FORUM on Corrections Research

A Study of 133 Suicides Among Canadian Federal Prisoners

Inmate suicide is documented as the leading cause of death in prisons in Canada⁽²⁾ and in Britain.⁽³⁾ It is also well known that the rate of suicide in penal establishments across North America and Western Europe is higher than that for the general population.⁽⁴⁾ For instance, in one of the most thorough studies of inmate suicide so far; Burtch and Ericson⁽⁵⁾ calculated that from 1959 to 1975, the suicide rate of inmates in Canadian penitentiaries was 95.9 per 100,000 prisoners. This is substantially higher than the corresponding rate of 14.2 per 100,000 for non-prison males in Canada.

Concern over this situation has led to the creation of a suicide prevention program by the Correctional Service of Canada. This program incorporates the concepts of the Province of Alberta's suicide prevention program.⁽⁶⁾ The design, refinement and success of such programs presuppose an understanding of the factors involved in inmate suicide.

A study of Canadian federal prisoners who committed suicide between 1977 and 1988 found that males were most likely to commit suicide, the most common method was by hanging and the act was frequently committed shortly after sentencing. Inmate suicide was also associated with single marital status, earlier suicide attempts, a history of drug or alcohol abuse and a previous history of psychiatric illness. There was, however, no significant relationship between suicide and age, offence type, previous convictions or length of sentence.

This preliminary research suggests a number of possible ways to help prevent inmate suicide. Suicide emerges, however, as a complex, multifaceted phenomenon that requires more research if we are to develop successful, long-term preventive strategies.

In this article, the results of this study are compared and contrasted with findings of other major studies on the topic.

Methodology

Conducted in 1990-91, the study attempted to examine all cases of inmate suicide committed in federal penitentiaries in Canada from 1977 to 1988. Canadian federal institutions detain only those offenders who have been sentenced to terms of imprisonment of two years or more. They do not contain prisoners on remand. The population examined in this study thus consisted entirely of sentenced prisoners serving two years or more.

The records of inmates who committed suicide between 1977 and 1988 were obtained from the records department of the Correctional Service of Canada. There were 133 files available for content analysis; a further 8 cases were not available to us.

Demographic Characteristics of Inmates

Consistent with previous major research, inmate suicide was found to be more common among men.⁽⁷⁾ Of the 133 suicides investigated, only 4 inmates were female and 115 (80%) were Caucasian.

Earlier research shows diverse results in the age distribution of inmates who committed suicide. One study found a higher incidence among younger individuals,⁽⁸⁾ while another showed a higher incidence among older inmates.⁽⁹⁾ Still another study indicated peaks of suicide activity at each end of the age spectrum - one peak at 15-to-19 years of age and the other at over 50 years old.⁽¹⁰⁾ In contrast, our study found the age of inmates who committed suicide to be evenly distributed with no remarkable peaks.

With respect to marital status, previous research in prisons⁽¹¹⁾ and jails⁽¹²⁾ showed quite conclusively that single inmates were more prone to suicide. In accordance, our findings showed that half of the inmates who committed suicide were reported as single (49.6%) and a further 12.1% were single through separation, divorce or bereavement. About 38% were married or living in common-law relationships. Of 130 prisoners on whom the information was available, approximately 60% claimed to have no children, 14% had one child and the remainder had more than one.

One explanation for the higher suicide rate among single prisoners is Durkheim's thesis of egoistic suicide which postulates that individuals with fewer communal ties are more likely to commit suicide. ⁽¹³⁾

Figure 1

Regional Distributions

The numbers of suicides occurring in each of the five different regions of the Correctional Service of Canada varied considerably. The Atlantic region registered 17 (14% of the total number of suicides); Quebec had 49 (40.5%); Ontario had 30 (24.8%); Prairies had 14 (11.6%); and the Pacific region had 11 (9.1%).

The following figure compares these results to the number and corresponding percentage of inmates housed in each region. ⁽¹⁴⁾ We can see that as a percentage of total suicides, the Atlantic region and particularly Quebec are higher than the corresponding percentage of the total population housed in those regions. In contrast, Ontario, the Pacific region and especially the Prairies show a lower percentage of the total number of suicides compared with their share of the inmate population. Precisely what accounts for this variation is unknown.

Circumstances of Suicide

Our findings concur with those of many earlier studies⁽¹⁵⁾ which found that hanging was the most common method of inmate suicide. Hanging comprised 80% of all cases where information was recorded, and at least two thirds of our total sample. While inmates in this study occasionally used clothes or cord to hang themselves, they almost always used bed linen.

In 96% of cases where information was available, and in two thirds of the complete sample, inmates committed suicide in their own cell. The time of suicide was spread quite evenly over the 24-hour period, although 40% (compared with the expected 33%) killed themselves or, more precisely, were found dead by staff between 8 p.m. and 4 a.m. Thirty-three percent were found between 4 a.m. and noon and 21% between noon and 8 p.m.

The time of the year when suicides occurred was very evenly distributed across the entire year and across various regions. This was in contrast to a seasonal fluctuation indicated by some studies.⁽¹⁶⁾

Criminal and Sentence Characteristics

Any conclusive association between violence and suicide is a matter of controversy.⁽¹⁷⁾ The results of this study showed a moderate incidence of violent crimes associated with suicide: 51 (38%) of the suicides had as their most recent offence a non-sexual offence of violence, while a further 34 (26%) had a robbery or weapons offence and 25 (19%) had property offences. Only one individual was a first-time offender. There were no obvious patterns when the frequency of suicide was examined according to sentence length.

With respect to location, one third of those who committed suicide had been in protective segregation at some time during the year before death, while 1 in 10 were in punitive isolation.

A high proportion of suicides occurred relatively soon after sentencing - one quarter within 90 days and about one half within a year of sentencing. The time between sentencing and suicide was not significantly affected by the type of offence or prior record, either singly or in combination.

Psychiatric History

As in many other studies,⁽¹⁸⁾ this study found a considerable incidence of previous psychiatric illness among those who committed suicide. Forty-four percent had at least one previous psychiatric hospitalization on record, most often in the year before the suicide. Thirty-two percent were recorded as being hospitalized in a psychiatric institution outside prison. Twenty-nine percent were recorded as having received psychiatric treatment as an out-patient. While no particular psychiatric diagnosis was on file for more than a few cases, a wide spectrum of diagnoses was observed across the sample as a whole, with no clear pattern emerging.

A history of attempted suicide for these inmates was also considerable. Of the 133 cases, 99 had previously attempted suicide at least once. Of these, just more than half had attempted suicide at least once while in prison, while a further quarter had attempted suicide more than once while in prison. One quarter were recorded as having made one or more attempts outside prison. Of those with at least one previous attempt, the vast majority (94%) had made that attempt within one year of the successful attempt.

The results also showed a positive relationship between alcohol or drug use and proneness to suicide. About two thirds of inmates who committed suicide were on record as having a history of alcohol abuse and just more than half had a history of drug abuse (54%). No information was available on drug or alcohol intoxication around the actual time of suicide.

Discussion

Research on prison suicide provides information that may be used to enhance preventive programs such as those recently introduced in the United States⁽¹⁹⁾ and Canada.⁽²⁰⁾ Our findings indicate that it would be useful to focus on simple institutional solutions, in addition to remedies aimed at individuals.

In institutions, measures can be taken to limit opportunities for committing suicide. For example, with respect to method, greater consideration should be given to such things as the type of bed sheet used, velcro fastenings instead of shoelaces, light fittings that detach under strain and other measures to eliminate opportunity. It is a valid argument to suggest that prisoners who are thought to be suicidal should not be allocated to single cells. We acknowledge that major efforts in this direction have begun and believe that it is in this sort of approach that effective prevention may be found.

With respect to remedies for individuals, in addition to the existing therapeutic and security practices, close attention should be paid to those inmates placed in punitive isolation or protective segregation, and to those who are single or have a previous history of suicide attempts or psychiatric disorder. Particular vigilance should be maintained on inmates soon after they enter the prison system. Given the high frequency of drug and alcohol abuse among prisoners who commit suicide, the availability and positive promotion of drug and alcohol programs could well be relevant. What is clear, however, is that inmate suicide is a complex problem requiring more research into all aspects of the phenomenon, of which only a few were discussed in this article.

---

Acknowledgments: We thank Ray Denson and Ken Pease for their translations of files written in French, Evelyn McCauley of the Research and Statistics Branch, Correctional Service of Canada for her invaluable assistance and Dorothy Little for her secretarial help.

⁽¹⁾Christopher Green, Consultant Psychiatrist, Stockton Hall Hospital, Stockton on the Forest, York, United Kingdom; Glenn Andre, Research Associate, Neuropsychiatric Research Unit, Regional Psychiatric Centre, Saskatoon, Saskatchewan; Kathleen Kendall, Contractor; Pilot Projects, Correctional Service of Canada, Regional Headquarters, Ontario; Terah Looman, Associate Researcher; Department of Psychiatry, Royal University Hospital, Saskatoon, Saskatchewan; Natalie Polvi, Institutional Psychologist, Warkworth Institution, Campbellford, Ontario.
⁽²⁾B.E. Burtch and R.B. Ericson, The Silent System: An Enquiry into Prisoners Who Suicide (Toronto: University of Toronto, Toronto Centre of Criminology, 1979).
⁽³⁾R. Smith, "The State of the Prisons: Deaths in Prison," British Medical Journal, 288 (1984): 208-212.
⁽⁴⁾P. Tournier; "Le suicide en milieu carcéral (1975-1978): Analyse statistique," Revue Internationale de Criminologie et de Police Technique, 36 (1983): 42-49.
⁽⁵⁾Burtch and Ericson, The Silent System, p.7.
⁽⁶⁾"The Bolt Report 1976: Alberta Report to the Task Force on Suicide." Presented to the Minister of Social Services and Community Health, Alberta.
⁽⁷⁾In addition to Burtch and Ericson, The Silent System, see D. Lester; "Suicide and Homicide in USA Prisons," Psychology Reports, 61 (1987): 126. See also R.J. Orlowski (1983) in Smith, "The State of Prisons: Deaths in Prisons," 208-212.
⁽⁸⁾B. Jaye-Anno, "Patterns of Suicide in the Texas Department of Corrections 1980-1985," Journal of Prison and Jail Health, 5 (1983): 82-93.
⁽⁹⁾E. Dooley, "Prison Suicide in England and Wales 1972-1987," British Journal of Psychiatry, 156(1990): 40-45.
⁽¹⁰⁾Burtch and Ericson, The Silent System.
⁽¹¹⁾Jaye-Anno, "Patterns of Suicide in the Texas Department of Corrections 1980 - 1985."
⁽¹²⁾A. Beigel and H. Russel, "Suicide Attempts in Jails: Prognostic Considerations," Hospital and Community Psychiatry, 23 (1972): 361-363.
⁽¹³⁾E. Durkheim, Suicide (Glencoe, Ill: Free Press, 1951).
⁽¹⁴⁾Statistics Canada, Adult Correctional Services in Canada 1989-90 (Ottawa: Statistics Canada, 1990). Figures for 1988-89.
⁽¹⁵⁾See, for example, Burtch and Ericson, The Silent System, p.30. See also R. Esparza, "Attempted and Committed Suicide in County Jails," in B. Danto (ed.), Jail House Blues (Orchard Lake, Mich.: Epic Publications, 1973), p.39.
⁽¹⁶⁾Dooley, "Prison Suicide in England and Wales 1972-1987." See also L. Hayes, "And Darkness Closes In - A National Study of Jail Suicides," Criminal Justice and Behavior, 10 (1983): 461-484.
⁽¹⁷⁾See, for instance, D.J. West, Murder Followed by Suicide: An Inquiry Carried Out for the Institute of Criminology (London: Heinemann, 1965). See also S.A. Backett, "Suicide in Scottish Prisons," British Journal of Psychiatry, 151 (1987): 218-221.
⁽¹⁸⁾See, for example, Jaye-Anno, "Patterns of Suicide in the Texas Department of Corrections 1980-1985." See also D. Topp, "Suicide in Prison," British Journal of Psychiatry, 134 (1979): 24-27. And see Burtch and Ericson, The Silent System.
⁽¹⁹⁾Orlowski, in Smith, "The State of Prisons: Deaths in Prisons."
⁽²⁰⁾E.H. Botterell, "Report on the Study Team's Investigation on Atlantic Region's Suicides." Unpublished report presented to the Commissioner of Correctional Services, 1984.