Written evidence from The Violence, Health and Society (VISION) consortium [EOV0036]

The Violence, Health, and Society (VISION) consortium is funded by the UK Prevention Research Partnership. We are a collaboration of epidemiologists, economists, data scientists, criminologists, evaluation experts, psychiatrists and more from multiple universities (including Bristol, City, Kings, Suffolk, and UCL). Our research brings data together from health and crime surveys, health services, police, solicitors, and third sector domestic and sexual violence specialist services. Together, over the course of our five-year project, we aim to improve the measurement of data on violence to influence policy and practice and reduce violence and the health inequalities that result.

Dr Jessica Phoenix is an Evidence Based Policing Specialist for a UK police force, where she oversees the force’s engagement with research, and assists with the implementation of research evidence into police policy and practice. Jessica’s research interests include policing, domestic abuse, repeat victimisation, and the measurement of violence.

The Section of Women’s Mental Health at the Institute of Psychiatry, Psychology and Neuroscience, King’s College London(SWMH) does research which informs policy and practice on women’s mental health. The SWMH leads research which aims to reduce the risk and impact of violence and abuse, with a focus on domestic and sexual violence and human trafficking, including through our leadership of the UKRI Violence Abuse and Mental Health Network.

VISION colleagues have also collaborated with the University of Essex on a Home Office funded perpetrator project.  The work led by Dr Katerina Hadjimatheou used unsupervised machine learning to analyse a longitudinal dataset of domestic abuse incidents from Essex Police.

Evidence and recommendations

Intervene early at both an individual and systems level

It is well established that violence against women and girls causes substantial harm to victim-survivors, even contributing to suicide (McManus et al., 2022). It is also increasingly recognised that those who use violence (Bhavsar et al., 2023), as well as those who come into contact with the prison service because they have committed crimes (Bebbington et al., 2021), are also more likely to experience poor mental health and suicidality. For these groups, violence is often used and experienced in a context of longer-term multiple adversity (Agenda Alliance and VISION, 2023). Men who’ve been to prison are about five times more likely than other men to have been raped in childhood (Bebbington et al., 2021). For some, therefore, the trajectory towards violence to others starts early. The most effective, and cost effective, point at which to disrupt the escalation in violence towards women and girls is therefore early intervention and properly resourced support for children and young people at risk.

Research evaluation efforts of the Drive Project Model at the system level suggests that an effective response requires a well-functioning, evidence-informed and equitable multi-agency ecosystem. Additionally, that there are acute systemic gaps in provision and development of interventions for marginalised groups, such as racialised communities (Adisa and Allen, 2020; Adisa et al., 2023). Individual-level factors relating to race/ethnicity escalates high-harm perpetration (Adisa et al, 2021). Alongside this, Environmental factors also continue to play a role in individual-level escalation in violence and harm, with responses to the Covid pandemic being one high profile example (Bhavsar et al., 2021). Evidence on interventions and responding to offender behaviours at a systems level remain lacking, and more systems-oriented research which takes into consideration the dynamics and multidimensional aspects of violence, and violence prevention, is needed to address this gap.

Longitudinal data needed

Mapping out life course trajectories in use and experience of violence is severely limited by the fact that violence is rarely asked about on longitudinal population surveys or birth cohort studies (Cook et al., 2022). Data that enable monitoring of change in use and experience over time is urgently needed, including questions about behaviours that would not meet criminal thresholds, may not be reported to police, and which span all types (physical, sexual, emotional, economic) and relationships (intimate, other domestic, acquaintance, stranger). Such data collections should cover the whole population, not be siloed to violence victimisation against women and girls. Violence is gendered, and whole population data collection is needed to understand that gender patterning, as well as in order to capture the occasions (albeit rarer) when men are victims and women perpetrators too.

Challenges of using police data to map repetition

We alert the committee to the work of Dr Phoenix (Phoenix, 2023). Her analyses of police data show that understanding repetition requires repeat demand to be more accurately measured in police data. Her work highlights how the ways in which police information systems are organised can hinder the effective measurement of repeat domestic abuse-related demand. Her observations of the police response to domestic violence and abuse and her close review of case files, led her to identify four key issues: (1) fragmented units of measurement across multiple information systems; (2) inconsistent recording of personal details; (3) multiple methods of identifying DVA; and (4) the embedding of information in free-text. Her work should be reviewed, and her recommendations implemented if escalation is to be better identified in police data.

Additional issues (identified in recent work by Dr Hadjimatheou and colleagues, 2023) relate to the recording of incidents including specifically ‘non-crimes’. Their longitudinal study of domestic abuse found that a significant proportion of incidents are recorded as non-crimes, that is, as behaviour that falls short of a crime or for which there is little or no evidence. Unlike criminal incidents, which are typically recorded in terms of their nature (e.g., assault, public order offence), non-crimes tend to be undifferentiated in the data. New research is needed mapping the sequence and chronology of non-crimes and crimes, and police forces should pilot recording the nature of non-crimes, so as to better understand their role in escalation pathways and better predict offending trajectories.

Insights from specialist services work with people who use violence

In a systematic review currently being prepared for submission looking at the effectiveness of UK-based support services and interventions for adults who have experienced domestic and sexual violence and abuse, we found evidence from eleven papers that advocacy/ Independent Domestic Abuse Advocate services increased the number of participants reporting a cessation of abuse from pre- to post-intervention (by 58.7%) (Carlisle and colleagues). Similarly, evidence from five papers reporting data on outreach services found that the proportion of participants reporting a cessation of abuse was 46.2% at post-intervention, compared to 0% at baseline. Therefore, such interventions and services appear to be effective for abuse cessation.

Mental health services and the escalation of violence against women and girls

We now have up to date evidence that perpetrators of partner violence are more likely to access mental health services in England than the general population and that a relevant sub-group of perpetrators of partner violence who have recent contact with mental health services have no history of criminal justice involvement (Bhavsar et al., 2022). Mental health service contact is common in the histories of perpetrators of domestic homicide (Home Office 2022, Chantler et al., 2019) however, we are aware of no available evidence in this context on escalation to homicide from non-criminal offending. This underlines the need to develop evidence urgently on trajectories of offending behaviours in perpetrators of partner violence (and other domestic abuse) to inform healthcare improvement and prevention. In line with wider evidence, we have found that association of partner violence with mental health service use is not attributable to a feature of “generally violent behaviour”, suggesting that perpetrators of violence against women and girls seen in mental health services may warrant specific response strategies, apart from broader responses to offending behaviour as a whole. Improved measurement of non-criminal behaviours, relevant to the Call for Evidence, is needed in mental health service populations.

Notes of caution on relying on risk assessments or making assumptions about severity

We also note that care is needed with any reliance on risk prediction tools and models and with implied severity hierarchies when thinking about escalation, for example in terms of the perceived relative severity of physical violence and threat and control. There are multiple dimensions to understanding potential increases in severity, including frequency, duration, pervasiveness/incident count, and physical and emotional injury, but ultimately to understand escalation we need to map changes in the harm experienced by victim-survivors. Additionally, focusing on offender profiles and histories as a fixed and bounded concept can limit understanding of perpetration and violence escalation, especially in cases where there are no convictions. Evidence from SafeLives has found that ‘less than 1% of perpetrators receive any form of intervention to address their behaviour’ (SafeLives). Dynamic understandings of offender profiles and intervention pathways are important.

References

Adisa, O., Allen, K., Manning, M., Ferreira, J. & Horvath, M. (2023) Drive Systems Change Evaluation: Final Report. Institute for Social Justice and Crime, University of Suffolk. (Report under embargo, public release date: Nov 2023).

Adisa, O., Bland, M. & Weir, R. (2021) Identifying predictors of harm within Black, Asian, and other racially minoritised communities. University of Suffolk. Link: https://openaccess.city.ac.uk/id/eprint/28047/1/FINAL%20Predictors%20of%20Harm%20UOS%20report.pdf 

Adisa O., and Allen K. (2020). Increasing safety for those experiencing family and intimate relationship harm within black and minority ethnic communities by responding to those who harm: Survey findings. University of Suffolk.       https://www.uos.ac.uk/sites/www.uos.ac.uk/files/Family%20and%20intimate%20relationship%20harm%20within%20black%20and%20minority%20ethnic%20communitiesFINAL%20%281%29.pdf 

Agenda Alliance and VISION (2023). Underexamined and Underreported-Suicidality and intimate partner violence: Connecting two major public health domains. https://www.agendaalliance.org/our-work/projects-and-campaigns/underexamined-and-underreported/

Bebbington, P. E., McManus, S., Coid, J. W., Garside, R., & Brugha, T. (2021). The mental health of ex-prisoners: analysis of the 2014 English National Survey of Psychiatric Morbidity. Social Psychiatry and Psychiatric Epidemiology, 1-11. https://link.springer.com/article/10.1007/s00127-021-02066-0

Bhavsar, V., Kirkpatrick, K., Calcia, M., & Howard, L. M. (2021). Lockdown, domestic abuse perpetration, and mental health care: gaps in training, research, and policy. The Lancet Psychiatry, 8(3), 172-174. https://www.thelancet.com/journals/lanpsy/article/PIIS2215-0366(20)30397-7/fulltext

Bhavsar, V., McManus, S., Saunders, K., & Howard, L. M. (2023). Intimate partner violence perpetration and mental health service use in England: analysis of nationally representative survey data. BJPsych open, 9(3), e64. https://www.cambridge.org/core/journals/bjpsych-open/article/intimate-partner-violence-perpetration-and-mental-health-service-use-in-england-analysis-of-nationally-representative-survey-data/142732D03264D9CB11B0EDCFFB0D7FF0

Barrow-Grint, K., Sebire, J., Turton, J., & Weir, R. (2022). Policing Domestic Abuse: Risk, Policy, and Practice. https://books.google.co.uk/books?hl=en&lr=&id=BfqGEAAAQBAJ&oi=fnd&pg=PA1983&dq=ruth+weir+domestic&ots=ftmnZaZ1Aq&sig=JemwTOvhpqG-VYdiXiK9_Tf7VS0&redir_esc=y#v=onepage&q=ruth%20weir%20domestic&f=false

Carlisle, S., Bunce, A., Prina, M., McManus, S., Barbosa, E., Feder, G., & Lewis, N. (under review). How effective are UK-based support interventions and services targeted at adults who have experienced domestic and sexual violence and abuse at improving their safety and wellbeing? A systematic review protocol.  https://europepmc.org/article/ppr/ppr691516

Chantler, K., et al. (2020). Learning from domestic homicide reviews in England and Wales. Health & social care in the community. 28(2): 485-493.

Cook, E., Markham, S., Parker, J., John, A., Barnicot, K., & McManus, S. (2022). Risk, responsibility, and choice in research ethics. The Lancet Psychiatry, 9 (1), 5-6.

https://www.thelancet.com/journals/lanpsy/article/PIIS2215-0366(21)00434-X/fulltext

Day, A. S., Jenner, A., & Weir, R. (2018). Domestic abuse: predicting, assessing and responding to risk in the criminal justice system and beyond. Women and the Criminal Justice System: Failing Victims and Offenders?, 67-94. https://link.springer.com/chapter/10.1007/978-3-319-76774-1_4

Hadjimatheou, Katerina and Quiroz Flores, Alejandro and Weir, Ruth and Skevington, Taylor (2023) Understanding Domestic Abuse Perpetrators: using unsupervised machine learning to analyse a longitudinal dataset of domestic abuse incidents from Essex Police. A Home-Office funded project. 2022. Project Report. University of Essex, London. https://repository.essex.ac.uk/35676/

Home Office (2022). Domestic Homicide Reviews:  Quantitative Analysis of Domestic Homicide Reviews October 2020 – September 2021. Home Office.

Phoenix, J. (2023). Improving Police Data Collection to Measure Repeat Demand: A Focus on Domestic Violence and Abuse. Policing: A Journal of Policy and Practice, 17, paad022. https://academic.oup.com/policing/article/doi/10.1093/police/paad022/7160184

SafeLives (2018) Introduction to Beacon: I’m a person. I don’t fit in to one of your categories. https://safelives.org.uk/sites/default/files/resources/Beacon%20specification%20-%20Appendix%20A_0.pdf

September 2023