Prevalence of Posttraumatic Stress Disorder in Prisoners

Abstract

People involved with criminal justice frequently are exposed to violence and traumatic experiences. This may lead to posttraumatic stress disorder (PTSD); however, no review, to our knowledge, has synthetized findings in this setting. We conducted a systematic review and meta-analysis to estimate prevalence rates of PTSD in prison populations. Original studies in which prevalence rates of PTSD in unselected samples of incarcerated people were reported were systematically searched between 1980 and June 2017. Data were pooled using random-effects meta-analysis, and sources of heterogeneity for prespecified characteristics were assessed by meta-regression. We identified 56 samples comprising 21,099 imprisoned men and women from 20 countries. Point prevalence of PTSD ranged from 0.1% to 27% for male, and from 12% to 38% for female prisoner populations. The random-effects pooled point prevalence was 6.2% (95% confidence interval: 3.9, 9.0) in male prisoners and 21.1% (95% confidence interval: 16.9, 25.6) in female prisoners. The heterogeneity between the included studies was very high. Higher prevalence was reported in samples of female prisoners, smaller studies (n ≤ 200), and for investigations based in high-income countries. Existing evidence shows high levels of PTSD among imprisoned people, especially women. Psychosocial interventions to prevent violence, especially against children and women, and to mitigate its consequences in marginalized communities must be improved. Trauma-informed approaches for correctional programs and scalable PTSD treatments in prisons require further consideration.

Figure 1.

Flow diagram showing the different stages of finding relevant studies on the prevalence of posttraumatic stress disorder in prison populations between 1980 and 2017.

Figure 2.

Prevalence meta-analysis of point prevalence of posttraumatic stress disorder in male and female prisoners from studies published between 1980 and 2017. Samples are sorted by sex and high- versus low- or middle-income countries, as well as by ascending prevalence rates within the subgroups. Sample weights from random-effects meta-analyses may not sum to 100, because of rounding errors. Smaller samples (n < 100) were aggregated for male (20, 41, 47, 53, 61) and female subgroups (36, 39, 42, 50, 56, 61). CI, confidence interval; MQ, Martinique; UK, United Kingdom; US, United States.

Figure 3.

Prevalence meta-analysis of lifetime prevalence rates of posttraumatic stress disorder in male and female prisoners from studies published between 1980 and 2017. Samples are sorted by sex and non-US versus United States, as well as by ascending prevalence rates within the subgroups. Sample weights from random-effects meta-analysis may not sum to 100, because of rounding errors. Smaller female samples (n < 100) were aggregated (39, 56). CI, confidence interval; UK, United Kingdom; US, United States.