Skip to main content
NCBI home page
eClinicalMedicine logoLink to eClinicalMedicine
. 2022 Jul 22;51:101560. doi: 10.1016/j.eclinm.2022.101560

Preventing suicidal and self-Injurious behavior in correctional facilities: A systematic literature review and meta-analysis

Stefan Stijelja a, Brian L Mishara b,⁎
PMCID: PMC9309412  PMID: 35898320

Summary

Background

Rates of suicide and self-harm are elevated in carceral institutions. Inmates are a vulnerable group since they are exposed to multiple risk factors. This paper critically reviews empirical research on programs to prevent suicidal and self-harmful behaviors in correctional facilities and summarizes effect sizes across studies.

Methods

We searched PsychINFO, PubMed, IEEEXPLORE and the CRISE Documentation Centre Database to identify relevant articles published before June 2022. Inclusion criteria were: peer-reviewed and with outcome data on effectiveness of prevention activities. Two reviewers independently assessed 905 articles to determine inclusion eligibility. Quality was assessed by two independent reviewers using the Quality Assessment Tool for Quantitative Studies. Meta-analyses using random-effect models were used to pool effect sizes for each outcome. This review was conducted in accordance with PRISMA guidelines.

Findings

Twenty-four of the 905 articles, published between 1980 and 2022, were included. Studies were frequently conducted in the United States (n=13; 54%) and used varying study designs; most frequently pre-post with no control group (n=9; 38%). Sample sizes and interventions varied considerably. Most were of moderate quality (n=21; 88%). On average, prevention programs in correctional facilities were effective in decreasing suicide deaths (pooled rate ratio of 0·35 [95% CI 0·23 to 0·55; p<0·001]; I2=68·01%), incidents of self-harm (pooled Hedges'g of -0·54 (95% CI: -1·03 to -0·05; p=0·031]; I2=81·34%), and suicidal ideation (pooled Hedges'g of -0·39 [95% CI: -0·65 to -0·14; p=0·003]; I2=47·09%).

Interpretation

Prevention activities are effective in reducing suicide death, self-harm and suicidal ideation in correctional settings. Multicomponent programs, which include several preventive activities, seem to be most effective in reducing suicide deaths. Future evaluation studies should control for confounding variables by including control groups, having larger samples and limiting attrition. Standards for suicide prevention in jails and prisons should be included in National suicide prevention strategies.

Funding

None.

Keywords: Prison, Inmates, Suicide, Self-harm, Prevention, Intervention, Best practice

Research in context.

Evidence before this study

Suicide is a leading cause of deaths in prisons and the rate of self-harm is much higher than in the general population. There has been limited research that evaluated programmatic interventions to prevent suicide and self-harm in prisons, and none used meta-analyses to summarize intervention effects across studies. A recent review of behavioural health interventions assessed the efficacy of psychotherapies with suicidal and self-harming individuals in prisons. It retained only six articles with empirical data. The last literature review of comprehensive suicide prevention programs and interventions identified twelve English-language articles, published from 1990 to 2012.

Added value of this study

We conducted an exhaustive systematic search for all empirical publications including research, identified 905 studies and retained 24 which provided data on effects on the three outcomes of suicide, non-fatal suicidal behaviors and suicidal ideation. We were able to conclude from the meta-analyses that prison suicide prevention programs were effective in decreasing suicide death, incidents of self-harm and suicidal ideation. We identified that the multi-component programs that include several prevention activities appear to be most effective in reducing suicide deaths.

Implications of all the available evidence

There is a need for more evaluation studies that control for confounding variables, have larger sample sizes and include comparable control groups. Nevertheless, the evidence from our analyses suggest that the implementation of suicide prevention programs in correctional facilities, particularly programs which include a variety of complementary actions and services, can save lives and reduce the incidence of non-fatal self-injurious behaviors. These findings add arguments for implementing suicide prevention programs in all correctional facilities by indicating that they can save lives and reduce the negative impacts of suicidal behaviors and ideation of prisoners, workers in correctional facilities and society at large.

Alt-text: Unlabelled box

Introduction

Suicide is a leading cause of death in prisons,1 and correctional facility suicide rates are higher than the general population.2 For example, in Austria the suicide rate for male inmates (130·8 per 100,000 population) is 3·5 times higher than the general population3; the UK female inmate suicide rate (83 per 100,000 population) is 20 times higher than the general population.4 Self-harm rates are also high in prisons, with an annual prevalence of self-harm in England and Wales of 5-6% for males and 20-24% for females.5 The human and financial costs have stimulated efforts to reduce suicide mortality and self-harm among inmates.6 The World Health Organization (2014) recommends that prison suicide prevention be included in national suicide prevention strategies.

Inmates have numerous risk factors for suicide and self-harm. The prevalence of mental health disorders in prisoners is high in both male and female: they have more psychotic disorders (4% for males; 4% for females), depression (10% for males; 12% for females), personality disorders (65% for males; 42% for females), alcohol and drug misuse/dependence (18-30% for males; 10-24% for females) and post-traumatic stress disorder (4-21% for males; 10-21% for females) than the general population.1,7, 8, 9 Prisoners are also exposed to unique risk factors associated with their incarceration, such as single-cell occupancy, hopelessness from serving a life sentence,10 and the institutional environment (e.g. overcrowded conditions, isolation protocols and risk of violence).11

There has been limited research that evaluated interventions to prevent suicide and self-harm in prisons, and none used meta-analyses to summarize intervention effects across studies. Winicov12 recently published a review which presented evidence from 6 studies of behavioral health interventions with suicidal and self-harming individuals in prisons, The last literature review on all types of suicide prevention programs and interventions (e.g., multifaceted programs) identified only twelve English-language articles, published from 1990 to 2012.11 It concluded that suicide prevention programs that addressed multiple risk factors appeared most effective in correctional settings. This paper updates our knowledge of evidence-based prevention in correctional facilities by presenting a systematic review, meta-analysis of results and a critical analysis of empirical research on programs to prevent suicidal and self-harmful behaviors in correctional facilities.

Methodology

The PRISMA guidelines13 for reporting of systematic reviews were followed. This research received no grants from any funding agency in the public, commercial, or not-for-profit sectors. The protocol for this review has not been previously published or registered.

Data source and search strategy

We searched PsychINFO, PubMed, IEEEXPLORE and the CRISE Documentation Centre Database, which contains an important body of studies on suicide and its prevention. IEEEXPLORE was included to identify the growing literature on the identification and prevention of suicide attempts through innovative technologies, such as image recognition techniques (e.g. a real-time recognition of suicidal behavior using an RGB-D camera;14). Searches were conducted in June 2022. As recommended by good practices in information science15,16 syntax specific to each database was developed using natural or controlled language. Supplementary Table S1 contains the syntax for PubMed. The other syntaxes are available upon request.

Study selection

The inclusion criteria were: 1) participants in a correctional facility (e.g., jail, prison or juvenile detection), 2) written in French or English; 3) no age or gender restrictions 4) published in a peer-reviewed journal; 5) includes an empirical analysis of the effectiveness of the prevention activities; 6) all study designs, except for review articles, were considered for inclusion. All papers published before 2022 were considered.

Outcomes of interest

In this review, suicide is defined as intentional self-inflicted death.17 Although self- harm can be classified as either with suicidal intent (i.e. suicide attempt) and non- suicidal, we adopted the broad definition commonly used in correctional institutions and previous literature reviews12: “any act where a prisoner deliberately harms themselves irrespective of the method, intent or severity of any injury.” Suicidal ideation in inmates was defined as the presence of thoughts about engaging in suicidal behaviours, as assessed by suicide risk assessment instruments or on the basis of placing of prisoners on suicide watch (i.e. intensive monitoring). We included all programs and interventions in carceral institutions with the expressed objective of decreasing the incidence of suicide, self-harm behaviors and suicidal ideation.

Data extraction and quality assessment

905 unique articles were identified and screened for eligibility based upon their title and abstract by the first author (SS) and the Director of the CRISE Documentation Centre (Luc Dargis), who are both experienced in screening and assessing empirical publications in previously published review articles. Both researchers screened the title and abstract and applied the inclusion and exclusion criteria. Inclusion was based upon consensus. If after discussion consensus was not achieved, the decision was adjucated by the second author (BLM). The main reasons for exclusion was lack of information on intervention effectiveness (i.e. an intervention or prevention strategy was described without evaluating its impact) and because articles did not meet inclusion criteria. SS read the full text of each study and assessed their eligibility.

The criteria for data extraction were adapted from the Cochrane Handbook for Systematic Reviews of Interventions.18 SS extracted the data on study setting, sample size, type of study design, nature of the intervention, outcome measures, and results.

Methodological quality was assessed using the Quality Assessment Tool for Quantitative Studies developed by the Effective Public Health Practice Project. The tool is psychometrically sound,19 and has been previously used in suicide research.20 SS and a skilled doctoral level research assistant independently assessed each of the following quality indicators as weak (1 point), moderate (2 points) or strong (3 points): Selection Bias, Study Design, Confounders, Blinding, Data Collection Methods and Withdrawal and Drop-Outs. Discrepancies were discussed and resolved. Domain scores were averaged to provide the total score: weak (1·00–1·50), moderate (1·51–2·50) or strong (2·51–3·00).21

Data analysis

Meta-analyses were conducted using Comprehensive Meta-Analysis (CMA), version 3 (Biostat, Englewood, NJ, USA). Analyses were conducted separately for suicidal ideation, self-harm and suicide deaths. Random-effect models were used since true effect sizes were likely to vary between studies.22 For data on suicidal ideation and self-harm, Hedges’ g was used to measure the effect size. Hedges’ g is a measure of standardized mean differences that corrects for small-sample bias (n ≤ 20).23 The interpretation is similar to Cohen's d, where an effect size of 0·2 is considered to be small, 0·5 is considered to be moderate and 0·8 is considered to be large.24 For studies where the main outcome was the suicide rate per 100,000 population per year, rate ratios were computed and pooled. When studies reported more than one outcome (e.g., suicidal ideation as measured by the Suicide Probability Scale and the Beck Scale for Suicidal Ideation) we instructed CMA to compute the mean of the outcomes for each study, and use this synthetic score as the unit of analysis.25 This approach prevents meta-analyses from underestimating the standard error of the summary effect. In the presence of zero events (e.g., no suicide deaths registered at post test), we used the continuity correction method (i.e. adding 0.5;26) to compute standard errors and the summary effect. Heterogeneity was assessed using Cochran's Q test and the I2 statistic (i.e. percentage of variability in treatment effect estimates that is due to between study heterogeneity rather than chance).27 Conventionally, I2 statistic of 25% denotes low heterogeneity; 50% moderate heterogeneity; and 75% high heterogeneity. As recommended in the Cochrane Handbook for Systematic Reviews of Interventions,18 the significance level for heterogeneity was defined as p<0.10. We generated and visually assessed funnel plots for presence of publication bias.28 Egger's regression test for asymmetry was also calculated; a significant test (p<0·1) suggesting publication bias.29 In the presence of publication bias, the Duval and Tweedie trim and fill method was used to estimate the number of missing studies and find the point estimate of the corrected overall effect size.30

Role of the funding source

There was no funding source for this study. SS and BLM had access to the data and jointly decided to submit this manuscript for publication.

Results

Study characteristics

Figure 1 presents the PRISMA flow diagram. We included 24 articles that reported on the effectiveness of 24 different interventions and prevention programmes. Of the 24, thirteen were in the United States, seven in the UK, two in Australia, one in Canada and one in Israel (see Table 1). Sample sizes varied considerably, from 10 inmates31 to an entire county jail population (e.g. 7500 inmates;32). Some studies had only men or women and others did not disclose their gender composition. The mean age was not consistently reported. The most common study design was pre-post with no control group (n=9), followed by time series study (n=5), randomized pre-post with control group (n=4), and retrospective cohort study (n=4). One study used a correlational design and another used a post-test only design. Ten studies evaluated the efficacy of psychotherapy.33, 34, 35, 36, 37, 38, 39, 40, 41, 42 Six studies analyzed comprehensive multicomponent suicide prevention strategies.6,32,43, 44, 45, 46 Two studies reported on the efficacy of peer support schemes.47,48 Two studies examined the effects of pharmacotherapy, clozapine31 and opioid substitution therapy.49 The other four studies were an evaluation of the efficacy of an Aboriginal art program,50 an assessment of the impact of an increased accessibility to a psychiatrist,51 examination of the effects of implementing therapeutic activities as alternatives to solitary confinement,52 and a study of the impact of general improvements in the jail environment on suicide mortality.53

Figure 1.

Figure 1

Open in a new tab

PRISMA flow diagram of the identification, screening, eligibility assessment, and inclusion of articles.

Table 1.

Suicide and self-harm prevention programs and interventions with empirical outcome data.

Image, table 1
Image, table 1
Image, table 1
Image, table 1
Image, table 1
Image, table 1
Open in a new tab

Most studies were of moderate quality (n=21), two were of weak quality (n=2) and only one was considered strong (n=1); Table 2. The methodological quality of studies varied considerably across quality indicators. No studies exhibited strong methodological features across the entire range of criteria. However, evaluation studies of multicomponent prevention strategies6,32,43, 44, 45, 46 exhibited high ratings on selection bias since their programs targeted entire prison or jail populations. Similarly, data collection methods were generally strong or moderate for all studies, as they all used psychometrically sound instruments or administrative data. Weak ratings were frequently recorded for study design and confounders. Most studies did not control for relevant confounders (e.g. age, length of incarceration, race), and several did not include a control group, and few studies randomly allocated their participants to experimental and control conditions.

Table 2.

Methodological quality assessment of included studies.

Image, table 2
Open in a new tab

Several studies reported various issues related to conducting an evaluation study in prison settings.33,36,38,44,47 These issues included the frequent and unpredictable transfer or release of prisoners that can increase the rate of attrition.33 Furthermore, the nature of correctional facilities can prevent researchers from carrying out randomization.47

Meta-analysis

Two studies were not pooled because their effect size could not be computed with the available information. In Hayes54 there was no comparison group or pre-intervention suicide rate, and data available in Tartaro & Levy53 were not amenable to the calculation of a rate ratio: the study had a Hedges’ g of -0·24, suggesting a low effect size.

Suicide deaths

Figure 2 presents the forest plot for the analysis of the effect of prison-based suicide and self-harm prevention programs on suicide deaths. The pooled rate ratio was 0·35 (95% CI 0·23 to 0·55; p<0·001), suggesting that the post-intervention group had 0·35 times the rate of suicide death compared to the pre-intervention group (i.e., a relative rate reduction of 65%). Heterogeneity was moderate and significant (Q-value = 18·76; d=6; p-value =0·005; I2=68·01%).

Figure 2.

Figure 2

Open in a new tab

Rate ratios [95% confidence intervals] for individual studies (squares and bars) and the pooled rate ratio [95% CI] using random effects model (red diamond). A 2-tailed p value of <.05 was considered statistically significant.

Self-harm

Figure 3 presents the forest plot for the meta-analysis of self-harm as the outcome. Panel A presents studies whose effect sizes were pooled using Hedges'g as the effect size index. Bursac et al.40 and Glowa-Kollisch et al.52 reported results that were not amenable to the calculation of a Hedges'g, their rate ratios were pooled (see Figure 3 Panel B). Results reported by Iancu et al.51 were excluded from our analyses since the setting (i.e. military prison) in which the program was implemented may not be comparable to the settings in other studies. The study population was imprisoned soldiers under obligatory service who could be released from service by obtaining a recommendation for discharge from a military psychiatrist. Thus, the observed increase in self-injurious behaviors following policy change may have been caused by an underlying increase in self-injurious behaviours that were possibly initiated in hopes of being discharged.51

Figure 3.

Figure 3

Open in a new tab

Panel A.

Hedges's g [95% confidence intervals] for individual studies (squares and bars) and the pooled Hedges's g [95% CI] using random effects model (red diamond). A 2-tailed p value of <.05 was considered statistically significant.

Panel B.

Rate ratios [95% confidence intervals] for individual studies (squares and bars) and the pooled rate ratio [95% CI] using random effects model (red diamond). A 2-tailed p value of <.05 was considered statistically significant.

In Figure 3 Panel A, the pooled Hedges'g was -0·54 (95% CI: -1·03 to -0·05; p=0·031), suggesting that the studies were moderately effective in reducing self-harm among inmates. Heterogeneity was high and significant (Q-value = 26·80; d=5; p-value =0·000; I2=81·34%). Similarly, in Figure 3 Panel B, programs were significantly associated with decreases in the incidence of self-harm (pooled rate ratio: 0·55 (95% CI: 0·36 to 0·83; p=0·005). Heterogeneity was moderate and significant (Q-value = 3·12; d=1; p-value =0·077; I2=67·93%).

Suicidal ideation

Figure 4 presents the forest plot for the meta-analysis on suicidal ideation as the outcome. The pooled Hedges’ g was -0·39 (95% CI: -0·65 to -0·14; p=0·002). On average, programs decreased suicidal thoughts by 0·39 standard deviations, as compared with inmates who did not receive an intervention. Heterogeneity was near moderate and significant (Q-value = 11·34; d=6; p-value =0·078; I2=47·09%).

Figure 4.

Figure 4

Open in a new tab

Hedges's g [95% confidence intervals] for individual studies (squares and bars) and the pooled Hedges's g [95% CI] using random effects model (red diamond). A 2-tailed p value of <.05 was considered statistically significant.

For suicide deaths, the Egger's test (p=0·248) was not significant, suggesting an absence of publication bias. A visual assessment of the funnel plot (Supplementary Figure S2) showed that two studies were missing to the right of the mean, according to the trim and fill method. The imputed pooled rate ratio was 0·38 (95% CI: 0·025 to 0·60), instead of 0·35 (95% CI 0·23 to 0·55), indicating that the validity of our results does not appear to be threatened by publication bias.

For self-harm and suicidal ideation, the Egger's test (p=0·498 and p=0·901, respectively) was not significant and the funnel plot (Supplementary Figures S3 and S4) was reasonably symmetrical, suggesting no publication bias.

For each outcome, caution should be exercised when interpreting results from Egger's test, since it is usually underpowered when used with less than 10 studies.28 We did not consider conducting a sensitivity analysis, excluding articles based on study quality, since the number of weak studies was small (n=2) and one study Tartaro & Levy53 was already excluded due to other reasons. We have not conducted subgroup analyses by country since the combination of outcome country-intervention would have yielded small meta-analyses of only 2-3 studies, which would have not provided helpful recommendations to policy makers or suicide prevention programme developers.

Discussion

This review found evidence supporting the efficacy of suicide and self-harm prevention programs in correctional settings. On average, the incidence of suicide, self-harm behaviours and suicidal ideation all decreased significantly following their implementation.

Concerning suicide deaths, five out of the seven pooled studies proposed multicomponent suicide prevention strategies. This is in line with the understanding of suicide as a multifaceted phenomenon. This suggests that programs seeking to reduce the burden of suicide in carceral institutions should seek to address multiple risk factors by including several strategies with robust evidence of their effectiveness, such as: 1) screening of inmates, 2) staff training in CPR and in crisis-intervention, 3) supervision of high-risk inmates, 4) proper communication between staff and inmates, 5) post-suicide administrative reviews, 6) staff debriefing, 7) improved clinical procedures, 8) ameliorated process for reviewing suicides, 9) restricted access to means and 10) provision of mental health treatment and support to inmates.32,43,45,46,55 However, in a national study of jail suicides, Hayes56 reported that while most facilities had a written suicide prevention policy, only 20% had written policies encompassing all the components of a suicide prevention program.

The reduction in self-harm is also notable, considering that self-harm in prison is associated with the risk of a subsequent suicide in this setting.5 However, an important limitation is the inconsistent and ambiguous definitions of self-harm which often includes both non-suicidal self-injury (NSSI) and intentional self-harm. Future studies should differentiate the two behaviors, which may have different aetiologies and could warrant different prevention strategies. In several of the studies that used suicidal ideation as the outcome variable, it is not clear if the thoughts about suicide were “serious” considerations of ending one's life, or thoughts about suicide with little actual risk of leading to a suicide attempt. Future research should be careful to use validated scales, ask about “serious” consideration of suicide and also include suicide risk assessments. Furthermore, studies included in this review had relatively small sample sizes. Larger studies are warranted to insure the robustness of the reported intervention effects. The same remark about sample sizes applies to studies with suicidal ideation as the outcome. Larney et al.49 is the only identified study to have evaluated an intervention program targeting drug or alcohol dependency among inmates. This is concerning, since alcohol use problems and a history of illicit drug use have been associated with an increased risk of suicide and self-harm in prison.10,57 Substance use disorder is highly prevalent among prisoners,8 and previous studies reported that this factor can successfully distinguish attempters from ideators in prison environments.58, 59, 60 Interventions tackling substance use should be an important component of suicide prevention strategies in correctional settings.

Only one study evaluated a culturally sensitive prevention program.50 The paucity of studies evaluating the effectiveness of culture- and gender-specific interventions is unfortunate, notably because Indigenous (Aboriginal) prisoners have a higher prevalence of suicide attempts than non-Indigenous prisoners.61 Similarly, male and female offenders are not exposed to suicide risk factors in the same proportion. Females offenders are more likely than men to report a lifetime history of suicidal ideation and attempts,59 have higher rates of borderline personality disorders,62 and more frequently have extensive childhood trauma histories.63 Tailored interventions may be called for to effectively address the needs of different groups of inmates.

With technological advancement, we anticipated that new surveillance tools will become available to prevent suicide in prisons and jails. However, while promising simulation studies have been published (e.g.14,64, 65, 66), none have been reported to be implemented in a real prison environment (i.e., with real prisoners). These technological advancements may one day help assist with the monitoring of suicidal inmates. However, they do not address the causes of suicidal behaviors, nor do they diminish risk factors or increase protective factors associated with suicide in carceral settings.

An emerging literature on the health of asylum seekers, who are sometimes in detention centres that resemble prison environments, suggests that those who are detained, across all types of arrangements (i.e. community-based arrangements, community detention, onshore and offshore detention), exhibit exceptionally high rates of self-harm. In the Australian asylum seeker population in 2015, self-harm rates were incredibly high: 260 per 1000 among asylum seekers placed in offshore detention.67 The extent to which programmes designed for persons who were condemned to prison for crimes may be pertinent for preventing suicidal behaviours with asylum seekers in detention has not been evaluated. The lack of interventions to address this critical issue is troubling.

Finally, aligning ourselves with authors who extensively studied suicide in correctional facilities,1,66 we believe that standards for suicide prevention in jails and prisons should be included in National suicide prevention strategies.

While we considered articles published in English and French, we may have disregarded studies published in other languages. Since we restricted our review to papers published in a peer-reviewed journal, we may have omitted evaluation studies found in the grey literature. Furthermore, while prison suicide rates vary considerably across time and country,67 suicide rates reported in some of the studies we included seemed relatively high before the intervention. It may be that some interventions were implemented in response to a cluster of suicide events in a particular setting. These high pre-intervention rates may have increased the effect size of the impact of the intervention. Without a control group, we have no way of knowing if the high pre-intervention rates would have decreased without the intervention, or if the observed effects are a result of regression to the mean, following a suicide cluster with an unusually high incidence of suicidal behaviors. Another hypothesis is that correctional settings can exercise greater control over their populations, thus facilitating the implementation of programmes and interventions. Studies included in the review were predominantly conducted in the United States and the United Kingdom. This may limit generalization to Low- and Middle-Income Countries and non-Western cultural settings. Finally, only one person (SS) extracted the data: this could have resulted in a greater risk of errors.

Results from meta-analyses indicate that programs significantly decreased suicide deaths, self-harm and suicidal ideation among inmates. However, variance in the effect sizes was moderate to high (I2 between 46·02% and 81·34%). This suggests that, depending on the outcome, 46% to 81% of the variability in effects is due to study differences (heterogeneity) and 19% to 54% is due to chance. Heterogeneity in program effects can be explained by differences in study populations (e.g. gender), the setting (e.g. size and nature of prison populations), duration of follow-up, the nature and dosage of the interventions received, and other unaccounted factors.22 While inconsistency is concerning and may limit the interpretation of the effect size to some extent, it is reassuring that at least the direction of the effect from each of the individual studies favored the intervention. Our small body of studies precluded the use of meta-regression or subgroup analyses to identify potential causes of heterogeneity. Future studies should try to elucidate which specific program or interventions works best for which subpopulations of inmates, and in what contexts.

Suicide and self-harm prevention present important challenges for correctional institutions. Our meta-analyses of the impact of suicide and self-harm prevention programs in correction settings suggest that programs can be effective in reducing the incidence of deaths by suicide, as well as the incidence of self-harm and suicidal ideation. Multicomponent programs seem to be most effective in reducing suicide deaths. However, our findings also highlight important methodological deficiencies in evaluation studies of suicide prevention programmes in correction facilities. Future studies should aspire to control for confounding factors by including control groups, using larger samples and limiting the attrition of participants. Standards for suicide prevention in jails and prisons should be included in National suicide prevention strategies.

Contributors

Both authors contributed substantially to the design of the work, interpretation of the data, intellectual content and writing of the manuscript.

Data sharing statement

All extracted and calculated data are available upon appropriate requests by emailing to the authors.

Declaration of interests

The authors have no conflict of interest to declare.

Acknowledgements

The authors thank Luc Dargis for his immense help in the bibliographical search. The authors would also like to thank the anonymous reviewers for their insightful comments which helped to improve the manuscript.

Footnotes

Supplementary material associated with this article can be found in the online version at doi:10.1016/j.eclinm.2022.101560.

Appendix. Supplementary materials

mmc1.pdf (254.6KB, pdf)

References

  • 1.Fazel S, Baillargeon J. The health of prisoners. Lancet. 2011;377(9769):956–965. doi: 10.1016/S0140-6736(10)61053-7. [DOI] [PubMed] [Google Scholar]
  • 2.Eck M, Scouflaire T, Debien C, et al. [Suicide in prison: epidemiology and prevention] Presse Med. 2019;48(1 Pt 1):46–54. doi: 10.1016/j.lpm.2018.11.009. [DOI] [PubMed] [Google Scholar]
  • 3.Fruehwald S, Frottier P, Eher R, Gutierrez K, Ritter K. Prison suicides in Austria, 1975-1997. Suicide Life Threat Behav. 2000;30(4):360–369. [PubMed] [Google Scholar]
  • 4.Fazel S, Benning R, Danesh J. Suicides in male prisoners in England and Wales, 1978-2003. Lancet. 2005;366(9493):1301–1302. doi: 10.1016/S0140-6736(05)67325-4. [DOI] [PubMed] [Google Scholar]
  • 5.Hawton K, Linsell L, Adeniji T, Sariaslan A, Fazel S. Self-harm in prisons in England and Wales: an epidemiological study of prevalence, risk factors, clustering, and subsequent suicide. Lancet. 2014;383(9923):1147–1154. doi: 10.1016/S0140-6736(13)62118-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Hayes LM. Prison suicide: an overview and guide to prevention The prison. journal. 1995;75(4):431–456. [Google Scholar]
  • 7.Fazel S, Danesh J. Serious mental disorder in 23 000 prisoners: a systematic review of 62 surveys. Lancet North Am Ed. 2002;359(9306):545–550. doi: 10.1016/S0140-6736(02)07740-1. [DOI] [PubMed] [Google Scholar]
  • 8.Fazel S, Bains P, Doll H. Substance abuse and dependence in prisoners: a systematic review. Addiction. 2006;101(2):181–191. doi: 10.1111/j.1360-0443.2006.01316.x. [DOI] [PubMed] [Google Scholar]
  • 9.Goff A, Rose E, Rose S, Purves D. Does PTSD occur in sentenced prison populations? A systematic literature review. Crim Behav Ment Health. 2007;17(3):152–162. doi: 10.1002/cbm.653. [DOI] [PubMed] [Google Scholar]
  • 10.Winicov N. A systematic review of behavioral health interventions for suicidal and self-harming individuals in prisons and jails. Heliyon. 2019;5(9):e02379. doi: 10.1016/j.heliyon.2019.e02379. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Barker E, Kõlves K, De Leo D. Management of suicidal and self-harming behaviors in prisons: systematic literature review of evidence-based activities. Arch Suicide Res. 2014;18(3):227–240. doi: 10.1080/13811118.2013.824830. [DOI] [PubMed] [Google Scholar]
  • 12.BL Mishara, M Tousignant. Comprendre le suicide, 2019, Presses de l'Université de Montréal; 2004. Montreal, Canada. Available from: http://www.deslibris.ca/ID/406470.
  • 13.Li B, Bouachir W, Gouiaa R, Noumeir R, editors. 2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA) 2017. Real-time recognition of suicidal behavior using an RGB-D camera. 28 Nov.-1 Dec. 2017. [Google Scholar]
  • 14.Bartels EM. How to perform a systematic search. Best Pract Res Clin Rheumatol. 2013;27(2):295–306. doi: 10.1016/j.berh.2013.02.001. [DOI] [PubMed] [Google Scholar]
  • 15.Shamseer L, Moher D, Clarke M, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ. 2015;350:g7647. doi: 10.1136/bmj.g7647. [DOI] [PubMed] [Google Scholar]
  • 16.Li T, Higgins JP, Deeks JJ. John Wiley & Sons; Chichester, UK: 2019. Collecting Data. Cochrane Handbook for Systematic Reviews of Interventions; pp. 109–141. [Google Scholar]
  • 17.Thomas BH, Ciliska D, Dobbins M, Micucci S. A process for systematically reviewing the literature: providing the research evidence for public health nursing interventions. Worldviews Evid Based Nurs. 2004;1(3):176–184. doi: 10.1111/j.1524-475X.2004.04006.x. [DOI] [PubMed] [Google Scholar]
  • 18.Clifford AC, Doran CM, Tsey K. A systematic review of suicide prevention interventions targeting indigenous peoples in Australia, United States, Canada and New Zealand. BMC Public Health. 2013;13(1):463. doi: 10.1186/1471-2458-13-463. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Armijo-Olivo S, Stiles CR, Hagen NA, Biondo PD, Cummings GG. Assessment of study quality for systematic reviews: a comparison of the Cochrane Collaboration Risk of Bias Tool and the Effective Public Health Practice Project Quality Assessment Tool: methodological research. J Eval Clin Pract. 2012;18(1):12–18. doi: 10.1111/j.1365-2753.2010.01516.x. [DOI] [PubMed] [Google Scholar]
  • 20.Riley RD, Higgins JPT, Deeks JJ. Interpretation of random effects meta-analyses. BMJ. 2011;342:d549. doi: 10.1136/bmj.d549. [DOI] [PubMed] [Google Scholar]
  • 21.Hedges LV. Distribution theory for Glass’s estimator of effect size and related estimators. J Educ Stat. 1981;6:107–128. [Google Scholar]
  • 22.Cohen J. L. Erlbaum Associates; Hillsdale, N.J.: 1988. Statistical Power Analysis for the Behavioral Sciences. [Google Scholar]
  • 23.Borenstein M, Hedges LV, Higgins JPT, Rothstein HR. Multiple outcomes or time-points within a study. Introduction to meta-analysis. John Wiley & Sons, Ltd; Chichester, UK: 2009. pp. 225–238. [Google Scholar]
  • 24.Sweeting MJ, Sutton AJ, Lambert PC. What to add to nothing? Use and avoidance of continuity corrections in meta-analysis of sparse data. Stat Med. 2004;23(9):1351–1375. doi: 10.1002/sim.1761. [DOI] [PubMed] [Google Scholar]
  • 25.Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;327(7414):557–560. doi: 10.1136/bmj.327.7414.557. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Sterne JAC, Sutton AJ, Ioannidis JPA, et al. Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials. BMJ. 2011;343:d4002. doi: 10.1136/bmj.d4002. [DOI] [PubMed] [Google Scholar]
  • 27.Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315(7109):629–634. doi: 10.1136/bmj.315.7109.629. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Duval S, Tweedie R. Trim and Fill: a simple funnel-plot–based method of testing and adjusting for publication bias in meta-analysis. Biometrics. 2000;56(2):455–463. doi: 10.1111/j.0006-341x.2000.00455.x. [DOI] [PubMed] [Google Scholar]
  • 29.Zarzar TR, Catlett TL, O'Connell MG, et al. Clozapine reduces self-injurious behavior in a state prison population. J Am Acad Psychiatry Law. 2019;47(1):61–67. doi: 10.29158/JAAPL.003818-19. [DOI] [PubMed] [Google Scholar]
  • 30.Cox JF, Morschauser PC. A solution to the problem of jail suicide. Crisis. 1997;18(4):178–184. doi: 10.1027/0227-5910.18.4.178. [DOI] [PubMed] [Google Scholar]
  • 31.Black DW, Blum N, McCormick B, Allen J. Systems Training for Emotional Predictability and Problem Solving (STEPPS) group treatment for offenders with borderline personality disorder. J Nerv Ment Dis. 2013;201(2):124–129. doi: 10.1097/NMD.0b013e31827f6435. [DOI] [PubMed] [Google Scholar]
  • 32.Long CG, Fulton B, Dolley O, Hollin CR. Dealing with feelings: the effectiveness of cognitive behavioural group treatment for women in secure settings. Behav Cogn Psychother. 2011;39(2):243–247. doi: 10.1017/S1352465810000573. [DOI] [PubMed] [Google Scholar]
  • 33.Perry A, Waterman MG, House A, et al. Problem-solving training: assessing the feasibility and acceptability of delivering and evaluating a problem-solving training model for front-line prison staff and prisoners who self-harm. BMJ Open. 2019;9(10) doi: 10.1136/bmjopen-2018-026095. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Pratt D, Tarrier N, Dunn G, et al. Cognitive-behavioural suicide prevention for male prisoners: a pilot randomized controlled trial. Psychol Med. 2015;45(16):3441–3451. doi: 10.1017/S0033291715001348. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Rohde P, Jorgensen JS, Seeley JR, Mace DE. Pilot evaluation of the coping course: a cognitive-behavioral intervention to enhance coping skills in incarcerated youth. J Am Acad Child Adolesc Psychiatry. 2004;43(6):669–676. doi: 10.1097/01.chi.0000121068.29744.a5. [DOI] [PubMed] [Google Scholar]
  • 36.Walker T, Shaw J, Turpin C, Reid C, Abel K. The WORSHIP II study: a pilot of psychodynamic interpersonal therapy with women offenders who self-harm. J Forensic Psychiatry Psychol. 2017;28(2):158–171. [Google Scholar]
  • 37.Camp J, Joy K, Freestone M. Does “Enhanced support” for offenders effectively reduce custodial violence and disruption? An evaluation of the enhanced support service pilot. Int J Offender Ther Comp Criminol. 2018;62(12):3928–3946. doi: 10.1177/0306624X17752254. [DOI] [PubMed] [Google Scholar]
  • 38.Bursac R, Raffa L, Solimo A, Bell C, Ford E. Boundary-spanning care: reducing psychiatric rehospitalization and self-injury in a jail population. J Correct Health Care. 2018;24(4):365–370. doi: 10.1177/1078345818792055. [DOI] [PubMed] [Google Scholar]
  • 39.Biggam FH, Power KG. A controlled, problem-solving, group-based intervention with vulnerable incarcerated young offenders. Int J Offender Ther Comp Criminol. 2002;46(6):678–698. doi: 10.1177/0306624X02238162. [DOI] [PubMed] [Google Scholar]
  • 40.Perry AE, Waterman MG, Dale V, Moore K, House A. The effect of a peer-led problem-support mentor intervention on self-harm and violence in prison: an interrupted time series analysis using routinely collected prison data. EClinicalMedicine. 2021;32 doi: 10.1016/j.eclinm.2020.100702. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Felthous AR. Preventing jailhouse suicides. Bull Am Acad Psychiatry Law. 1994;22(4):477–488. [PubMed] [Google Scholar]
  • 42.Hastings GE, Vick L, Lee G, Sasmor L, Natiello TA, Sanders JH. Nurse practitioners in a jailhouse clinic. Med Care. 1980;18(7):731–744. doi: 10.1097/00005650-198007000-00003. [DOI] [PubMed] [Google Scholar]
  • 43.Hayes LM. From chaos to calm: one jail system's struggle with suicide prevention. Behav Sci Law. 1997;15(4):399–413. doi: 10.1002/(sici)1099-0798(199723/09)15:4<399::aid-bsl284>3.0.co;2-o. [DOI] [PubMed] [Google Scholar]
  • 44.Hayes LM. Prison suicide: an overview and guide to prevention (part 3) Crisis. 1995;16(2):63–65. doi: 10.1027/0227-5910.16.2.63. [DOI] [PubMed] [Google Scholar]
  • 45.Kovasznay B, Miraglia R, Beer R, Way B. Reducing suicides in New York State correctional facilities. Psychiatr Q. 2004;75(1):61–70. doi: 10.1023/b:psaq.0000007561.83444.a4. [DOI] [PubMed] [Google Scholar]
  • 46.Hall B, Gabor P. Peer suicide prevention in a prison. Crisis. 2004;25(1):19–26. doi: 10.1027/0227-5910.25.1.19. [DOI] [PubMed] [Google Scholar]
  • 47.Junker G, Beeler A, Bates J. Using trained inmate observers for suicide watch in a federal correctional setting: a win-win solution. Psychol Serv. 2005;2(1):20–27. [Google Scholar]
  • 48.Larney S, Gisev N, Farrell M, et al. Opioid substitution therapy as a strategy to reduce deaths in prison: retrospective cohort study. BMJ Open. 2014;4(4) doi: 10.1136/bmjopen-2013-004666. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Rasmussen MK, Donoghue DA, Sheehan NW. Suicide/self-harm-risk reducing effects of an Aboriginal art program for Aboriginal prisoners. Adv Ment Health. 2018;16(2):141–151. [Google Scholar]
  • 50.Iancu I, Bodner E, Sarel A, Einat H. Changes in mental health policy and their influence on self-injurious behaviours in the Israeli military prison system. Can J Psychiatry. 2007;52(9):591–597. doi: 10.1177/070674370705200907. [DOI] [PubMed] [Google Scholar]
  • 51.Glowa-Kollisch S, Kaba F, Waters A, Leung YJ, Ford E, Venters H. From punishment to treatment: the “clinical alternative to punitive segregation” (CAPS) program in New York City jails. Int J Environ Res Public Health. 2016;13(2):182–192. doi: 10.3390/ijerph13020182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Tartaro C, Levy M. Predictors of suicide in new generation jails. Justice Res Policy. 2008;10(1):21–37. [Google Scholar]
  • 53.Fazel S, Cartwright J, Norman-Nott A, Hawton K. Suicide in prisoners: a systematic review of risk factors. J Clin Psychiatry. 2008;69(11):1721–1731. [PubMed] [Google Scholar]
  • 54.Favril L, Yu R, Hawton K, Fazel S. Risk factors for self-harm in prison: a systematic review and meta-analysis. Lancet Psychiatry. 2020;7(8):682–691. doi: 10.1016/S2215-0366(20)30190-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Fazel S, Yoon IA, Hayes AJ. Substance use disorders in prisoners: an updated systematic review and meta-regression analysis in recently incarcerated men and women. Addiction. 2017;112(10):1725–1739. doi: 10.1111/add.13877. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Favril L, Indig D, Gear C, Wilhelm K. Mental disorders and risk of suicide attempt in prisoners. Soc Psychiatry Psychiatr Epidemiol. 2020;55(9):1–11. doi: 10.1007/s00127-020-01851-7. [DOI] [PubMed] [Google Scholar]
  • 57.Favril L, Stoliker B, Vander Laenen F. What differentiates prisoners who attempt suicide from those who experience suicidal ideation? A nationally representative study. Suicide Life-Threat Behav. 2020;50:975–989. doi: 10.1111/sltb.12638. [DOI] [PubMed] [Google Scholar]
  • 58.Favril L, O’Connor RC. Distinguishing prisoners who think about suicide from those who attempt suicide. Psychol Med. 2019;51(2):1–8. doi: 10.1017/S0033291719003118. [DOI] [PubMed] [Google Scholar]
  • 59.Larney S, Topp L, Indig D, O'driscoll C, Greenberg D. A cross-sectional survey of prevalence and correlates of suicidal ideation and suicide attempts among prisoners in New South Wales, Australia. BMC Public Health. 2012;12(1):1–7. doi: 10.1186/1471-2458-12-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Black DW, Gunter T, Allen J, et al. Borderline personality disorder in male and female offenders newly committed to prison. Compr Psychiatry. 2007;48(5):400–405. doi: 10.1016/j.comppsych.2007.04.006. [DOI] [PubMed] [Google Scholar]
  • 61.Clements-Nolle K, Wolden M, Bargmann-Losche J. Childhood trauma and risk for past and future suicide attempts among women in prison. Women's Health Issues. 2009;19(3):185–192. doi: 10.1016/j.whi.2009.02.002. [DOI] [PubMed] [Google Scholar]
  • 62.Bouachir W, Gouiaa R, Li B, Noumeir R. Intelligent video surveillance for real- time detection of suicide attempts. Pattern Recogn Lett. 2018;110:1–7. [Google Scholar]
  • 63.Bouachir W, Noumeir R, editors. 7th International Conference on Imaging for Crime Detection and Prevention, ICDP 2016. Institution of Engineering and Technology; 2016. Automated video surveillance for preventing suicide attempts. [Google Scholar]
  • 64.Lee S, Kim H, Lee S, et al. Detection of a suicide by hanging based on a 3-D image analysis. IEEE Sensors J. 2014;14(9):2934–2935. [Google Scholar]
  • 65.Hedrick K, Armstrong G, Coffey G, Borschmann R. Self-harm in the Australian asylum seeker population: a national records-based study. SSM Popul Health. 2019;8 doi: 10.1016/j.ssmph.2019.100452. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66.Favril L, Wittouck C, Audenaert K, Vander Laenen F. A 17-year national study of prison suicides in Belgium. Crisis. 2019;40(1):42–53. doi: 10.1027/0227-5910/a000531. [DOI] [PubMed] [Google Scholar]
  • 67.Fazel S, Ramesh T, Hawton K. Suicide in prisons: an international study of prevalence and contributory factors. Lancet Psychiatry. 2017;4(12):946–952. doi: 10.1016/S2215-0366(17)30430-3. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

mmc1.pdf (254.6KB, pdf)

Articles from eClinicalMedicine are provided here courtesy of Elsevier

RESOURCES