While experiences and reactions are complex and highly individualised (VAWnet, 2006), a large body of research and practice links sexual victimisation with experiencing a range of mental health struggles, including general psychological dysfunction, post-traumatic stress, depression, anxiety, suicidal ideation/attempts, substance abuse, sexually risky behaviours, and eating disorders to name a few (Campbell et al., 2009; Dworkin et al., 2017; Tewksbury, 2007). Conversely, mental health difficulties can elevate vulnerability to sexual victimisation (Miles, 2019), and, when a history of sexual assault already exists, the risk for repeated sexual violence increases (Lovell et al., 2021).
In this context, Kaul and colleagues (2024) conducted a systematic review and meta-analysis of peer-reviewed studies examining the prevalence and risk of sexual victimisation among users of psychiatric services. Their aim was to assess prevalence and relative risk of sexual violence victimisation for male and female service users depending on the type of mental health services attended: inpatient, outpatient, and mixed services.
The negative impact of sexual violence victimisation on mental health has been extensively studied. This review provides data on the prevalence and risk of sexual victimisation in people who already use mental health services.
Methods
The authors assessed prevalence and risk in the past year (fluctuating depending on primary study’s date) and in adult lifetime victimisation. The latter was defined as experiences having occurred ≥16 years old. Studies on childhood sexual victimisation, grey literature and non-English language papers were excluded.
Using the WHO definition of sexual violence, three large electronic databases were searched from inception to 18 July 2022. Included studies:
- analysed male and/or female mental health service users aged ≥18 years;
- measured adult lifetime and/or past year, and relative risks of sexual victimisation, or had sufficient data for additional odds ratios analyses;
- employed experimental, before and after, interrupted time series, cohort, case-control, and cross-sectional study designs.
Evidence quality was assessed using a modified Newcastle-Ottawa Scale. Variability in findings across studies (i.e., heterogeneity) was calculated using I2 statistic. Random effects meta-analyses were conducted for studies reporting prevalence by sex.
Results
Study characteristics
Post screening, 26 studies encompassing 197,194 service users were included. Twenty-three studies were conducted in “high income countries”, particularly the United States (8), and 2 were from “low- and middle-income countries” (India and Brazil).
The majority (19) were of medium quality, 6 were high quality, and 2 were low quality. Disaggregated data from twenty studies were subsequently included in the meta-analyses.
1. Past year and adult lifetime prevalence
Eleven studies explored past year prevalence and fourteen explored adult lifetime prevalence.
Female mental health service users, regardless of and across settings had overall higher prevalence of sexual victimisation than males, both in the past year (13% vs 3%) and across their adult lifetimes (28% vs 8%). There were no studies of past year prevalence for females in inpatient only settings, but adult lifetime prevalence in inpatient settings still show overall higher prevalence for females than males (21% vs 6%). This disparity remained true in outpatient and mixed services. Some of the highest prevalence in women was noted in samples of episodically homeless women, and those with severe mental illness. Amongst men, the highest prevalences were noted in military veterans, especially when they had attempted suicide. Prevalence results increased slightly when low quality studies were removed from analyses, usually by up to 2%. Heterogeneity was high between studies: between 85% and 97%.
2. Overall risk
Mental health service users overall had higher risk of past year victimisation than the general population. The authors state that, at the lower end, they had two times the odds of victimisation and, at the top end, “the largest study found their sample of 936 psychiatric service users were over 17 times more likely to experience sexual violence than 32,449 general population controls.” It is unclear if this risk was as high for adult lifetime victimisation as, according to authors’ supplementary materials, lifetime risk was reported only for those in outpatient settings.
The odds were higher for male service users compared to controls, even when the studies involved people from non-urban areas or excluded patients with alcohol and substance abuse – known risk factors for victimisation. Female mental health service users had higher odds than men of past and adult lifetime victimisation compared to controls.
Results of this review indicate that regardless of settings and sexes, mental health service users had higher prevalence of past year and adult lifetime sexual victimisation than controls.
Conclusions
Mental health service users had higher prevalence of sexual victimisation, both in the past year and across adult lifetime, compared to controls. The higher prevalence remained across mental health service settings – although there were some large variations in reported prevalence for men. Female mental health service users had particularly high rates of sexual victimisation, especially when adult lifetime prevalence was considered.
Mental health service users have higher prevalence and risk of sexual violence victimisation than controls. The prevalence and risk of victimisation is higher for women than men.
Strengths and limitations
The review provided up-to-date prevalence and risk information by including studies spanning 1983 up to 2020. It focused on selecting data that could be disaggregated by mental health service user sex (important for understanding trends given the gendered nature of sexual violence) and further presenting it by type of setting attended by the service users (important when trying to understand type, complexity, and unique needs of patient presentations to develop tailored support strategies). Nonetheless, there are some limitations.
The authors themselves note the review is limited by the exclusion of grey literature and papers not in English. Grey literature, such as institutional reports, are particularly useful for current and fast data, which research project timescales might miss. For example, a Care Quality Commission report (2018) speaks of rates and risks of sexual abuse within mental health inpatient wards.
These decisions may have led in part to the relatively small sample of reviewed papers. While this shows the paucity of appropriate research in the field, it also restricts the results’ overall generalisability and trustworthiness. For example, higher past year risk for female outpatients was based on 2 studies. Notably, results on lifetime risk of sexual violence in male and female were also based on only 2 papers.
Considering the methodological quality, inconsistencies, and paucity of data in the primary studies, as noted by the authors, they were unable to:
- Calculate pooled odds ratio of sexual victimisation for neither sex, meaning that while findings on unequal prevalence and risk remain true, the true extent is unclear.
- Account for confounding factors (e.g., substance use, age, ethnicity), which are important in understanding prevalence and, especially, risk of sexual victimisation.
- Fully disaggregate victimisation by mental health diagnoses and all types of services.
- Generalise findings across all service user samples, regardless of mental health diagnosis, social characteristics, and settings.
It is also worth noting that different papers used different control groups, but in the reporting of odds ratios it is not always clear who the comparison population comprises. As the results and their magnitude are relative to the baseline comparison, this detail should not have been omitted.
The reporting of this review makes it unclear who the comparator group is, and whether results represent odds ratios, prevalence ratios, rate ratios or relative risk ratios.
Implications for practice
Kaul et al’s (2024) review uncovers several areas for much needed improvement and opens timely discussions in research and practice. Despite methodological limitations of the included papers, the review’s findings are consistent to similar, older research (Khalifeh et al., 2016; Maniglio et al., 2009).
The testimonies of sexual violence victims are often not given due weight by investigative and criminal justice agencies and may be dismissed as unreliable because of their psychiatric diagnosis (Wieberneit et al., 2024). The review’s evidence on the higher prevalence and risk of victimisation amongst service users helps reduce this bias.
With heterogeneity between the studies reaching nearly 98%, the review highlighted the considerable discrepancies in the quality and conduct of studies measuring prevalence and risk of sexual victimisation in psychiatric populations. There appears to be a lack of consensus on: how sexual violence is defined, which and how participant samples are selected, how dynamic risk factors are accounted for in understanding sexual abuse types and mental health presentations, and how both past and lifetime prevalence/risks are measured. Joining other researchers calling for consistency, Kaul et al. (2024) note the “need for a comprehensive and consistent measurement framework for sexual violence to enable reliable and comparable prevalence data to be collected”.
This consistency should also translate to practitioners and services. Data sources and measurement tools are primary concerns, which impact actual practice and support provided. As Kaul and colleagues point out, clinicians do not routinely ask about sexual victimisation and definitions of violence are country- and culture-dependent and prone to personal biases. Sexual victimisation is linked to a range of negative psychosocial and economic consequences and this review found substantially higher prevalence and risk of victimisation for people already attending mental health services. This underscores the need for accurate and ethical screening to become routine, which would allow for better, patient-tailored, and more effective support.
Finally, there was a high variation of reported sexual violence in the male sample (e.g., 0% to over 20%). This speaks of the higher under-reporting of men’s sexual victimisation, as well as the methodological limitations of research studies not routinely disaggregating data by sex. Research and practice should note overall sex discrepancies in prevalence/risk as well as within and across different psychiatric settings if accurate support strategies and comprehensive understanding is to be achieved.
Research and practice should consistently and comprehensively measure the prevalence and risk of sexual victimisation.
Statement of interests
Ioana Crivatu has no conflicts of interest to declare.
Editor
Edited by Laura Hemming.
Links
Primary paper
Anjuli Kaul, Laura Connell-Jones, Sharli Anne Paphitis & Sian Oram. (2024). Prevalence and risk of sexual violence victimization among mental health service users: a systematic review and meta-analyses. Social Psychiatry and Psychiatric Epidemiology, 59(8), 1285-1297.
Other references
Campbell, R., Dworkin, E., & Cabral, G. (2009). An ecological model of the impact of sexual assault on women’s mental health. Trauma, Violence, & Abuse, 10(3), 225-246.
Care Quality Commission. (2018). Sexual safety on mental health wards.
Dworkin, E. R., Menon, S. V., Bystrynski, J., & Allen, N. E. (2017). Sexual assault victimization and psychopathology: A review and meta-analysis. Clinical Psychology Review, 56, 65-81.
Khalifeh, H., Oram, S., Osborn, D., Howard, L. M., & Johnson, S. (2016). Recent physical and sexual violence against adults with severe mental illness: a systematic review and meta-analysis. International Review of Psychiatry, 28(5), 433-451.
Lovell, A., Majeed-Ariss, R., & White, C. (2021). Repeat attenders are disproportionately vulnerable: An exploration of revictimisation at Saint Mary’s sexual assault referral Centre. Journal of Forensic and Legal medicine, 80, 102158.
Maniglio, R. (2009). Severe mental illness and criminal victimization: a systematic review. Acta Psychiatrica Scandinavica, 119(3), 180-191.
Miles, L., Valentine, J. L., Mabey, L., & Downing, N. R. (2022). Mental illness as a vulnerability for sexual assault: A retrospective study of 7,455 sexual assault forensic medical examinations. Journal of Forensic Nursing, 18(3), 131-138.
Tewksbury, R. (2007). Physical, mental and sexual consequences. International Journal of Men’s Health, 6(1), 22-35.
The World Health Organization. (2012). Sexual violence.
Wieberneit, M., Thal, S., Clare, J., Notebaert, L., & Tubex, H. (2024). Silenced survivors: A systematic review of the barriers to reporting, investigating, prosecuting, and sentencing of adult female rape and sexual assault. Trauma, Violence, & Abuse, 25(5), 3742-3757.
