In this issue (page 1109),1 Wendy Wobeser and colleagues present an analysis of mortality rates and causes of death among people in involuntary confinement in Ontario from 1990 to 1999. A particular strength of this paper is its epidemiologic approach, which involved examining reports of deaths in federal penitentiaries, provincial prisons and police cells. It is worth noting that it is usually a challenge for researchers to get valid information from such a range of different public authorities. The authors found 308 inmate deaths, of which 291 are described in detail. Rates of death by poisoning, suicide, homicide and natural causes were compared with the rates among Canadian men in the general population aged 25–49 years. The striking results are that rates of death by both violent and natural causes in custody exceeded by far the rates in the general population. Death by accidental intoxication (drug overdose) was 20 and 50 times more common among those in provincial and federal custody, respectively, and suicide by strangulation was 10 and 4.5 times more frequent. Furthermore, an excess number of deaths from cardiovascular disease was found as well, affecting a disproportionate number of young people.
Both illicit drug use and suicide are major health problems in prisons. Studies conducted in several countries have found an increase in suicide rates in prisons over the last 5 decades.2,3,4 However, most of the results presented in the current literature are epidemiologic descriptions of suicide rates rather than detailed analyses of the personal characteristics, social background, and criminal and psychiatric history of incarcerated people who commit suicide. As yet, we do not have much evidence concerning the social integration of those who have committed suicide compared with that of inmates who survive custody.5 There has been little interest in whether the documentation by nonmedical prison staff of suicidal ideation and intent (suicide threats and reports of attempted suicide and self-harm) leads to adequate intervention efforts. In fact, in more than 20% of cases of people who committed suicide in prison in Austria, nonmedical staff had documented signs of suicidality without starting further preventive action (e.g., referral to psychiatric care). Prison suicides are often preceded by signs of suicidal intent, and these signs should, at a minimum, prompt the provision of further psychiatric care.6 Research into suicide in prison must be directed toward the exploration of protective factors (i.e., factors that reduce the probability of suicide) and environmental factors that may influence suicidal behaviour (e.g., prison regimes and prisoners' social networks).7
Gore and Bird8 have drawn attention to the unexpectedly high prevalence of illicit drug use in prisons in the United Kingdom, which may also account for high suicide rates in custody. Illicit drug and alcohol abuse problems have been found to be, like other psychiatric disorders, common characteristics of those who commit suicide in prisons all over the world.3,4,5,9 These results correspond well to the findings of Bland and colleagues,10 who researched whether individuals with mental disorders were overrepresented among those in various legal difficulties as compared with the general population. Among prison inmates, the authors found a lifetime prevalence of 87% for substance abuse, 56.7% for antisocial personality disorder, 22.8% for affective disorders, 15.6% for anxiety/somatoform disorders, 2.2% for schizophrenia and 1.1% for mental retardation. All diagnoses were significantly overrepresented in the incarcerated group, leading to the conclusion that the most disadvantaged population of mentally disordered patients has entered the correctional system after the reduction in the number of places in institutions intended to care for the mentally ill.11,12 In spite of the high (and probably increasing) prevalence numbers of inmates who are mentally ill, there is only very little or no treatment or rehabilitation given in prisons in the Western world.10 Therefore, a further increase in deaths from drug abuse and suicide may have to be expected, if the policy of care for incarcerated individuals is not changed. We have conducted research on suicide in custody in Austria, evaluating all self-inflicted deaths in prison (population 8 000 000; total average daily prison inmate population about 6800). With the exception of preventive initiatives in local county jails, programs for suicide prevention have not yet been implemented here, but the Ministry of Justice supports our research, through providing access to the relevant files and statistical data, and intends to stop the significant increase in inmate suicides.
So far, there is very little information available concerning natural deaths in custody. Natural deaths are the minority compared with violent deaths; nevertheless information is needed concerning deaths from cancer, cardiovascular problems and other diseases. Reporting of high death rates in prison can help to focus attention on prison medical services and can facilitate the implementation of preventive programs, which have proved to be effective.13
This study by Wobeser and colleagues,1 with its careful classification of deaths in custody in the most populous province in Canada, could be used in the development of evidence-based programs to prevent avoidable deaths in custody. The aim of such programs should be to guide prison authorities in setting priorities for the allocation of their (usually very limited) health care budgets: What proportion should be directed toward suicide prevention, or to the management of illicit drug abuse, or to screening programs to identify physically ill inmates? However, we still do not have enough scientifically confirmed evidence regarding both natural and violent deaths in custody to develop preventive strategies that will truly be effective. Further work has to be done to find out more about individual risk factors, precursors of deaths and preventive factors.
𝛃 See related article page 1109
Footnotes
-
Contributors: Following discussions between the authors about the research paper considered here, Dr. Fruehwald wrote the first draft of this paper. Both authors contributed to revisions of the manuscript, and both authors have approved the final version.
Competing interests: None declared.