is not so straightforward.

(ISVA, Force A)

Victim-survivors’ thoughts on ‘dignity and respect’

The second component of PJT is dignity and respect, conceptualised by Holder^{[footnote 153]} as meaning that victim-survivors are valued as a person (not just a witness) and that the allegations were taken seriously. Our consultation panels agreed with the value of ‘dignity and respect’, but highlighted the risk that it could be enacted in superficial ways:

Respect isn’t just about overt politeness. It isn’t just about the words the police use and how they say those words, it’s also about the decisions they make and the reasons they use to justify them.

(Victim-Survivor, National Panel 4)

Similarly, some victim-survivors felt that police were too far from being able to act or think respectfully:

I don’t think you can ask or talk about respect in terms of the police. I think we need to go a step back and ask them, ‘Do you actually see people who walk in a police station as human beings?’ And I don’t think the answer would be yes. And if you don’t see people as human beings, then all the other stuff that we’ve spoken about today is not going to come

(Victim-Survivor, National Panel 4)

Victim-survivors’ thoughts on ‘trustworthy motives’

Mazerolle et al^{[footnote 154]} defined the next component of PJT, trustworthiness, as the extent to which citizens feel police are sincere and open, as well as motivated by the best interests of the public. Victim-survivors agreed that trust matters, but mostly talked about the lack of trust they now had:

When you lose trust in the police as well, I don’t think it comes back neither. Like, you struggle for years and years after as well, reporting things to the police.

(Victim-Survivor, National Panel 3)

How are you treated is the main thing, if you’re not treated right, then how can you trust the police? If they disrespect you and you find they say one thing, and then they don’t update when they said they would. Yeah, that’s really, really awful. And really makes you feel let down and insecure and maybe unsafe, if you don’t know what’s going on. 

(Victim-Survivor, Panel 1)

This highlights the links between trust, respect, and safety. Feeling unsafe was a theme across all victim-survivor experiences, so it is important to consider how PJT can reflect this.

Victim-survivors also had little confidence that police forces want to achieve meaningful change, as well as little trust in the ability for recommendations alone to improve policing:

Can I just ask; when you get your findings and you go to the police, is it just going to be recommendations again? Because this is what we see, like you know the independent enquiry, it’s just recommendation after recommendation and the lessons learned, and nothing’s changing.

(Victim-Survivor, Force C)

Like it doesn’t even make sense to teach the police that they need to respect people, or that trustworthiness is important, because they don’t understand what it means. They apply their own definition which they just made up in their heads somehow, and yeah, they just perpetuate their old patterns.

(Victim-Survivor, National Panel 4)

In order to trust police motives, victim-survivors therefore talked about the need for police to open themselves up to scrutiny, be willing to be held to account, and to do more than tinker at the edges of the status quo. Some victim-survivors also made it clear that they were fatigued with rhetoric about taking sexual violence seriously without criminal justice agencies also being held to account. For example, victim-survivors repeatedly talked about a need to consider accountability of police and/or the Government:

Victim-Survivor 1: We have a police problem…

Victim-Survivor 2: That is bigger than us and sexual violence

Victim-Survivor 1: Yes [name], we are not holding the police to account for many, many different things (National Panel 4)

Let’s give our, our government a list of women who they are responsible for the death of because they didn’t treat them properly…

(Victim-Survivor, National Panel 3)

A final element of trustworthiness in the existing literature is competence^{[footnote 155]}. This was important to victim-survivors, who noted that while good treatment is important, a police officer still needs to be effective:

They didn’t treat me badly in terms of like personal interactions, they weren’t rude to me, but they were completely hopeless in terms of actually conducting a competent investigation

(Victim-Survivor, National Panel 4)

I didn’t feel like I was in a position where I was being unfairly treated because of who I was, I did feel like they are useless at what they do.

(Victim-Survivor, National Panel 4)

Victim-survivors’ thoughts on ‘neutrality’

Perhaps the most contested element of PJT is that of ‘neutrality’, described by Murphy and Barkworth^{[footnote 156]} as unbiased decision-making that consistently applies legal rules. Victim-survivors talked about their fear that police decision-making would be biased, suggesting the importance of ‘neutrality’ as defined by PJT:

I have history with the police. As a teenager I was in and out of prison and things like that so when the police did nothing basically about the rape, I thought to myself, ‘Well, I’m in a lower class of third or even fourth class, you know, because I’ve been in trouble with the police, the police aren’t interested anymore about me.’ That’s the assumption I had.

(Victim-Survivor, National Panel 5)

My family are all from gypsies and travellers, so obviously going in, they already have a, a jaded view of the gypsy/traveller community, and then obviously me going in, making a report, it was a bit of a, ‘are you sure? Did it happen? Is that not normal with your lot?’ So, I think there has to be a massive awareness for different cultures.

(Victim-Survivor, National Panel 3)

I have to point out this is a societal prejudice against people with a learning disability; you know, people are repulsed by them… I’ve heard officers be like, ‘you know what, she should be lucky that happened, really, because no one else can find her attractive’.

(ISVA, Force A)

However, ‘neutrality’ had negative connotations for victim-survivors in all consultation panels. For some, it had connotations of being cold and uncaring:

Victim-Survivor 1: I just think that neutrality is hand-in-hand with fairness. It’s very, very similar. You think, this is fair, that’s fair. The preconceived ideas. You need someone who’s completely neutral to any kind of situation who’s not coming in with any opinions. So, they’re completely neutral to anything that you tell them.

Victim-Survivor 2: I don’t know, I don’t… to me, it’d come across as a different way, like, somebody who’s remaining neutral is like somebody who’s just having, like, that monotone voice, like, it’s just, like, another procedure that they’ve got to do. (National Panel 3)

Neutrality is also widely criticised in academic literature as a male-dominant construct that does not account for different starting points and the barriers that may be faced by different victim-survivor groups (Smith, 2021). This was reflected in the concerns of some victim-survivors that ‘neutrality’ suggested that decision-making can be objective:

I just don’t think it’s realistic to say that the police will be completely neutral, because virtually nobody else in the world is, unfortunately. So, there’s a lot you can do, to try to make sure that they treat people fairly and with an open mind, and don’t make assumptions. But I kind of think it’s better to start from the viewpoint that actually everybody is biased in different ways, and generally not everyone’s always aware of it, and then try to row back from that.

(Victim-Survivor, National Panel 4)

Victim-survivors’ concerns about safety are not currently included in PJT

Our initial discussions also suggest some important concepts that are not explicitly addressed within PJT. Safety (for themselves and others) was a significant concern for many victim-survivors throughout the whole process and beyond, regardless of how recently the offence(s) took place:

I’m really scared; I’m, I may get killed, I know that.

(Victim-Survivor, National Panel 2)

I’d argue that you’re in more danger after your case is finalised than you are during it.

(Victim-Survivor, National Panel 3)

Victim-survivors welcomed it when police offered measures that contributed to their safety. It was apparent that victim-survivors often sought protection from the police, but there was a perception that officers do not always understand a victim-survivors’ fears about safety:

So I had gone from feeling, well at least I feel safe because he doesn’t know my name now, to being hugely anxious because actually he’d known my name all along. They gave me a panic alarm, which was amazing, but I was made to feel like I was being overly anxious about it

(Victim-Survivor, National Panel 4)

I think respect is to not make you feel crazy. Like, after, whether you get a conviction or not, whether that person’s in prison or not, you still have that fear that someone’s gonna come for you.

(Victim-Survivor, National Panel 3)

I felt as if they weren’t, weren’t protecting me from him, because I live in a small town.

(Victim-Survivor, National Panel 1)

Victim-survivors’ fears extended beyond physical safety, to concerns about their ability to meet the demands of the criminal justice process and to cope with the impact it might have on them:

The fear of going to court, being ripped apart because you can be, you can be vague, you may not be able to give dates or times or all these bits that the police will want you to, the police can’t supply it. And also the fear of being triggered and having lots of flashbacks, and having to go through all that and not necessarily getting the therapy and support you need to deal with all of that.

(Victim-Survivor, National Panel 1)

While PJT does not explicitly include safety as one of its principles, a procedurally just approach was viewed as central to securing victim-survivor safety and survival:

The role of the police is to protect the public and, essentially, to protect life. Now, when they compromise on any of the aspects we discussed – respect, dignity, personal approach, voice, time, resources or fairness, when they bring their own bias and judgment, they essentially compromise people’s very survival.

(Victim-Survivor, National Panel 4)

As a pillar, we will continue to reflect on the role of safety and whether it is best understood as an outcome of procedural justice, or a component of procedural justice itself.

Acknowledgements

We thank the officers, specialist support service workers and victim-survivors who participated in our work during Year 1. We especially want to acknowledge the invaluable contributions from victim-survivors who generously shared their experiences with us. Thank you also to our whole research team, who have worked incredibly hard.

Suggested Citation

Smith, O.; Johnson, K; and Brooks-Hay, O. (2022). Detailed Learning from Pillar Three of Soteria: Embedding procedural justice and engaging victims. Part of Operation Soteria Bluestone.

Appendix 10: Pillar Four – Officer learning, development and well-being

Lead: Dr Emma Williams Team: Dr Nicky Miller; Richard Harding, Dr Arun Sondhi, Dr Linda Maguire; Jenny Norman; Daniela Abinashi and Dr Rachel Ward

Introduction

Over recent years, changes to police education at entry level and processes of continuous professional development (CPD) have been central to the debate about the professionalisation of the police service. In relation to improving the police response to, and investigation of rape and other sexual offences (RAOSO), the need for more effective training has been a repeated theme throughout government, third sector and academic reviews aimed at exploring the continuing problem with attrition and unequal distribution of justice at the police stage of the criminal justice process. Upskilling officers and ensuring access to the specialist knowledge required to change the attitudes and behaviours of investigating officers is critical for a more professional approach to this highly complex crime. Whilst some police organisations have made structural changes to establish specialist units and ‘ring fenced’ officers to investigate rape, the assumption that improvements to or an increase in police training will lead to improved responses has yet to be explored in the research relating to this area of policing. Furthermore, despite there now being a standardised curriculum in the College of Policing’s Serious Sexual Assault Investigation Development Programme (SSAIDP), local police areas develop the training content and delivery style independently with no oversight from the College itself.

The well-being of officers is another area of national interest at the current time. The National Police Wellbeing Service (NPWS) was launched in 2019 to provide support and guidance for police forces across England and Wales. It aims to make improvements and build organisational health and officer well-being through several evidence-based activities and resources. Part of their work has been to develop an evidence based ‘Investigator’s Toolkit’ which signposts officers to support in regard to building resilience and welfare needs. Whilst there is a wealth of literature on the trauma experienced by officers involved in complex crimes^{[footnote 157]}, there are limited insights into the influence of organisational factors on officer stress, or indeed any barriers that might impact on decisions to access well-being support offers – particularly in relation to investigating RAOSO.

Investigating such complex crimes as RAOSO requires specialised and highly skilled investigating officers who are equipped with the relevant knowledge to deliver in their role^{[footnote 158]}. As a result of this Pillar Four is the central enabler to delivering more professional investigations and effective victim care through the new National Operating Model which is being built on the foundations of Operation Soteria Bluestone.

The aims of Pillar Four were:

• To understand the extent to which officers are equipped with the specialist knowledge required to investigate and respond RAOSO and their ability to apply it in practice

• To explore the support for the welfare needs for RAOSO investigators

• To understand the drivers of welfare issues for RAOSO investigators

• To understand the limitations to learning in the field of RAOSO

Background

Valuing and cultivating learning and knowledge is critical for organisational development and improving service delivery^{[footnote 159]}. Enabling practitioners to access learning resources and CPD is central to both, individual professional identity and organisational commitment^{[footnote 160]}. Indeed, notions of professionalism link to narratives about organisational legitimacy, trusted and autonomous decision making, competence, and self-identifying as a professional^{[footnote 161]}. However, research has found that access to learning and CPD is sporadic in policing and difficult for many^{[footnote 162]}. Set against the backdrop of increasing demand, complex victim needs^{[footnote 163]} austerity and a shortage of experienced investigators^{[footnote 164]} the ability for officers to access any learning offered is further restricted.

Therefore, the theoretical framework on which Pillar Four is built is from the literature on organisational justice. Pillar Three and the focus on victim treatment is firmly built on the notion of procedural justice but there is wider evidence that links the delivery of fair treatment externally (for example to victims) to officers’ perceptions of fairness and equity inside organisations. Organisational justice recognises the requirement of inclusive participation and the role of the workplace to empower people to deliver their roles effectively and in professional way (Nilsson and Townsend, 2014). Several studies describe how fair and just organisations are associated with feeling valued, loyalty, a commitment to organisational goals and an emotional attachment to their police force^{[footnote 165]} Furthermore, the benefits of accessing and applying specialist knowledge and CPD include improvements in officers’ confidence to deliver work roles and an increase in professionalism^{[footnote 166]}. Being equipped with the specialist knowledge required to deal with the complexity of RAOSO is critical for both victims and the well-being of officers and, as researchers^{[footnote 167]} argue, this needs to be framed within the context of organisational justice. Competence and self-efficacy are linked to professional knowledge attainment^{[footnote 168]} and can counter the development of compassion fatigue and burnout^{[footnote 169]}.

Many authors have debated the complex structural and cultural issues related to what counts as knowledge within policing and much of this relates to the perceived undermining of experiential knowledge which officers gain over their time in policing^{[footnote 170]}. As a result of this Pillar Four included the recognised positive outcomes of reflective working practices as a further way of both capturing learning from officers themselves and for applying more formal learning in a practical and real way^{[footnote 171]}. As part of the application of organisational justice theory Pillar Four considered officers’ perceptions of the opportunities offered by the organisation to effectively apply their learning in practice as part of this review.

Methodology

As a result of the research questions involved, Pillar Four took a mixed methodological approach by using both qualitative and quantitative data collection methods. The research questions are:

• What learning material and delivery style is involved for RAOSO officers at a local level and how does the material attend to the SSAIDP curriculum and evidenced good practices of delivering learning to the police?

• What are the barriers, and enablers to officers being able to access learning and well-being support?

• What resources are available for RAOSO officers locally?

• Is there a relationship between the offer of learning in the organisation and the well-being of officers involved in RAOSO investigations?

• How are officers enabled to apply and reflect on their learning in practice?

Qualitative research:

Across the four pathfinder police forces a total of 28 interviews and 23 focus groups, comprising 119 participants were undertaken through a purposive sampling strategy developed in collaboration with local police Pillar Four leads. Due to the nature of the research questions specific officers and roles were required, and key individuals were therefore targeted within the sample. The Interviews were held at strategic and operational levels with officers and staff in the Senior Leadership Team, learning and development leads, well-being co-ordinators and business architect teams. Focus groups were comprised of first response officers, RAOSO investigators, supervisors from PC to DCI level. The table below outlines the numbers of participants in each data collection activity.

Table 1: Year 1, Pillar Four research activity – participant numbers for interviews, focus groups and survey

In addition to the interviews, force learning materials and all local well-being resources were reviewed for content. We were also invited to observe training sessions in all force areas.

The recorded focus groups and interviews were transcribed, coded, and thematically analysed using NVivo^{[footnote 172]}. The authors used an open coding framework to organise the analysis which enabled the identification of the top-level themes or conceptual categories arising. Two researchers analysed the data to ensure independent inter-rater reliability.

The learning material reviewed for each course included:

• Lesson plans

• Course material (i.e., PowerPoint slides)

• Supporting reference material

It should be noted that the learning material supplied for the non-specific sexual offences courses was only that which related to RAOSO investigations; this did include material around the investigative process, the building blocks of an investigation, decision making and sources of bias, as well as case law and forensic processes.

The learning material was analysed and considered against the SSAIDP curriculum, the current evidence based related to an understanding of RAOSO (such as rape myths, victim vulnerabilities and trauma). The style of delivery was reviewed against the evidence of effective pedagogical approaches for delivering learning. It should be noted that not all the learning was observed and therefore assessments were made based on the review of materials without seeing it delivered in action. The research team are aware of the resulting limitations in relation to observing the style of delivery for those courses.

Quantitative research:

The primary focus of Pillar Four’s quantitative analysis was the development and administration of a Learning and Wellbeing Survey applied systematically to the five pathfinder police forces. The Bluestone team coproduced the survey in partnership with ex and serving police officers to incorporate components relating to the work-learning environment using validated tools developed in organisational behavioural domains. These included additional questions on the training received by RAOSO officers. Health and well-being sections include questions from the NHS Staff Survey^{[footnote 173]} to allow for direct comparisons. Additional questions focus on burnout symptoms using a licensed and validated schedule measuring emotional exhaustion, depersonalisation and personal accomplishments^{[footnote 174]}. Other questions focus on perceptions of attitudes towards victims/survivors, organisational priorities and barriers to service delivery.

Underpinning the survey analysis are theories relating to Job-Demands Resources (JD-R), whereby burnout and work-related stress is the pathological indicator for jobs with high demands and low available resources^{[footnote 175]}.

The survey was administered online and 538 responses were received across the five police forces (37% response rate). There was some difficulty in determining the denominator of police officers eligible. This was due to operational reconfiguration towards generalist models and some officers may have been invited to respond to the survey but did not have a RAOSO caseload, thereby dampening the survey response rate. Analysis of survey data in conjunction with qualitative evidence allowed for a deeper understanding of the interplay between work demands, learning climates and increased depersonalisation, manifested as compassion fatigue towards victims^{[footnote 176]}. Furthermore, we used multivariate analyses to understand the correlates of the three components of burnout.

Summary of findings

Whilst there were some nuances between the 4 pathfinder forces, the analysis revealed three core strategic themes to the research undertaken in Pillar Four. Whilst these themes are discussed separately in this report there are clear mutual dependencies between them.

1: The de-professionalisation of the RAOSO officer role

2: Structural and systemic barriers for enabling learning and organisational health

3: Individual resilience to ‘cope’ with the lack of resilience for officers provided by the organisation

The de-professionalisation of the RAOSO officer role

Capacity building not capability building

Due to demand levels and a general lack of detectives within policing there is a tendency to make transactional decisions about building capacity within teams rather than considering officers’ competence and capability to deliver professionally in their roles. This is against a backdrop of the Police Uplift and direct entry detective programmes which have resulted in very young in-service officers being placed in teams investigating highly complex crimes without the learning input required. Following completion of their PIP2^{[footnote 177]} course and the 12-month work-based assessment of competence portfolio, required to be certified as a detective, officers can then access the SSAIDIP and investigate RAOSO. However, the research found, across the sites, officers who had received no specialist training to investigate RAOSO.

…I’ve got somebody that I’m tutoring today that did their course [PIP2] in June, and they’ve already been allocated a RASSO job. That’s how strapped we are at the moment. I think it’s irrelevant whether you’ve got that extra training or not…

(Force A)

Individual and role appropriate training requirements are not being effectively identified through supervision or being prioritised. Furthermore, access to any learning offered within the forces was inhibited by concern about the time officers must take out to attend courses and how this impacts on the demand they are trying to manage. The survey revealed across the 4 pathfinder police forces that over 80% of respondents do not feel they ‘have time to participate in training’ and yet just 2.7% feel ‘I do not need to participate in learning and development because I am competent in my job’. This shows clearly how much officers would like to receive extra learning input. As this officer states:

Obviously, they are working detectives on the frontline of investigations with caseloads, workloads dealing with serious physical and sexual violence, and they just can’t be released in large numbers in order to fulfil the training and get their accreditation. Because obviously, that would leave nobody to do the job. It’s about finding that sweet spot of the training offer in comparison with what the force can support

(Force D)

This results in officers feeling that there is no value placed on equipping officers who work in this area with the learning they need. This is further impacting negatively on recruitment and retention within the teams.

The workload is excessive, and staffing is too low. This causes officers anxiety every day and I know the majority of staff are working longer hours without pay just to try and keep their crimes in some sort of control. I am also concerned that we are now having student officers coming into the dept dealing with high risk investigations and not having the experience and the dept not having the officers to support them on a daily basis

(Force C)

‘I’m smiling out there all the time. I’m Mr Positive. We can do this. We’re part of a team. In here (office), I have my head in my hands most of the time. I stare at a sheet that’s full of vacancies or people on secondment and attachment and think, I don’t know how we can do this’ (Force B)

The challenges noted by investigating officers were replicated amongst those in supervisory roles who have strategic oversight of the investigations and the needs of their teams.

…There is no teaching, no training and I think there could be, maybe a better input on what is expected of you as an OIC. But you literally just have to do or die…

(Force C)

In the absence of formal learning sources practitioners are learning from their own and their peers’ trial and error experiences with limited time for reflecting on and evaluating their decisions. This approach to learning is unlikely to address culturally entrenched practices and beliefs around the complexities of dealing with RAOSO. For example, rape myths, victims’ vulnerabilities and investigative approaches.

Omnicompetence and lack of specialist knowledge

RAOSO differs from other investigations as it involves an assessment of consent and, often involves individuals that are known to each other. Victims of RAOSO typically have complex vulnerabilities which require specialist understanding, particularly in the context of relating such factors to attrition^{[footnote 178]}. Currently, within the context of RAOSO there is, predominantly, a reliance on omnicompetence, with officers and investigators operating as ‘investigative generalists’. It is through the provision of expert knowledge that individuals are enabled to effectively deal with the ambiguity presented by these cases, ask critical questions and reflect on their practice. There is a concerning lack of specialist knowledge within investigation teams and there is no question that this is impacting on attrition, the support given to victims and officers confidence in their role. The survey showed that just over 30% agree ‘my organisation provides me with training in advanced skills’.

The department is on its knees and gradually getting worse. There [are] hardly any experienced officers left within the department and the ones that are left are actively seeking new jobs. We deal with some of the most horrific cases within the force, yet we are given no time to do a thorough investigation and often they are left in the hands of probationers with little or no experience at dealing with RASSO offences

(Force C)

This can result in a reliance on internal police knowledge about what constitutes real rape when officers make assessments about victim credibility. The survey found that over 30% of respondents felt they could tell whether a case will progress based on their early assessment of victim credibility.

…You apply your own learned experience from your evidential test, your public interest test as you would do with any other case; knowing your basic points to prove and the nuances of an investigation from your basic training. Whilst we have specialist investigators who may pick up specialist skills along the way, at supervising level if you hadn’t actually done that, then it’s very little traction to pick up…

(Force A)

A further negative outcome of the lack of specialist knowledge was the incongruence not being able to effectively deal with victims’ needs had for officers with their own personal desire to do their best for victims of this crime. This incredibly powerful quote sums up this section perfectly and highlights the extreme issues that RAOSO officers are facing:

I do not know how [much] longer I can manage all this on my own without an investigation going seriously wrong or me becoming seriously ill. I am exhausted and the force frankly could not give two hoots and just want more and more from me…I don’t think they know what a terrible state we are in. We are not giving the victims a good level of service and we are destroying our staff

(Force C)

The delivery and content of formal learning is inconsistent, lacks sufficient assessment and there is a clear absence of evidence base to underpin learning (and practice)

The review of the learning material revealed significant gaps in knowledge relating to academic research, didactic presentations which did not link key relating factors together (such as victim vulnerabilities and attrition), a strong emphasis on functional skill attainment, limited real life examples or local data for officers to consider the learning in context and opportunities for critical engagement, reflection and application of learning were limited. The offer of learning was also varied across the sites. One force had truncated courses together to at least deliver something and 3 (and at the time of writing this report all 4)^{[footnote 179]} had paused the delivery of the SSAIDIP to focus on other areas of training need. Delivery approaches varied but broadly the integration of (limited) formal and (predominantly) workplace learning was not evident nor was an evaluation of the learning delivered.

Aside from one force which had more positive views from officers on the learning climate, less that 45% of respondents in the survey felt that the learning and development offer from their force met their needs. Conversely, over 90% of respondents said they would like to participate in learning events if they were available.

Whilst the perception of the learning offer in Force D was more positive than others, the research revealed unequal opportunities to access learning across different teams. A recently reconfigured rape investigation team in Force D felt overwhelmed by demand and limited resources whereas other standard investigation teams did not share this view. These two quotes highlight this:

Personally, speaking I’ve got massive resource issues within my teams. We’ve lost a lot of experience. So, we commenced the teams in [Month], then there was uplift in specialist crime who happened to be based within our BCU, which then meant that I lost a lot of experienced officers to other roles. So, we’re struggling as far as resources are concerned. And whoever we now have in are obviously going to be less experienced. We’re looking at some of those who have very little experience within investigative environment at all. So, we are struggling

We’ve had quite a number of abstractions when I first inherited the team in [Month] because we’d had a number of serious incidences and major crime had taken a lot of our resources plus a couple of vacancies. But since the new year I’ve been quite well backed by my SMT, and I’ve only got the one vacancy left on the team at the minute. And probably 80% of my team are PIP accredited as well, which makes a big difference. They’ve completed their portfolio. They’ve got a fair understanding of all types of investigations as well as experience of dealing with rapes… he may not have as much support or resources as perhaps we’ve got somewhere else

Ultimately whilst this force had in place a specialist team ringfenced for RAOSO, the officers were not given the specialist skills and knowledge that forms the basis to what specialism is.

In summary the lack of specialist knowledge offered across the sites has de-professionalised the role of RASSO investigators. This perceived lack of organisational value placed on both upskilling officers and seeing the value of this knowledge for purposes of justice is impacting on officers decisions to take on these roles. The connection between competence, officer confidence and the provision of and ability to access learning is compounding the use of cultural knowledge as officers learn from ‘trial and error’ experiences. Pillar Four argues that, in line with organisational justice theory, the workplace has a responsibility to enable professionalism in this complex area of policing.

Structural and systemic barriers for enabling learning and organisational health

Transactional decisions made with limited time for strategic understanding or potential risks

In all sites there was a reliance on transactional decision making with limited strategic consideration to the enablers required to drive successful reform in the longer term. It is this that has impacted over time on the lack of real change in the policing of rape and the historical problems with attrition rates at the police stage. Indeed, individual, and organisational learning aimed at changing the behaviour of officers is one of the main enablers to genuine change.

All forces had at some point over recent years made changes to rape teams, usually based on recommendations from reviews and/or HMICFRS inspections, the extent to which, in Force C had created confusion and fatigue for officers. As this officer explains:

You may have found already with surveys, staff feel fatigued. That’s not just [CONSULTANCY FIRM], in [FORCE C], we like to keep asking people for surveys. We like to keep changing things. I get if you stand still, you never get anyway, but equally, staff are tired. And some may struggle to differentiate between [CONSULTANCY FIRM], MOP CPRL, service transformation, continuous improvement, we’ve had lots of things going on

(Force C)

Whilst this is implicit in all of the forces, Force B have instilled a strong performance culture and layers of bureaucracy which were perceived very differently between the senior team and the operational officers. There appeared to be a lack of understanding concerning the impact of the force processes on the ground on both victims and officers. These two quotes evidence this:

Strategic: ‘I think the supportive leadership model that we’ve got in [FORCE B] is the right one. I’m not saying it’s always perfectly used, but on the whole, recognising the impact upon individuals, I think we are a lot less hierarchical in terms of commanders’

Operational: ‘The way things are in our force at the minute, we’re measured on absolutely everything. And I mean measured on your victims, your crimes, your outcome rates, your resolved rates, your created crimes…we are measured on absolutely everything and we are criticised for everything as well. … And I think that has a massive negative effect on our officers’ wellbeing’.

Officers felt that these processes of ‘management’ combined with the level of demand influenced decisions about case prioritisation. Furthermore, findings from the survey also suggest an inherent tension between a desire to meet organisational or corporate targets (increased charges, convictions etc.) and to provide a victim-orientated service regardless of the outcome. In two forces (B and C) this related to an understanding of risk and the application of THRIVE to assess cases. Ultimately this meant that stranger cases were predominantly prioritised and new cases coming in prioritised over routinely updating victims involved in ongoing investigations over time.

It’s impossible, it really is […] I think juggling wise, I always drop something. I cannot do this impossible workload 100%. I would have to drop something to manage something else

(Force A)

Far too many jobs to investigate meaning that we are giving the public a terrible service. I feel embarrassed when I have to explain to victims that due to a high volume of work we cannot offer the service they deserve. The department is on its knees and gradually getting worse

(Force C)

Strategic understanding of demand and use of local data and information limited

The organisations were not effectively managing demand or understanding what demand involves through the use of locally held data. The research team heard stories of deflecting and managing demand as methods of coping. This was compounded by the lack of specialist knowledge and complexity of some of the cases coming in. The application of rape myths to assess ‘good’/’bad’ victims in terms of those that may more successfully progress to charge and prioritisation of new cases over ongoing cases are some examples of the way officers manage the demand through deflection.

Indeed, the survey found that in one force where demand was particularly high, 43% of respondents thought that some victims were more deserving of their time than others. With these attitudes there is a real risk that myths will be applied to prioritise caseloads. This is implicit in this quote:

They are all passionate and committed. What, I genuinely feel is that they are broken. They have been so overwhelmed in work, work lives, that they are not able to do their best. And actually, as a result, that, is where some of the corner cutting comes from and where some of the sort of victim credibility [assessment], for example, comes from

(Force B)

Demand analysis would also facilitate the identification of learning needs and focused development options based on the local typologies of cases being reported locally. Strategically equipping officers with the knowledge, they need to support vulnerable victims and investigate these complexities in different ways would also help manage the demand more effectively. This, linked with the transactional nature of resource deployment, means that officers are unable to cope with either the type of cases coming in or the number. Indeed, the survey found that less than 25% of respondents felt able to effectively manage their workload and nearly 80% feeling they do not have enough time to get their work done.

…I just want to say that when Rape Investigation Team was introduced, we were promised something that hasn’t been delivered. Because obviously they said they wanted to professionalise it… They would be working alongside us as well. And that just hasn’t happened. So, we thought we’d have that support around us. So, I think that we do initiate a good service to victims [unclear]. And our workload is going up and up…

(Force D)

Time for effective supervision

Demand had additional consequences for supervision of both cases and staff. Case reviews, critical discussion about decision making, welfare needs being identified and processes for organisational learning were compromised by the firefighting nature of demand management in forces. Furthermore, addressing the accumulative nature of dealing with these offences over time is not picked up through supervisory sessions with staff. Force D reported more highly on perceptions of being to access their supervisors in the survey than other forces, but this is at risk. As this supervisor told us:

As a supervisor on the rape investigation team, the current demand is unmanageable for me as the supervisor of over 100 live cases. The large workload impacts on my effectiveness as a supervisor across all aspects. Having worked as a supervisor in a range of other roles and departments, this is the most stressful and busiest role / department which I have worked in. I believe that the only way to reduce this demand is bring in additional supervisors to the department to share the workload and enable more effective closer scrutiny and action plans for ongoing cases

The lack of experience in teams caused additional issues for the supervisors as they pick up workload that new investigators cannot deal with.

I think it’s worth also adding that we’re talking about the amount of work the DIs and DCIs place into the actual physical investigations and backfilling the lack of experience on the teams. But it’s also that management expert aspect of looking after our teams. Our teams, because they are young in-service, they don’t necessarily have that coaching and level of ability, and all the other elements

(Force C)

In summary the implications of these issues are severe for victims, justice outcomes, officer welfare and professional development. This ‘firefighting’ approach to demand management is creating potentially unethical decision making about certain cases which is not being recognised because of the lack of case reviews occurring in the forces. The perceived legitimacy of the change to RASSO internally is undermined due to the transactional nature of the process, the lack of understanding of the impact it is having on staff welfare and the genuineness of the police to change and improve for victims.

Individual resilience to ‘cope’ with the lack of resilience for officers provided by the organisation

The final theme relates to the individual resilience provided by officers to support the investigation of rape through these adversities. ‘Toxic’ coping strategies are significantly impacting on officers’ welfare and home life in a climate where supervision is lacking, time to access support is limited and officers feel guilty about taking time off and leaving increasing demands to be dealt with by their colleagues. The organisational responsibility for the impact on officers’ wellbeing was highlighted throughout this research.

…Having to work on our days off as well. I’ve done that a number of times just to catch up with bails where I’m doing excess hours from home just so that when I come in, I’ve got a bit of breathing space, so it impacts your family life and home life massively as well…

(Force D)

The findings from the survey point toward high levels of perceived ill-health and stress among RASSO officers. The stress level is higher amongst RASSO officers compared to NHS staff (64.5% compared to 46.8% in the NHS based upon the 2021 NHS staff survey). Sixty per cent of RASSO officers admit to coming to work in the past three months despite being ill (compared to 54.5% of NHS staff)^{[footnote 180]}. Burnout symptoms for emotional exhaustion were shown to be high relative to studies undertaken in the NHS.

Analysis of survey data in conjunction with qualitative evidence highlights the interplay between work demands and increased depersonalisation, manifested as compassion fatigue towards victims^{[footnote 181]}. In addition, there is supporting evidence of officers continuing to work despite ill-health due to strong task cohesion^{[footnote 182]}. The analysis also found strong effect sizes for work demands, the effect on work-life balance, pressure on an individual to come to work despite being unwell from managers and colleagues, and variability in team cohesion (job demands). In contrast, a positive learning climate and satisfaction with organisational health policies offset burnout (job resources).

In all force areas officers are exhibiting at least equivalent or higher levels of emotional exhaustion than medical staff surveyed in two studies that focused on measures of burnout using the same measurement tool (Maslach’s Burnout Inventory)^{[footnote 183]} during Covid and feel less of a sense of personal achievement in their work. Furthermore, in one force particularly (B) the acceptance and normalisation of welfare issues simply being part of being in this role was particularly concerning. As well as demand factors there is a clear link between officers’ perceptions of their organisations learning climate and commitment to learning and officer well-being and a relationship between having access to organisational support and well-being. Under 35% of officers currently feel that their force takes positive action of health and well-being.

I feel stressed all the time. I feel, bad tempered, exhausted and frustrated

(Force D)

Insufficient training with no open and transparent route to obtaining it. No time for CPD or complete required portfolios for CoP courses. Organisation refuses to respond effectively to capacity and capability issues i.e. staffing and training. No adequate proactive response to welfare i.e. reduce workload which is the catalyst. Organisation does not respond to root level issues which creates stress and frustration i.e. inadequate training, IT, laptops, vehicles, resources, staffing

(Force C)

Burnout is high, but you keep going for your colleagues. There is no help for the high workload

(Force C)

The organisation is doing very little to empower officers in their role both in relation to their welfare and their skills. Research^{[footnote 184]} suggests that for a police force to be operating in a procedurally just manner, it must first create organisationally just internal operations. It is difficult to expect the police to treat individuals with procedural justice if they themselves are not treated with dignity and respect by their leaders and organisation. This research highlights the link between burnout, learning and occupational welfare support. Police officers are the conduit between organisational justice and procedural justice – their perceptions of procedural fairness and respect in the workplace may have a subsequent impact on the delivery of procedurally fair, transparent, and inclusive justice for victims. This is key in RAOSO investigations.

In summary the level of burnout amongst this cohort of officers is overwhelming. A culmination of unmanageable demand, a lack of specialist knowledge and ability to access learning plus limited time to access support is significantly affecting officer wellbeing. The risks for the organisation, victims, and its most valuable assets – the officers – are clear. These are organisational responsibilities and as explored in the theoretical framework for Pillar Four the risk to internal and external organisational legitimacy is significant.

Acknowledgements

We would like to say a big thankyou to all the officers and staff who gave us their thoughts on this important subject for the research. Their insights and contributions were invaluable.

Appendix 11: Pillar Five – data and performance

Pillar lead: Jo Lovett Pillar team: Gavin Hales, Liz Kelly, Fiona Vera-Gray, Gordana Uzelac, David Buil-Gil and Andy Myhill.

Introduction

All police forces hold, and are continually gathering, significant amounts of data on the cases that are reported to and investigated by them. Pillar Five set out to explore what these data could tell us about the nature of rape and other sexual offence (RAOSO) cases coming into the four forces, both at the individual force level and collectively. We also sought to assess how and whether forces currently use data to inform their response to these crimes, including the availability and quality of data that any such findings could be based on. The objectives were therefore to:

• provide in-depth analysis of data on RAOSO cases at force-level and across forces to identify patterns and trends relating to RAOSO reporting, case progression and outcomes;

• improve understanding of why a large proportion of reported RAOSO cases are not progressing, specifically those finalised as outcomes 14, 15 and 16;

• assess how forces record, analyse and utilise police-held administrative data;

• provide advice and guidance on how to do this more effectively.

The independent analysis of data both within and across all five pathfinder forces was intended to provide new evidence on the nature of reported RAOSO cases and the patterns associated with case progression and outcomes, feeding into local and national policy developments and responses. The rationale was that better use of police data would assist individual forces to develop a clearer strategic understanding of the nature of rape and sexual assault locally and, through this, greater awareness of what investigative strategies and approaches might be required to achieve better outcomes. Additionally, ensuring that performance monitoring more accurately captures trends in local police responses to rape and sexual assault should assist in identifying key areas of challenge requiring improvement.

Background

Research over the past four decades has shown that the majority of rape cases reported to police do not progress beyond the police investigation^{[footnote 185]}. Whilst reporting figures have followed a relatively constant upward trajectory for many years, the increases have intensified in the past five to ten years. At the same time, charges and prosecutions have ‘plummeted’^{[footnote 186]}, leading to searching questions about why rape investigations are failing, while some have asked whether rape has been effectively ‘decriminalised’^{[footnote 187]}.

The Home Office crime outcomes framework was revised in 2015 to move away from a simplistic focus on detections towards wider outcomes, and to provide greater transparency about how police deal with the crimes reported to them^{[footnote 188]}. Police must report on these outcomes to the Home Office through the Home Office Datahub. The recent UK Government End-to-End Rape Review on Findings and Actions ^{[footnote 189]} and HMICFRS and HMCPSI joint thematic inspection^{[footnote 190]} highlighted an increase in victims withdrawing from the process (designated outcomes 14 and 16 within the framework), and there is increasingly a public narrative that asks why so many victims are disengaging from the police process^{[footnote 191]}.

These trends need to be seen against a wider backdrop of changes to crime recording practices. Home Office Counting Rules (HOCR), the national framework for interpreting, classifying and counting crime^{[footnote 192]} and the National Crime Recording Standard (NCRS)^{[footnote 193]} were introduced in the early 2000s to address inconsistent recording practices. The HOCR and NCRS are aligned to ensure greater validity and reliability of police recorded crime data and resulted in an immediate increase in the number of crimes recorded^{[footnote 194]}. Following ongoing recording issues noted particularly in respect of sexual offences^{[footnote 195]}, there have been further changes and new guidance including: the introduction of one crime per victim/suspect in rape crime records in the HOCR; clarifications in the NCRS as to which third party reports should be seen as in victims’ interests and therefore be recorded; and an overall shift in the spirit of how and when an incident should be crimed. Adherence to the NCRS and HOCR has become increasingly forensic following the focus on crime recording under the HMICFRS ‘Crime Data Integrity’ (CDI) inspection programme. There has been a specific focus on rape and other sexual offences within the CDI inspections and reports, as historically levels of under-recording and ‘no-criming’ of these offences have been highlighted as a matter of particular concern.^{[footnote 196]} No criming is where additional information comes to light which determines that the crime has not occurred. This was found to be an area of particular concern during early HMICFRS CDI inspections but subsequent inspections have seen marked improvements.

General issues with police data have also been noted: for example, missing data in police records of rape offences has been flagged up in a number of reviews and inspections^{[footnote 197]}, as have some of the issues faced by police analysts in terms of IT infrastructure and systems, lack of technical support, and training and development needs^{[footnote 198]}. The importance of collecting better data on the characteristics of users of the CJS in order to monitor access to justice and equalities has also been underlined^{[footnote 199]}.

Methodology

We began Year 1 focusing on RASSO (Rape and Serious Sexual Offence cases) as per the programme brief. During the course of the deep dives, it became apparent that there was a lack of consensus about the definition and use of RASSO among police forces and other agencies, although most broadly interpreted this to mean rape and other penetrative or contact sexual offences. We have primarily focused on rape cases in the analyses below, but where relevant we have explored the additional sexual offences in our sample. As a programme, we have adopted the term RAOSO (Rape and Other Sexual Offences) to avoid implying that some offences are more or less serious than others.

We employed a combination of quantitative and qualitative methods^{[footnote 200]} in the deep dives. Using a mixed methods approach enabled triangulation between data sources to test consistency and provided a more comprehensive understanding. The quantitative and qualitative strands of work were mutually informing, and provided breadth and depth of findings. For example, interviews with police staff and the case file analysis provided context and explanation for some of the higher-level statistical findings and also informed our understanding of gaps in the quantitative data. Similarly, the quantitative work provided an overview of over-arching trends within which to situate the qualitative findings.

Analysis of depersonalised data on all RASSO allegations reported to each force in 2018-2020/21 (81,705 cases comprising 47,213 rapes and 34,492 SSOs)

These datasets were extracted and collated by each force and shared with the research team in Excel form. They cover victim, suspect and offence characteristics, as well as information on case outcomes. The datasets consist of all RASSO cases reported to police in the calendar years 2018 to 2021. Two forces fell slightly short of that, as one was only able to supply data to the end of 2020 and another to the end of November 2021, as the timing of the deep dives fell during 2021. The data were subjected to exploratory quantitative analysis. We refer to this as the ‘2018-21 dataset’.

Case file analysis of rapes closed as outcomes 14, 15 and 16 (741 cases)

In Force A, we examined 100 case file records of each outcome type from the period immediately preceding each deep dive, but due to the scale of the task we reduced this to 50 in Forces B, C and D. To ensure consistency in the sampling across forces, we focused only on rape cases here. A small number of cases were excluded as they were duplicates or files were missing. Case files were shared with the research team in two different ways depending on the data sharing agreements negotiated with each force. Forces A and D supplied redacted, anonymised extracts from the case including the investigation log and police actions file in PDF form, while Forces B and C controlled remote access was granted to police systems via secure police laptops so that records could be consulted directly. From the case file records, the research team extracted information recorded about the offence, victim and suspect, reporting, outcomes and timescales. This included capturing detailed reasons why cases were not proceeding so as to shed light on the range of factors underpinning this with the aim of developing possible sub-classification codes. In addition, qualitative research memos were completed to summarise each case, provide broader narrative context, plot an investigation timeline and describe key decision-making relating to outcomes. Qualitative data on evidential issues and rationales for case outcomes were subjected to thematic content analysis in order to generate descriptive codes about why cases were closed. We refer to this as the ‘case file dataset’. The case file analysis presented here is based on 591 cases from Forces A, B and C^{[footnote 201]}.

Interviews with police staff (37 interviewees)

Qualitative semi-structured interviews were conducted with strategic and operational police leads, analysts, crime management staff and force crime registrars in each force. These were identified by our pillar leads within each force as being most relevant to our area of enquiry. Interviews focused on participants’ experiences of how data is used, how performance is monitored, crime recording and issues relating to case progression within their force. The interview transcripts were subjected to thematic analysis^{[footnote 202]}.

Survey with analysts in one force (24 responses)

In Force A, which had a large number of analysts, we distributed an online survey to all intelligence, central strategic, and central and area-level performance analysts.

Document review

Relevant documents in each force were reviewed, including problem profiles and other analyses, performance frameworks and reports, and policies related to crime recording, investigation and outcomes.

The table below provides a breakdown by force of the full data corpus achieved.

Table 1: Pillar Five data corpus

Summary of key findings

The following sections document our key findings in relation to the four pathfinder police forces: quality and use of police data; the profile of RAOSO cases; crime recording and outcomes; and victim engagement. Key findings include:

• Police data are incomplete and inconsistent in a number of areas.

• Among recorded rapes, between one quarter and one third involved intimate partners, but there was a relatively high proportion of missing data on suspect-victim relationships in each force.

• Consistently across the four forces, around one-third of police recorded rapes were flagged as domestic abuse-related.

• In the case of outcome 14, half of all reports were made by a third party only or involved a disclosure that was not intended to constitute a formal report (‘telling not reporting’).

• The combination of over-recording and not ‘no criming’ results in the denominator for charge rate calculations being higher, and resulting apparent charge rates being lower than would otherwise be the case.

• The increasing volume of police recorded rapes has coincided with austerity; all else equal, the result will have been fewer resources available per recorded rape investigation.

• In all forces outcomes codes are frequently misapplied, and in some forces they are particularly poorly supervised.

• In the 2018-21 dataset, outcomes, and especially charge rates, vary systematically by suspect/victim relationship type.

Quality and use of police data

The four pathfinder forces generated data on a huge number of cases over the four-year period examined here – over 80,000. In general, these datasets are a very rich resource, particularly where we have been able to drill down into individual case files to gain a more complete picture of the case and development of the investigation.

Nevertheless, across all four forces we found that in a number of aspects police data are incomplete and inconsistent. This makes strategic analysis difficult and the validity and reliability of some findings uncertain. It should be recognised that, often, the primary details relating to offences are recorded at the point of an initial report being made, which can be a pressured and difficult time for both victims (or third parties reporting) and police staff alike, and it may not be possible to obtain all of the information that is of interest to analysts for a variety of reasons, including that victims/third parties may not know or wish to volunteer it. However, if these details come to light further into the investigation they can still be captured.

Key gaps typically concern data on victim and suspect ethnicity, both self- and police-defined, and suspect-victim relationship (see Table 2).

Table 2: Missing data across forces in the 2018-21 dataset (rape cases)

Notes:

Force A: Data for 2018-20. Victim with a URN; numbered suspects; suspect-victim relationship combination of 2 variables.

Force B: Data for Nov 2018-21.

Force C: Data for 2018-21. Suspect-victim relationship for 2018-20 only; combination of 2 variables (56% missing on main variable).

Force D: Data for 2018-21. Susp ethnicity excludes where there was no identified suspect

Exploration of the case files showed that while information may be missing from the structured fields where it should be recorded, some of this is known to police, as it is recorded elsewhere in case records, such as in the investigation log. Details of victim and suspect sex were known in virtually all cases in the case file dataset, and there were notably fewer gaps in relation to ethnicity and especially suspect-victim relationship. However, in Force C, data gaps remained significant (see Table 3).

Table 3: Missing data across forces in the case file dataset (rape cases)

Notes:

Victim sex, age and ethnicity data were often redacted on the PDFs in Force A, so the true level of data available is unclear.

Other protected characteristics, such as disability, are rarely recorded systematically, if at all, and two forces have now created a combined sex/gender identity category^{[footnote 203]}, meaning that sexual offences against trans/non-binary people cannot be disaggregated.^{[footnote 204]} Suspect data are frequently incomplete, and this is not entirely limited to cases where suspects are unknown or unidentified. Dates, including dates of birth and particularly offence dates, are also subject to error. For example, in some cases where a precise offence date is not known, the report date is entered instead, which limits the possibility to analyse time taken to report and investigation lengths. Another gap across all forces is police investigative actions (for example, conducting victim and suspect interviews), which tend to be recorded in narrative logs rather than in structured fields, making systematic analysis and monitoring difficult.

The knock-on effect on analysis and the misinterpretation of data because those fields are not being entered properly, is huge, it totally changes our profiles (analyst Force D).

Factors affecting poor data completion identified by police staff interviewed included stretched resources (for example, not enough call handlers for the volume of calls) and heavy workloads for investigators, and we identified a lack of guidance for officers on how to enter data in case records and how to interpret variables or assign categories. In one force, we were told that a change of system in which certain fields were no longer mandatory had led to a fall in completion rates. There was also an absence of a corporate narrative about the importance of good data, not just as an administrative task, but as something of benefit to officers and the organisation as a whole – as a clear record of the investigation, as intelligence, and as a source of strategic understanding, organisational learning and development. Additionally, data completion was not something that appeared to be routinely checked at supervisory reviews or at the point of file closure in any force.

Analysts frequently reported that police IT systems were not designed to support strategic analysis.

It’s a terrible system for analysis. It wasn’t set up for analysts

(analyst, Force B).

[It] doesn’t code the data the way that we need it for our report

(analyst, Force D).

It was probably fit for purpose when it was built, but there’s a lot of things that I would like to have included in there

(analyst, Force A).

One issue is that a lot of valuable information about both the offence and the investigation is held in narrative logs or attachments rather than in structured, searchable fields, which are much harder to analyse systematically.

There is a lot of data hidden underneath [the structured fields] that I can’t get. It’s not part of my role to read through 20, 30 pages of [investigation log]

(analyst, Force A).

In two of the four forces, analysts are mitigating these data issues by creating parallel datasets that they manually populate from case files, which they then use for analysis. However, this is labour-intensive, often making it hard to look beyond small samples of cases; it entails a duplication of effort and diverts time away from analysis itself. The National Intelligence Model states precisely the opposite, that: “Analysts must be given dedicated time to undertake quality analysis and should not be side-tracked into performing basic information management tasks”^{[footnote 205]}. Although these processes reveal data errors in the original record, in almost all cases they are not being corrected in the main recording systems.

Despite recommendations by the joint police and Crown Prosecution Service (CPS) inspectorates^{[footnote 206]} that forces should have relevant and up to date problem profiles, none of the forces had one on rape or RAOSO. One force had a thematic profile focusing only on stranger rapes, which are a minority of all cases. It should be noted that two forces had begun developing a new profile during the deep dive.

In each force, there were multiple data systems in use and poor links between them. There was particularly a lack of access to, or possible linkage with, CPS data. A partnership project between Avon and Somerset Constabulary and CPS South West is developing a prototype tool to link data across criminal agencies to address some of these issues locally.

Analytical support seems to vary greatly between police forces, both in terms of resourcing and levels of experience and specialisation.

Profile of reported RAOSO cases

Some findings from the analysis of the 2018-21 dataset have been presented in the main report. Here, we report on several key characteristics of the cases coming into these four forces during this period. Mindful of the caveats regarding data quality and gaps outlined above, when reporting these findings, where possible we look at comparisons between forces to identify whether there are areas of consistency and other data sources, such as the case files and interviews, to cross-check and contextualise.

Offence types

With the exception of Force A, where the majority of offences were rape, there was a roughly a 50/50 split between rape and other sexual offences.

Table 4: Breakdown of RASSO offences by rape / SSO

Note: (1) An apparent error with the data provided by Force A meant some other serious sex offences were missed^{[footnote 207]}, understating their proportion of overall RASSO offences. In national data, the proportion is closer to 50/50.

Suspect-victim relationship profile

Forces A, B and C were relatively consistent in terms of the suspect-victim relationship. In all three, the largest group involved intimate partners, comprising between one quarter and one third of cases, while between 10% and 17% were strangers. Force D was affected by high levels of missing data in respect of this variable.

Table 5: Suspect-victim relationship profile of police recorded rapes

Notes: (1) 2018-20; (2) 2018-21.

Consistently across the four forces, around one-third of police recorded rapes were flagged as domestic abuse (DA)-related, as were around one in ten other serious sex offences. In most forces, there was consistency between the DA variable and the suspect-victim relationship variable(s) (noting that not all DA cases will have involved intimate partners).

Table 6: Proportion of cases with domestic abuse flag compared with intimate relationships

Notes: (1) 2018-20; (2) 2018-21.

Timing of reporting / recording

Figure 1 is from Force A and shows both the particular delay between offences and reports in familial rape cases, but also the way across all relationship types the data have a heavily skewed distribution, with a long tail (offences reported later). Using more detailed relationship categories in Force B (not shown), rapes committed by ‘persons of trust’ were reported even later, with the median being around 38 years (at least some of which were linked to large scale allegations of historical institutional abuse that were being investigated by the force).

Figure 1: Rape allegations: time from offence to report (days)

A consistent pattern is seen across forces whereby the age of the victim at the time of the offence is inversely related to the time taken to report to police.

With the exception of Force D, where force analysts and our case file review highlighted concerns about the recording of historic offence dates, consistently around 7 out of 10 rapes with victims aged 12 and under were reported more than 5 years after they occurred (the figures for the Forces A to C were 68%, 69% and 72%). By contrast, older victims were much more likely to report rapes closer to the time of the offence, as this chart for Force A illustrates (see Figure 2).

Figure 2: Victim age at time of offence and time between offence and report to police

Data for this chart was not available to publish accessibly. Figure 2 is a chart showing victim age at time of offence and time between offence and report to police. It demonstrates that older victims were much more likely to report rapes closer to the time of offence.

Crime recording and outcomes

There have been marked increases in police recorded rape offences, nationally and in all of the deep dive areas over the past decade, although the scale of increases and resulting rates (police recorded rapes per 100,000 residential population) have varied widely around the respective national averages (see Figure 3, in which the pathfinder forces are highlighted, alongside those with the highest and lowest rates of police recorded rapes). However, the Crime Survey England and Wales estimates that there has been relative stability in the prevalence of rape over the last 15 years^{[footnote 208]}.

Figure 3: Rapes recorded by police forces in England and Wales 2012/13-2021/22

Data for this chart was not available to publish accessibly. Figure 3 is a graph showing rapes recorded by police forces in England and Wales. It shows that there have been marked increases in police recorded rape offences, nationally and in all of the deep dive areas over the past decade, although the scale of increases and resulting rates (police recorded rapes per 100,000 residential population) have varied widely around the respective national averages (in the graph, the pathfinder forces are highlighted, alongside those with the highest and lowest rates of police recorded rapes).

Our analysis supports the view that changing crime recording practices have increased the number of rapes recorded by police forces (more in some than others) and have almost certainly seen the profile of recorded rapes change over time. As noted above, key formal changes include one crime per victim/suspect combination and the recording of all third-party reports as rape, many made without the knowledge or consent of the victim. We discuss the detail and consequences of these changes as seen in the in-depth case file review below.

Attribution of outcomes

In all forces outcomes codes are frequently misapplied, and in some forces they are particularly poorly supervised. In at least one quarter of the case file sample, details documented in the file suggested that the wrong outcome had been assigned. We have found a number of examples of:

• Cases where there is a named suspect being finalised as outcome 14

• Cases where there is no identified suspect being finalised as outcome 16

• Cases where the suspect has died being finalised as outcome 14, 15 or 16 rather than outcome 5 or 12.

Outcome finalisation processes differ across forces, with varying layers of oversight and parties responsible for this, which has implications for how carefully the outcome is checked, but also for how long the process takes. This in turn has a bearing on analysis of investigation timescales, depending on when the investigation end is calculated as being – the point of submission for closure or the point of outcome finalisation. In the case file analysis, this varied widely between forces, although there were marked differences within forces by outcome type.

In the 2018-21 dataset, outcomes, and especially charge rates, vary systematically by suspect/victim relationship type. For example, charge rates for ‘stranger 1’ and familial offences are typically much higher than where allegations occur between current/former intimate partners or friends and acquaintances. This may reflect the complex reasons why victims may not want or intend police action, as it is cases with current/former partners that are particularly being picked up through the ‘telling not reporting’ category – for example, through the DASH or disclosed as part of another investigation/report. The differences also likely reflect the varying salience of consent. Related to this, we saw little understanding of how to build cases where there is a consent defence, particularly where there had been a prior relationship.

Any differences in the relationship profile of recorded rape cases between forces may therefore be expected to result in differences in overall charge rates. To mitigate this, ideally, charge rates would be calculated and published by suspect-victim relationship type to allow meaningful comparisons to be made.

Reasons why cases designated outcome 15 are not proceeding

Analysis of the sample of case files designated outcome 15 (where there is a named suspect but evidential difficulties prevent further police action) shows that evidential issues cover less than half of cases (see Figure 5). In around one fifth of cases, the assessment of the victim’s credibility appeared to be a key factor. Small proportions of cases were linked to victim withdrawals, third-party reports and possible no crimes (these are discussed in more detail during the discussion of outcomes 14 and 16 below).

Figure 5: Reasons why outcome 15 cases are not proceeding

Within the category of evidential issues, police most commonly stated that the evidence was not strong enough or raised concerns about whether the victim was, in fact, consenting. Evaluations of the strengths and weaknesses of a case often located ‘strong’ evidence in a way that suggests a stranger-based model guides the approach to case-building. Evidential issues were thus regularly about a lack of digital, forensic or CCTV evidence, and the suspect making a defence of consent was regularly listed as a weakness in relation to making a decision to charge. Detailed analysis of what evidence is understood as strong or weak in a case is something that will be explored further in our Year 2 work. Where victim credibility was a factor, this was usually linked to perceived inconsistencies in the victim’s account, complex mental health and/or substance issues, previous allegations, and the victim’s willingness to handover their phone and other digital equipment or consent to the police contacting third parties such as their GP, for supporting material.

The analysis of outcome 15 cases, particularly, highlighted that the law on consent appears to be poorly understood by some rape investigators. There is evidence in the case file data of consent being presumed unless non-consent is made explicit by the victim and a lack of applied understanding of the ‘freedom’ qualification in the statutory definition of consent.

Reasons why cases designated outcome 14 and 16 are not proceeding

There is a public narrative of the CJS failing victim-survivors who want action – and there are instances of this among the outcome 15 cases – but the scale is being amplified through current recording practices. Our exploratory analysis^{[footnote 209]} of cases finalised as outcome 14 and 16, which are commonly described in terms of ‘victim withdrawal’^{[footnote 210]}, showed that the actual ‘withdrawal’ of the victim from the investigation accounted for only 16% of outcome 14 cases and 34% of outcome 16 cases (see Figure 4). We included within this, both cases where there was a clear withdrawal via a retraction statement, and those where the victim ceased cooperating with the investigation, often not responding to police attempts to contact them. Even with the inclusion of the latter, less formal ‘withdrawal’ these data show that the majority of cases finalised as outcome 14 and 16 do not result from the victim removing support from a previously supported investigation. For around a quarter of these, fear/ambivalence about the CJS process was a key issue. Other factors included needing to prioritise other matters in their life, such as health or education.

Figure 4: Reasons why outcome 14 and 16 cases are not proceeding

It is important to understand the full range of rapes reported to and/or recorded by the police, especially those not proceeding, and how (particularly for stakeholders) this may differ from the profile of victims and offences seen by ISVA services. Insights from the case files call into question the proportion of cases closed at outcome 14 and 16 that are not proceeding due to a lack of victim engagement.

Analysis suggests that Home Office crime outcome codes responsibilise victims and offer few clear insights into the reasons for attrition. Although subtle changes have been made over time to the wording used in the outcome framework^{[footnote 211]}, outcomes 14 and 16 are nevertheless commonly referred to as ‘victim-based’, despite coming under the broader heading of ‘evidential difficulties’, and legacies of previous wording remain built into police systems. This language is reiterated by officers throughout the decision-making process, leading to a script of rape victims as the reason why cases fail within policing (through terms like ‘unwilling’, ‘uncooperative’, ‘refusing’ and ‘withdrawal’).

The language of ‘victim engagement’ places the responsibility on victims for ensuring that cases can go forward, whereas there were some examples where police ‘lost’ the victim through failure to make timely contact. The onus for engagement was placed firmly on victims rather than on the police, whereas the police must also take some responsibility for finding more creative ways to engage and support victims, including when they do not want to proceed, recognising that mental health issues, trauma and their circumstances may affect their (non)response (see Pillar Three). This also points to national issues about limitations of the outcome codes to accurately reflect what is going on. Language used by officers in the case files here frequently positioned victims negatively, using terms like ‘obstructive’, ‘challenging’, ‘refusing’, ‘unwilling’ and ‘uncooperative’ (see also Pillar Three). There was even one case where the officer responsible for liaising explicitly disengaged themselves from the victim, saying:

This report can be closed, until the victim makes contact and agrees to speak to police in the right manner.

In-depth case file review shows that even where offences had been recorded without the knowledge or support of a victim-survivor, victim-survivors expressing a negative reaction to a formal investigation was labelled as them not supporting or cooperating.

Within the sample of outcome 14 and 16 cases that we analysed there are three categories of cases that account for a substantial proportion of these records made without the knowledge or support of a victim-survivor. The first is reports made to the police about the victim but not by the victim. These third party cases were not normally supported by the victim, and reports were often made without their knowledge, which could cause distress and a sense of betrayal in relation to those who had made the report. We have also found some third-party reports being recorded, and finalised under ‘victim-based’ outcomes, where the person who the third party is claiming to be the victim states that that no offence has occurred (see third example in Box 1 below).

The second is constituted by what we have defined as ‘telling not reporting’. Telling not reporting cases led to the recording of cases, often non-recent, that victims had chosen not to report at the time and did not wish to report now. This includes examples where police are recording rape offences disclosed to them during the DASH ^{[footnote 212]} risk assessment process, as victims here are ‘disclosing/answering’ a question on sexual violence not actively ‘reporting’ it. (This was also the case for 17% of recorded rapes examined in the 2019 London Rape Review^{[footnote 213]}). Other instances, as shown in Box 2, are where a crime victim is offering an explanation – such as that they do not want to talk to a male police officer because of previous sexual violence experiences – and a disclosure made in the course of a different, separate investigation. Their recording meant that some level of investigation ensued, which victims did not welcome and were often distressed by, and this sometimes affected their trust and confidence in the police.

In the case of outcome 14, half of all reports were made by a third party only or involved a disclosure that was not intended to constitute a formal report (‘telling not reporting’). Similarly, well over one third of outcome 16 cases were made up of third-party reports (16%) and where the victim was ‘telling not reporting’ (23%) (see Figure 4). The examination of these cases clearly showed that not every rape recorded by the police results from a victim contacting the police to report that they have been raped (let alone wanting it to be investigated). There was also a small group of cases where victims were unable to participate in an investigation or give a full account, often due to chronic mental health issues or crises.

Third party cases

1. A psychiatrist reports that during a consultation, a woman said that she was raped. The psychiatrist provides the woman’s name and says he does not have her permission to report this to the police but she does agree to getting medical assistance. The police talk to a doctor at the SARC who says this breaches client-patient confidentiality, but that he would contact the victim about coming to the SARC. When the woman is contacted by the SARC she is surprised at contact and angered that there had been police contact as she did not wish to report the incident or have any investigation conducted. She raised concerns in regard to her doctor contacting the police. She did not want any police contact. This is recorded as a report of rape and NFA’d under outcome 14.

2. A woman is contacted by a domestic abuse service following a referral from the MARAC process, as she has been designated high-risk, has mental health issues, is homeless and has a history of domestic abuse. She discloses to the service being raped by her ex-partner, and they inform police. When police contact her to follow up, she says she did not wish for police to be informed and is shocked that they are calling her. She is not ready to talk to police. She is described as an ‘unwilling’ victim.

3. A woman is coming home from a night out in a taxi. She is drunk and has an argument with driver who forces her to get out of the car. She then goes to a garage and sits down, crying. A worker at the garage called the police, tells them there is a woman crying outside and suggest she may have been raped. When the police come and speak to her, she says she was not raped and explains what happened. She wants to make a complaint against the taxi driver. This is recorded as a rape and finalised at outcome 14.

Telling not reporting cases

1. A woman is in a police car after there had been a dropped 999 call where shouting could be heard and a woman telling a man to get out of her address. While in the car she says that she is nervous of men and wanted to talk to the female police officer because she had previously been raped and had reported this to the police. This is recorded as a report of rape even though the officer states, in requesting finalisation, that “this offence was mentioned in passing”.

2. A woman reports domestic violence to police. As part of the last assault the abuser said in front of her children that she had been raped. She told police that he said this and that it upset her because she did not want her children to know. She says she does not want to report the rape, it happened 20-30 years ago, and she had already reported it to another force who had investigated it. [This force] cannot find the report [made to the other force] and so record this as a report of rape and then NFA it under Outcome 14.

In a small proportion of outcome 14 and 16 cases, information in the case file indicated there was a possible ‘no crime’ (which is to say that evidence suggested that an assault had, in fact, not taken place). This was more prevalent among cases designated outcome 14 (see Box 3).

Examples of possible no crimes

1. A man with mental health issues, PTSD and learning difficulties living in supported housing makes a third-party report to police saying he can hear a friend being raped. When police attend, he says he heard the rape in his head, then says it was not in his head and takes them to the door where he heard it. The two residents there say they are okay and nothing has happened. The caller has made similar calls previously and staff say his condition has worsened. There is no identified victim or suspect, but case finalised at 16.

2. A duplicate crime report that has already been crimed is closed as outcome 16 rather than being cancelled.

3. A prank call from a child, with no clear allegation, victim or suspect, is recorded and finalised as outcome 16.

The requests to no-crime that we have seen in the case file analysis are invariably refused through reference to HO counting rules. Scrutiny of crime recording by HMICFRS, especially of rape in relation to ‘no crime’ decisions came about in the wake of numerous inspections and reviews that found police were under-recording sexual crime and ‘no criming’ cases that should have been crimed. The approach of HMICFRS has made force crime registrars highly risk averse and consequently reluctant to authorise ‘no crime’ decisions. Qualitative interviews with police officers, crime management teams and Force Crime Registrars indicated that the bar is set too high for ‘no criming’. Investigators report that they often do not bother requesting no crime decisions as the process is so onerous and unlikely to be authorised.

Going through the no crime process is horrendous […] you have to provide additional evidence to say it didn’t happen, then it goes to a panel. You can waste six months to a year for it to come back and then in very few cases, well you would never get one back

(Officer, Force A).

The difficulty of ‘no criming’ means that officers are assigning different, often victim-based, outcomes to these cases, further accentuating the view that is the reluctance of victims that means cases do not proceed.

[I]f it’s getting classed as rape, we have to close it under outcome code 14, 15, 16 or 18

(Police officer, Force A).

The bar for ‘no criming’ is now that beyond reasonable doubt a crime did not happen, that is to say evidence is required that proves a crime did not happen – the same as for a conviction. While the bar is set high to prevent vulnerable victims from being pressurised or disregarded when arriving at a decision that a crime of rape did not occur, in practice it is creating a paradox. Current recording practices mean that though there is little evidence needed for a rape to be recorded, as it is recorded before being investigated, Additional Verifiable Information (AVI) is needed to prove it did not happen. Connectedly, once recorded, victims are subjected to similar levels of disbelief and scrutiny when they say something has happened as when they say it has not. In both instances, the word of the victim is not taken as credible evidence of what has or has not happened.

In addition, in some forces, a small number of victims account for a disproportionate volume of recorded rapes, typically where their reporting behaviours are linked to underlying mental ill health. Several such individuals from at least two forces were evident in the case file sample. Most of these reported incidents are being ‘crimed’, although some forces are developing local policies to ‘no crime’^{[footnote 214]} subsequent allegations of a similar nature linked to specific individuals and circumstances in the future. This is intended to operate on a limited, case-by-case basis with oversight from the force crime registrar. Alongside this, there needs to be a clear safeguarding approach and a transparent system of monitoring this kind of repeat offending.

The unintended consequences of Home Office counting rules for RAOSO

In all of the above categories (third party cases, ‘telling not reporting’, and no-crime cases), crime records are being generated in order to comply with the HOCR because a crime has been disclosed. In-depth case file analysis and interviews found examples of investigating officers stating explicitly that offences were being recorded for “Home Office reporting rules only.” While crime recording has an important role in registering the extent of crime, acknowledging harms to the victim and highlighting the potential risks posed by suspects, our analysis suggests that such recording practices and the processes flowing from them are having significant unintended consequences across four interlinked dimensions: victim-survivors, data integrity, police workload, and wider policing culture.

Victim-survivors: The finalisation of these cases as outcome 14 or 16 attributes the reason for the case closure to victims for cases they never reported. Victim-survivors are finding themselves involved in a police investigation they did not wish to take place, and this can place them in a position of further possible harm or distress, as illustrated below.

I am not and was not wanting to record this as a crime and prosecute or have [suspect] arrested or interviewed. I am still in contact with him for the purposes of him seeing our child and if he was to be interviewed and/or charged I know this would cause a great deal of problems as I know he wouldn’t react very well. For the purposes of trying to keep the situation between us as calm as possible for as long as possible and prevent escalation I do not want him to be interviewed on this matter.

(Victim in outcome 14 case)

An Asian woman reports her husband, who is a police officer, for domestic abuse, especially his controlling behaviour while he is out of the country. She answers yes to the sexual violence question on the DASH and a rape is recorded. The woman is clear that she will not give a statement about the rape. The abusive man is arrested on re-entry to the country and interviewed about a rape allegation, the case is NFA’d at outcome 16. There is no record of any action being taken about her initial complaint. (Outcome 16 case)

Where the sexual violence has come to light through the DASH or a domestic abuse investigation, police pursuing a RAOSO charge also has the potential to de-prioritise what victim is actually reporting – domestic abuse – and may deter them from pursuing any police action in the future. Rather than increasing confidence it has the potential to decrease it. Finally, there is evidence in police files of these records being listed in future reports of rape as a weakness (i.e. counting against the decision to charge a future, often unrelated, report of rape).

Data integrity: For data integrity, current recording practices are making it impossible to establish the true number of victims who are withdrawing from a report they made to the police with the intention of having it investigated. The case files suggest that police forces are over-estimating reporting and the extent to which victims are withdrawing. Baselines are increased by recording any mention of rape and the inability to no crime. The automatic recording of cases disclosed through processes like the DASH is also complicating any findings in relation to perpetrator relationship and outcome, which appear to show particularly low charge rates compared to most other relationship types.

Police workloads: There are also important implications of these findings for police in terms of workloads and resources from the administrative load of cases that are never going to proceed, as these need to be recorded, processed and responded to. This inevitably will be compounding findings from other pillars in relation to police workload and wellbeing.

Policing culture: Finally, there are implications for policing culture and public confidence in the CJS. Criming cases that victims do not want reported and then attributing these to a ‘victim’-based’ outcome perpetuates the idea of rape victims as the reason cases fail (the terms ‘unwilling’, ‘reluctant’, ‘refusing’ recur in case files) within policing, at the same time as leading the public and policy makers to question what is wrong with policing and the CJS that makes victims want to withdraw. Finally, current recording practices are contributing to an overall negative picture of victim-survivors in wider policing culture as victim-survivors are positioned as obstructing the policing of a criminal offence by being ‘unsupportive’, ‘reluctant’, or ‘unwilling’ to support an investigation they never wanted or expected in the first place.

Reporting and charging rates

Taken together, the findings outlined above mean that rising reporting and falling charge rates need to be properly contextualised. With the changes in crime recording practices, it seems likely that a proportion of rapes that are recorded today would not have been 10 years ago – so much of the increase is not in reporting but recording practices. The combination of over-recording and not ‘no criming’ results in the denominator for charge rate calculations being increased, and resulting apparent charge rates being lower than would otherwise be the case. The now recorded rapes have characteristics that mean they may be less likely to proceed to a charge in that there is no victim-survivor seeking an investigation.

It should also be noted that the increasing volume of police recorded rapes has coincided with austerity; all else equal, the result will have been fewer resources available per recorded rape investigation. Elongated police and wider CJS timescales mean that some rapes recorded in a given year may not be finalised until several years later, with charges typically taking the longest of all Home Office outcomes. Charge rates for a given year therefore continue to rise over subsequent years. This makes looking at data over time series increasingly important.

The onus for engagement was placed firmly on victims rather than on the police, whereas the police must also take some responsibility for finding more creative ways to engage and support victims, including when they do not want to proceed, recognising that mental health issues, trauma and their circumstances may affect their (non)response (see Pillar Three). This also points to national issues about limitations of the outcome codes to accurately reflect what is going on.

Appendix 12: Pillar Six^{[footnote 215]} – Examining, understanding and improving the use of digital material in RAOSO investigations

Pillar Lead: Tiggey May, Pillar team: Dr Catherine Talbot, Dr Rachel Skinner with Samantha Atkinson, Arianna Barbin, Chantelle Butt, Emilly Holtham, Tamzin Jeffs, Amrana Latif, Asmaa Majid, Georgie Markham-Woods, Tiago Garrido Marques, Elena Reid, and Louise Trott

Introduction

Pillar Six was introduced into Operation Soteria Bluestone following digital deep dives conducted in four forces in England as part of a Home Office funded Science, Technology and Research grant aimed at examining the digital RAOSO investigative journey. The deep dives followed a similar methodological approach to Pillar One-Five of Operation Soteria Bluestone where emerging findings from Pillars One-Five were also uncovering the problematic nature of digital evidence in RAOSO investigations; mainly concerning the handling of victims’ mobile phones (Pillar Three), the lack of a digital investigative strategy (Pillars One & Two), the absence of digital training for many RAOSO officers (Pillar Four) and the lack of any quantifiable digital performance data (Pillar Five). The overall aim of our Year 1, Pillar Six research has been to examine and understand the importance, prominence, and problems digital evidence presents to a RAOSO investigation and investigators in four forces across England and Wales.

Background

Ten years ago, digital data was viewed as the periphery element of criminal investigations. Now, the volume of personal data and digital material held on mobile devices, PCs and tablets has amplified at such an unprecedented rate that the Digital Forensic Science Strategy in 2020^{[footnote 216]} stated that over 90% of recorded crime now has a digital element (NFSS^{[footnote 217]}, 2020; Wilson-Kovacs, 2021^{[footnote 218]}). Despite the prominence of digital data, there remains a limited understanding, outside of Digital Forensic Units (DFUs), of how best to carry out a digitally competent, forensically compliant investigation^{[footnote 219]} (Vincze, 2016^{[footnote 220]}). Unsurprisingly, the volume of social media and other digital evidence routinely found in digital investigations is such that the “quintessential “smoking gun” is increasingly being viewed as the quintessential “needle in a haystack” (Brown, 2015^{[footnote 221]}).

The digital challenges facing the police today are a web of interconnected issues relating not only to equipping the police with the necessary technology to access and extract relevant data (Casey, Katz and Lewthwaite, 2013^{[footnote 222]}; Muir and Walcott, 2021^{[footnote 223]}) but also to providing police officers and staff with the necessary skills and training to be able to tackle the investigative challenges digital data poses (Home Office, 2015^{[footnote 224]}; Muir and Walcot, 2021), whilst remaining mindful of the associated ethical and privacy rights of victims, suspects and witnesses (Aminnezhad, A., and Dehghantanha, A. 2014^{[footnote 225]}, Muir and Walcott, 2021). In addition, there are institutional challenges which relate to ensuring that all digital data held by the police are stored, transferred, and deleted in compliance with the relevant data protection laws governing these processes (Al-Khateeb, H. M. and Cobley, P; 2015^{[footnote 226]}).

Following the collapse of the R v Allan^{[footnote 227]} trial, the digital evidence challenges faced by policing and subsequently the broader criminal justice system were brought into sharp focus^{[footnote 228]}. The rapid growth in the importance of digital forensics has meant that local procedures, policies and resources have struggled to keep pace with demand^{[footnote 229]}. With increasing attention being focused on digital forensics, numerous policies have evolved at speed, creating an overall state of digital confusion and a lack of consistency across many forces. (May et al., in preparation^{[footnote 230]}). In addition, analytical tools which should assist investigators have been variously described by officers and police staff as “time-consuming”, “inaccessible”, and “unable to communicate” with existing systems (Schreuders et al., 2018; McPhee and Rumney 2020; Rappert, Redfern and Wilson-Kovacs 2021). The result of which has been that manual, time consuming redaction, inputting and analysis is still being undertaken by officers in many force areas. In addition, respondents in Schreuder’s study (2018) also reported that their case management software was incompatible with other police systems, resulting in an ambiguity in record keeping and data duplication over many different systems (Schreuders et al., 2018).

Despite the increased attention and scrutiny that digital investigations are now under, and the increased policy, legislative, practice and guidance documents now available from the College of Policing, the National Police Chiefs’ Council, the Crown Prosecution Service, The Home Office, the Information Commissioner’s Office, the Victims’ commissioners and a number of public sector and charitable bodies working in this area, it is still widely viewed as an area of policing which remains ‘wanting’, regarding officer and staff understanding of the quality assurance measures, bespoke training, clear guidelines, appropriate technology, and investment (Foster et al. 2019; Rappert, Wheat and Wilson-Kovacs 2020; Muir and Walcott 2021; May et al. In preparation).

Methodological Approach

The methodological approach adopted by Pillar Six^{[footnote 231]} was mixed method and involved:

• Analysis of 137 anonymised RAOSO case reviews which included digital data, over a three-year period from 01/01/2018 to 01/01/2021^{[footnote 232]}

• Quantitative analysis of 14,708^{[footnote 233]} data strands relevant to police digital data^{[footnote 234]}

• 40 in-depth qualitative interviews with RASSO/Criminal Investigation Department (CID) officers and DFU staff

• 14 in-depth qualitative mapping interviews to understand the structure, capability, and capacity of digital and RAOSO systems and processes

• A review of 141 force policy, procedure and guidance documents relating to the: acquisition, seizure, analysis, storage, transfer, and deletion of digital data/evidence

• Documentation analysis of flow diagrams and documents provided by the four forces outlining the use of digital material and technology throughout the RAOSO investigative lifecycle, from incident notification to CPS file submission.

Methodological Rationale

The case file and quantitative data were requested to enable us to better understand what digital data is routinely collected in RAOSO investigations, its value to the investigation and the differences in data requests and analysis between victim, suspect, and witnesses. The case reviews also enabled collection of important factors involved in collecting digital material in terms of why it was collected, and what, for example, was extracted from a device. In-depth interviews were conducted to understand the decision-making processes behind the digital investigative strategy and officer/police staff views on the current digital “as is” in each force. The interviews also provided us with an in-depth understanding of officer/staff views regarding the strengths and weaknesses of each force’s digital capability, competence, and vulnerability. The mapping interviews provided us with a comprehensive understanding of the digital investigative RAOSO life cycle, from inception to case conclusion. Finally, the documentation review allowed us to analyse the digital policies, procedures, and guidance available to officers on each of the core processes involved in a digital investigation.^{[footnote 235]}

Data Limitations

Three of the four forces were unable to provide quantitative digital data, due to the time that would have been needed to extract these data, and the data being held across multiple systems none of which linked up. The fourth force, whilst providing over 14,000 data strands, had missing or no digital data in a significant proportion of cases. Whilst all of the forces provided case file data, we were limited to a relatively small number of cases due to the difficulty of drawing out the data, which is stored across multiple systems and had to be manually filled by officers, thus impacting on their workloads.

Findings

The four forces experienced digital challenges and displayed good practice across three key areas:

1. Institutional

2. Investigative

3. Technological

Interestingly, many of the concerns raised by interviewees did not relate to the technology itself, but rather the human factors which underpin the use of technology. We present our findings under the three key themes to have emerged from our deep dive research.

Institutional challenge 1: An absence of integrated digital evidence systems

Across the four forces there was a combination of data capture and redaction systems and processes in place. Some forces purchased digital systems; others had created their own in-house systems. Some were manually redacting files, others had simple redaction tools. Data analysis tended to be undertaken on key word searches, within defined time parameters, between named individuals or from certain platforms such as WhatsApp or Tinder. Officers from all four forces believed that the lack of technology and training in this area was inadequate, although all were aware of the work currently being undertaken by the National Police Chiefs’ Council, a number of digital forensic services, the Police Digital Service, the Accelerated Capability Environment, and the Government aimed at trying to resolve these issues. Without integrated data capture, redaction and analysis systems in place, we were unable to capture the entirety of the digital RAOSO cycle. The absence of a digital case management system able to capture and store data from both incident and investigation resulted in officers having to manually enter the same data across multiple systems, increasing both the administrative burden and the possibility of error/inconsistency. For new officers to a RAOSO team, there was an absence of any digital/investigative flow diagrams or documentation describing the RAOSO digital journey from incident notification to file submission to CPS, resulting in officers learning from others, which ultimately includes both exemplary and bad practice. For most forces, the DFU forensics network was separate from, and inaccessible via, the force network^{[footnote 236]}. In the absence of cloud technology, USBs were being used to transfer data between networks or officers had to travel to DFUs to view large files, in some forces this one journey would easily be a day’s work. This challenge also explains why the data for case reviews and large quantitative data sets were difficult to collect.

To get to the DFU I have to drive; it’s three hours there and three hours back^{[footnote 237]}, a whole day to view one file. It’s such a waste of resources

(RASSO officer Force D)

Institutional challenge 2: Staff resourcing

Staff resourcing was a challenge experienced across all four forces. A shortage of trained RAOSO staff was highlighted as one of the main reasons for the logjam in suspect phones waiting to be downloaded, which was exacerbated by the volume of RAOSO cases within the system at any one-time requiring phone downloads. The lack of trained staff was highlighted as the main contributing factor, increasing the workload pressure on trained staff. The result was a backlog of, predominantly suspect phones^{[footnote 238]}, waiting to be downloaded, which delayed investigations and increased the waiting times for victims, witnesses and suspects alike. In addition to forces moving to a more proportionate approach to phone downloads, the introduction of the Level 1^{[footnote 239]} phone download kiosks was, also reported to be assisting with reducing the time taken to extract and analyse the phones backlog in all four forces.

There are a staggering amount of digital devices and they take an awful lot of time to analyse, download, etc. Be it here, or at our specialist unit we just don’t have the man hours that we need. That’s not just a statement, we just don’t have enough cops…staffing is far worse than anyone realises

(RASSO officer Force B)

Institutional challenge 3: Storage and deletion of data

Despite all forces/officers being bound by the principles of the Management of Police Information (MoPI), which sets out how long data should be stored for dependent on crime type and age of victim, and that every force should have a Data and Governance Lead able to provide guidance on individual force retention and disposal schedules, the storage and deletion of digital data was found to be a challenge, with some participants expressing confusion about policies which were either unknown or not being followed. For example, in one force, when asked about data storage protocols for RAOSO cases, participants cited varying lengths of data storage time ranging from six months to 30 years. From the case reviews, there was inconsistency both across and within forces in how digital data is stored, some cases cited storing phone downloads/similar content on USB and DVD/Disks, whereas some data was stored on secure servers with high levels of security. Concerns were also raised by some interviewees about the policies and protocols regarding staff access to phone downloads, highlighting that access to a victim or suspect’s phone data was not limited to those working on the case.

If it’s stored in the property store, for instance, that can be taken out by any officer that wants to

(Officer Force B).

Every police officer has access to the software, so they could access the file and then look at it, you just need to know the reference number

(Officer Force E)

I can access any of them [phone downloads] they’re not password secured; I can look at all of them…they should only be open [accessible] to the people that are invested [working] on that case

(RASSO Officer Force D)

CCTV and pictures are often stored on a system called Evidence Works, nicknamed within our force evidence doesn’t work…… We’ve had notorious issues with it. It doesn’t always upload; it’s always going offline, and things take ages to upload.

(Officer Force E)

Institutional challenge 4: Training and professional development

Compounding the issue of officer and staff resourcing was the perceived lack of appropriate training and expertise outside of DFUs, regarding analytical capability, data interpretation, and the presentation of digital material in RAOSO cases, which was considered vital, but across the four forces, lacking. Interviewees felt that the current training was basic, too generalised and lacked any specific training around the use of digital material in RAOSO cases^{[footnote 240]}. Interviewees believed the current approach to be “worryingly problematic” given the pivotal role of digital material in many RAOSO investigations. Participants in two of the forces said they tended to learn on the job, often however, from relatively inexperienced RAOSO staff. There were also concerns regarding the lack of expert digital knowledge, which officers thought was needed from the outset of an investigation, across the four forces within RAOSO units and CID departments, which was said to create investigative time delays which resulted in important evidence being missed. Interviewees stated that access to digital expertise should be considered both routine and essential, not a luxury add on, especially given the rapid advances in technology.

It would be good if everyone was given at least some sort of a basic foundation of digital knowledge (Officer Force D)

So, you’re being taught by somebody with very little experience, our training is a photocopy of a photocopy of a photocopy. You only get taught what your tutor knows, and then you only teach what you know and so the image degrades, the quality degrades. We learn to investigate off one another, to the benefit of some and the detriment of others (Officer Force F)

Investigative challenge 1: A lack of formal protocols^{[footnote 241]}

One key investigative challenge concerned a lack of formal protocols surrounding digital evidence in RAOSO investigations. In Force B officers highlighted how they tended to operate independently, with few protocols in place aimed at ensuring consistency of approach with regards to digital material and no standardization when interrogating data. Across the four forces, there were no guidance documents which specifically addressed:

• Reasonable lines of enquiry

• Record keeping at key stages in the digital investigative process: seizure, extraction, analysis, storage, transferring and deletion

• Victim, witness, and suspect consent to acquire digital devices^{[footnote 242]}

• Data destruction

Whilst force documentation mentioned all the above, none were covered in detail, and none provided illustrative examples for officers to follow. Similar to reasonable lines of enquiry, the importance of keeping up-to-date records throughout the different stages of the process was mentioned, however, there was an absence of dedicated guidance and supervision, which interviewees ultimately thought led to differing practice being adopted within and across force areas. Another important theme absent from force guidance documents was an explanation and definition of victim and witness consent. Documents referenced and mentioned the importance of victim and witness consent however, for new RAOSO officers, there was no further explanation. In essence, the notion of consent was highlighted as being important, but no explanation of what consent means what to do if consent is withdrawn and the circumstances under which this should be obtained were provided. Data destruction was another important administrative and legal process which lacked guidance across the four forces. There was also little mention of this issue within any other force guidance. This finding was supported by the mapping interviews.

there’s no specific guidance or document that tells us what we’re going to be looking at, or when we need to look at it.

(Officer Force B)

Investigative Challenge 2: Volume of data and proportionate downloads

Across the four forces, participants identified the problematic nature regarding the volume of digital data available to them, highlighting it as one of the key digital challenges facing the police, particularly in RAOSO investigations. Several participants voiced their concern about how fundamentally ill-equipped policing is to manage the current volume of digital data, which will only increase and in all likelihood become more complex as new digital devices and apps enter the market. For officers and DFU staff, the constant digitization of our lives and the volume of data this brings to even the simplest of investigations, is demanding and was also described as “demoralising”. The case reviews found that the majority of analysis involved key word searches and restricting initial data to within specific date/time parameters, the latter of which had the highest proportion of relevant material found. Whilst participants described using key word searches to limit the data download, none had received any external training from experts in intimate partner/acquittance violence, coercive control or manipulative practices aimed at shaming and/or silencing a victim. As such keyword searches tended to be basic and sexual: such as “nob, dick, boobs, fisting, drugs, porn, fuck, finger, tits, underwear, consent, headlock, role play, victim, police, rape and penis”. In addition, RAOSO officers across the forces stated that they had received little training on how to decide upon and define extraction parameters.

Officers across the four forces stated that there were very few quality control measures in place to ensure that a RAOSO investigators extraction parameters, analysis and review of phone download data was proportionate and appropriate for the investigation.

On a recent investigation the phone download was colossal, there were half a million text messages alone

(Officer Force E)

Similar to findings from Pillars One and Two on suspect focused investigations and Pillar Three on victim engagement, RAOSO interviewees from Pillar Six, whilst understanding the importance of “proportionate downloads”, also emphasised how entire downloads from both victims and suspects were both “preferable” and “beneficial” to the investigation. One officer, from Force E, viewed bulk downloads as a tool which ultimately saved investigative time: “doing a bulk download would save us time for sure, if we just take everything at the first opportunity”. However, another stated that a full download can produce a “few terabytes of data, and that’s then a biblical problem”. The desire for a bulk download, articulated by some officers, may, however, be more indicative of how the digital investigative strategy in some RAOSO investigations is divorced from the wider investigative strategy; and as such, not integrated or thought through using the same principles.

We should always get the most out of the device…. once that phone leaves your custody and goes anywhere else, it could be wiped, it could be corrupted, it could be broken, it could be damaged. It could be changed

(Officer Force D)

Investigative Challenge 3: Tensions with the CPS

Across all four forces, participants spoke of a ‘culture clash’ between the police and CPS. A common frustration was of the CPS frequently requesting full downloads of victims’ phones or asking the police to request access to third party material, such as doctor, social service, or school records. In addition, interviewees expressed frustration at the CPS’ refusal to accept evidence in certain formats, such as screenshots of messages, which, whilst such measures may not be considered forensically sound, officers believed was less intrusive and avoided victims having to hand over their mobile phone to have data extracted from it. Whilst there is still an absence of technology which fully supports selective extraction^{[footnote 243]}, officers thought that the ‘police’ were slowly moving towards a more proportionate approach to digital data extraction, participants felt this had yet to be fully embedded within the CPS. Participants also spoke of poor communication between themselves and the CPS, with submissions reportedly often sitting in a portal queue for long periods of time before receiving a response from a CPS lawyer.

The CPS system is very behind…it’s bureaucratic, difficult and [processes are] really time consuming. There’s a lot of issues where they say they don’t receive things and that you must resend them, and that you have to put evidence in a certain format with a certain naming convention. It’s just all very onerous

(Officer Force F)

Technological Challenge 1: Technological resources

Throughout our study, participants highlighted how they struggled to access the necessary hard and software to process mobile phones and other digital devices in a timely and effective manner. The software used for even the simplest of digital interrogations was criticised for its lack of innovation and sophistication. Officers were also critical of the ability of policing to keep up with the rapid development of technology. Simply having access to the necessary software was raised as an issue. A number of participants highlighted how different departments tended to share access to specialised software, which resulted in staff from across a number of departments, who access the same software, having to wait to use it, adding delays into the digital element of many investigative process. For officers, this created unacceptable delays for both victim and suspect.

What keeps people’s data more secure makes it more inaccessible for us, it’s an arms race that we [the police] are losing (Officer Force F)

There is a frustration around waiting for the tools that enable you to do your job properly

(Officer Force E)

We only have two laptops capable of doing cloud extractions for an office of 20 staff

(DFU Force D)

Technological Challenge 2: Technological capability

Across the four forces, technological capability was spoken about as being ‘just about good enough’. And “it’s not fantastic…it’s sufficient”. Whilst technological capability is being addressed by the Police Digital Service through the Digital Enablement Programme, constant phone upgrades and app developments with increased encryption means that specialised and expensive software quickly become obsolete. The capability of the software available to policing was said to impact on the transfer of information from the police to the CPS, who at the time of writing had a 1MB limit on document transfer, which means that most submissions to the CPS need to be broken up, a process which is incredibly time-consuming for officers and police staff and one that has a significant impact on officer workload, victim frustration and criminal justice case delays.

If we had more trained people, it [digital investigations] could be done a lot better

(Officer Force B)

Imagine you have a rape case with multiple documents and each one of them invokes the same procedure. It makes people cry. You see them on the scanners for two or three days at a time. When you ask them what they’re doing, it’s stuff that you or I could have attached to an email and sent in seven seconds. Instead, we have someone on a photocopying scanner for about a day

(Officer Force B)

Investigative good practice

Whilst there were many challenges highlighted to us by digital and RAOSO officers and police staff, there were also officers striving to implement good practice and innovation across the four deep dive forces.

Institutional good practice: Specialism

Not all forces had introduced specialists, however, in Force D, and E specialism was starting to be introduced. In Force D, digital media investigators (DMIs) were employed to conduct the vast majority of phone extractions, saving RAOSO investigators time, whilst also allowing a RAOSO investigator to discuss their digital strategy with a digital specialist. Force F were in the process of introducing dedicated phone download and extraction posts, aimed at saving investigators time whilst also recognising that specialist skills are needed to work with victims to develop a plan in the least intrusive, most collaborative way. Access to this type of expert knowledge was highlighted by interviewees as a possible way to reduce delays, save money, and ultimately improve RAOSO investigations.

We’re going to have these core specialists (phone download and extraction staff who are going to nail this down for us. It’s going to save us work and it’s going to improve the process for the victims, I hope.

(Officer Force D)

Across the four DFUs, RASSO and CID units, officers generally agreed that specialist roles and knowledge were deemed good practice in RAOSO investigations, not only regarding the handling of digital material but also the investigation as a whole. One force discussed their recently established dedicated rape investigation unit team. Whilst acknowledging that it was too early to establish any long-term impact resulting from this institutional change, interviewees believed that the re-structure was viewed by officers as a positive move. Officers reported that the introduction of the dedicated rape unit now allows RAOSO cases to maintain a priority without removing staffing resources from other important cases, such as major crime, benefits of this approach highlighted by interviewees included providing a better service to the victim, improving officer morale, and lightening the current excessive workload pressures.

I’d like to think by having a dedicated team we’re providing a better victim service because we only focus on rapes. The people that are on the team are here because it’s an area we’ve either got expertise or interest in, and we’re not trying to manage burglaries and robberies and everything else alongside it. We can focus a lot more time into this one specific area. So, I’d like to think it’s made improvements for victims. I think it’s had an improvement for officer morale because the people that are working on the team want to be here, and I know it’s helped CID on non RASSO workloads, because they’ve seen a reduction in their workloads so they can be more productive. So, I think holistically it’s been quite positive for the officers and maybe the victims.

(Officer Force D)

We’re going to have these core specialists (phone download and extraction staff) who are going to nail this down for us. It’s going to save us work and it’s going to improve the process for the victims, I hope.

(Officer Force D)

Investigative good practice: Victim sensitivity

Across the forces, we found a shifting emphasis from wholesale digital data grabbing to a more proportionate victim-centred approach. Many officers highlighted how reticent they would be to hand their phone to a police officer they had never met before, highlighting perhaps a greater sensitivity towards the victim’s experience. Whilst not all officers shared this view, there were many who had actively moved away from requesting a full phone download, instead using a more proportionate approach, with parameters having been established with the victim, and consent gained through the use of Digital Processing Notification forms^{[footnote 244]}. This was, however, coupled with the frustration from officers at the perceived slow progress on finding solutions to selective extraction and automated analysis.

Interviewees also stressed the importance of returning victims’ phones as soon as possible after extraction. All four forces highlighted their commitment to achieving the Government’s ambition “that no adult rape victim will be left without a mobile phone for more than 24 hours in any circumstances.”^{[footnote 245]} Whilst there has been change in this area of policing, officers were aware that there is still a long way to go to convince victims that their phone is not going to be subjected to a full and thorough ‘digital strip search’.

Since Operation Soteria has come out, the victim’s phone in rape cases is our top priority. We do kind of drop anything that we’re doing when a victim rape case comes in.

(Officer Force D)

We’ll always now go for less intrusive means, and I think that applies equally to victims and suspects because they’re both deserving of privacy and certainly nobody’s been convicted of anything at that point

(Officer Force F)

Technological Good Practice: Home Office Funded Technology Uplift

Whilst not all of our five forces had received a Home Office funded technology uplift, those that had, stated that the uplift had improved the technological resources available to them and as a result, their perception was, that they had improved the service provided to victims and suspects. Digivans were reported to be a ‘game changer’ in terms of allowing trained digital media investigators to visit a victim at a location convenient to them and, if appropriate, have data extracted from their device. If the victim preferred to attend the station, this option is now available as an appointment, arranged at a convenient time for the victim. Prior to this approach being adopted a victim often handed the phone to the investigating officer not always knowing when it would be returned. Replacement phones were also offered to victims in some of the Operation Soteria Bluestone forces, although these were described by some officers as embarrassing, as they lacked any credit, were old models and had not been charged prior to being handed to the victim.

Innovation and promising practice

In addition to the Force technology uplifts, we also identified examples of local innovation regarding the current digital funding/technology/skill challenges facing forces. In Force F, the DFU had developed a third-party material question set, the aim of which is to inform the data extraction and investigative strategy. Officers reported that the approach assisted them to sift, expel and process the vast volumes of data acquired and significantly reduced the need for bulk downloads. Officers described the question set as a “valuable tool” for RAOSO cases, providing a systematic strategy through a set of questions which both the victim and suspect provide answers to identify the information needed.

The Detective Inspector of this team highlighted the importance of developing a clear digital strategy, not only in terms of the time it saves an investigation but also in reducing the intrusion to the victim and suspect and reducing the volume of data needing to be interrogated. The question set is not perceived as intrusive and moves away from the culture of “capture everything”. The same force was also exploring screen mirroring data capture, this process was reported to assist officers to capture data that is sometimes difficult to download in a readable format, e.g., WhatsApp messages. Instead of downloading this type of message, it is recorded using a mirror. Mirroring and screen-recording allows officers to capture the material they are specifically interested in.

We now have some questions sets specifically for RASSO, which are done as part of the interview with the victim, and the suspect, which should be done before the download…it’s really important that we know how these people are communicating as part of the interview so that we can then do our digital strategies correctly and make sure we don’t just download the phones for the sake of it

(Officer Force F)

We’re trying out this other bit of kit at the moment that allows us to do consensual video screen capture of the phone. Where the phone is plugged in to essentially a large screen and actions on the phone are recorded.

(Officer Force F)

Acknowledgements

The Pillar Six team would like to express our thanks to the officers and police staff across the four deep dive forces, who provided us with quantitative data, which was incredibly difficult to extract; gave up their time to be interviewed, providing us with their thoughts and insights on the challenges they face and the innovation they are trying to implement. We would also like to thank the Op Soteria Bluestone Home Office and MOPAC teams for their support and for keeping us on track. Finally a special thanks to the research teams from Pillars One-Five and to Betsy Stanko and Katrin Hohl for welcoming Pillar Six with such supportive open arms.

Appendix 13: Published academic papers

The following academic papers have been published based on the Project Bluestone pilot. The articles are published “open access” which means they can be read and downloaded free of charge, and without a need to register.

Hohl, K., Stanko, E.A. (2022). Five Pillars: A Framework for Transforming the Police Response to Rape and Sexual Assault. International Criminology, 2, 222–229. https://doi.org/10.1007/s43576-022-00057-y

Dalton, C. T., Barrett, S., Horvath, M. A., & Davies, K. (2022). A systematic literature review of specialist policing of rape and serious sexual offences. International Criminology, 2, 230–252. https://doi.org/10.1007/s43576-022-00062-1

Hohl, K., Johnson, K., & Molisso, S. (2022). A Procedural Justice Theory Approach to Police Engagement with Victim-Survivors of Rape and Sexual Assault: Initial Findings of the ‘Project Bluestone’ Pilot Study. International Criminology, 2, 253–261.

https://doi.org/10.1007/s43576-022-00056-z

Williams, E., Norman, J., Ward, R., & Harding, R. (2022). Linking professionalism, learning and wellbeing in the context of rape investigation: Early findings from Project Bluestone. International Criminology, 2, 62–275. https://doi.org/10.1007/s43576-022-00059-w

Norman, J., Fox, A., Harding, R., Majid, A., Williams, E., Davies, K., & Horvath, M. A. (2022). Critical reflection: The importance of case reviews and reflective practice in rape and serious sexual offences investigations. International Criminology, 2, 2, 276–285.

https://doi.org/10.1007/s43576-022-00061-2

Maguire, L., & Sondhi, A. (2022). Stress-related psychosocial risk factors among police officers working on Rape and Serious Sexual Offences. The Police Journal, online first.

https://doi.org/10.1177/0032258X221128398

Lovett, J., Hales, G., Kelly, L., Khan, A., Hardiman, M., & Trott, L. (2022). What can We Learn from Police Data About Timeliness in Rape and Serious Sexual Offence Investigations in England and Wales? International Criminology, 2, 286–298.

https://doi.org/10.1007/s43576-022-00069-8

Stanko, E. A., & Crew, S. (2022). From Project Bluestone to Operation Soteria Bluestone: An Academic-Police Collaboration. International Criminology, 2, 299–304. https://doi.org/10.1007/s43576-022-00065-y

https://doi.org/10.1007/s43576-022-00056-z

to Rape and Sexual Assault. International Criminology.

Lovell, R., Huang, W., Overman, L., Flannery, D., & Klingenstein, J. (2020). Offending histories and typologies of suspected sexual offenders identified via untested sexual assault kits. Criminal Justice and Behavior, 47, 470-486.

Woodhams, J. & Bennell, C. (Eds.). (2014). Crime linkage: Theory, research, and practice. CRC Press.

Demerouti, E., Bakker, A. B., Nachreiner, F., & Schaufeli, W. B. (2001). The job demands-resources model of burnout. Journal of Applied psychology, 86(3), 499.