Canadian Task Force on Preventive Health Care

Full Text Review

Please note: In 2003, the CTF updated its Grades of Recommendations to include an "I Recommendation" for situations where insufficient evidence exists to allow a recommendation to be made.  (Formerly, these situations were captured under a "C Recommendation".)  This change is not retroactive, and all "C Recommendations" made prior to 2003 have not been reevaluated in light of the new "I" recommendation grade.  For a discussion of these recommendation grades, please link to the 2003 article in the Canadian Medical Association Journal here.

Prevention of Suicide

Prepared by Jane E. McNamee, MA, Research Associate, Department of Psychiatry, Chedoke McMaster Hospitals and David R. Offord, MD, FRCPC, Professor of Psychiatry, McMaster University, Hamilton, Ontario

These recommendations were finalized by the Task Force in March 1994

Up Contents

Chapter Overview

In 1989 the Task Force recommended that primary care physicians routinely evaluate suicide risk among patients in high-risk groups. Available interventions include counselling, follow-up and, if necessary, referral to a psychiatrist.< 1,2>Updated information shows changes in age-groups at increased risk for suicide.<3> Highest-risk groups in 1960 were men aged 70 years or more, and women aged 65 to 69 years. In199 1 highest-risk groups were men aged 20 to 24 years, and women aged 40 to 44 years. Other identified high-risk groups include people who have previously attempted suicide; those witha history of psychiatric illness; people with a history of drug and alcohol abuse especially those living in isolation; those with a chronic or terminal physical illness; Native people; people with a family history of suicide; and persons in custody. Additional groups identified recently are Aboriginal and Native youth in northern communities and new immigrants. There is some evidence that physician education in treatment of affective disorders can reduce suicide rates. Studies evaluating suicide prevention in school-based intervention programs, community-based suicide prevention centres, or hospital-based intensive follow-up situations have shown that none reduced the incidence of suicide significantly. Studies evaluating medical treatment for reduction of suicidal ideation or depression have shown some reduction of depression. However there is insufficient evidence to recommend for or against the combination of routine evaluation of suicide risk with intervention programs.

Burden of Suffering

Suicide has accounted for about 2% of annual deaths in Canada since the late 1970s. (Unless otherwise stated, mortality rates cited here refer to deaths per 100,000 population). Between 1960 and1991 the suicide rate in Canada increased from 7.8 to 12.8, peaking at 14.0 in 1986. In 1991, Alberta had the highest suicide rate (17.8), followed by Prince Edward Island (16.8) and Quebec (15.7). Between1960 and 1991 the greatest increase in the provincial suicide rate was a three-fold increase in Quebec.

Eighty percent of all suicides reported in 1991 involved men. The male:female ratio for suicide risk was 3.8:1 (4.0:1 in 1960). In 1991 the highest suicide rate in males was in the group aged 20 to 24 years(33.3), followed by those aged 25 to 29 (29.7), and those aged 30 to34 (29.2). The highest rate in females was in the group aged 40 to44 years (8.3), followed by those aged 45 to 49 (7.7), and those aged30 to 34 (7.3). Other groups, in decreasing order of risk, were malesaged 35 to 39 (27.2), and 50 to 54 (26.7). Other at-risk groups among females were those aged 55 to 59 (7.2), 25 to 29 (7.1), and 65 to 70(7.1). For both sexes the overall rates decreased with age, but began to increase again in later years.

Preferred methods of suicide remain unchanged. In 1991, males chose firearms (36%), hanging (30%) and by gas vapours (10%).Females chose to ingest solid or liquid substances (38%), hanging (24%)and drugs or medication (12%).<3>

In both males and females, the greatest increase between1960 and 1991 occurred in the 15-to-19-year age group, with a four-and-a-half-fold increase for males, and a three-fold increase for females. Among males, other age groups at greater risk in 1991, compared with1960, were those aged 10 to 14 (3.5:1), and 20 to 24 (2.7:1). Among females, other groups at greater risk in 1991, compared to 1961,were those aged 35 to 39 (2.2:1), and 40 to 44 (1.8:1). Despite the considerable increase in the suicide rate in young age groups, few reliable predictors of suicide in young people have been identified. One unexpected finding is that, since 1969, the rate of suicide in15-19-year-old males has been higher in Canada than in the U.S., by as much as 50 percent, in many years.<4> The potential years of life lost (PYLL) to age 75 in 1986 due to suicide were 122,908 per100,000 population, 97,613 among males and 25,295 among females in Canada.<5>

Suggestions that Canadian suicide rates might be affected by under-reporting have recently been refuted. One study<6> found that under-reporting suicide in Canada did not substantially alter findings, suggesting that most conclusions based on official rates are essentially correct.

The public health impact of attempted suicide is substantial; the burden on casualty, general medical and psychiatric services is considerable. The risk of suicide after an attempt has been reported from 26.9:1 to 100:1. Among such individuals it is estimated that one-third have reported previous episodes of self harm and 15-20%will repeat within 3 months. The risk is highest within the first3 years, especially during the first 6 months after an attempt.<7>

The mentally ill (those with affective disorder, schizophrenia, neurosis, personality disorder or organic brain syndrome) and people with drug and alcohol problems are at greater risk (by a factor of2.4 to 23 times) than the general population.<8,9> Suicide is the chief cause of premature death among schizophrenics with a rate of 350-600 per 100,000 schizophrenic persons.<10> Depression and alcoholism are associated with suicide in elderly males living alone.<11> People with a chronic or terminal illness are at increased risk, from 4:1 among cancer patients to 66:1 among those with acquired immune deficiency syndrome (AIDS). In patients with symptomatic HIV infection the risk is thought to be even higher.<12>

Suicide rates in the Canadian Native population are more than twice the sex-specific rates, and three times the age-specific rates of non-Native Canadians (56.3 for Native males and 11.8 for Native females). Among Aboriginal males, the rate for the 15-24 year age group was more than double that for all Aboriginal males(90.0:39.0).<13> Suicide among northern Native youth has reached epidemic proportions. In Alberta the rates in the northern region were 80.1, in the central region, 71.2, and in the southern area,35.3.<14> An extremely high overall rate (180.2) has been found for 10-19 year-old Native males living on the northern coast of Labrador.<15> The 1991 Aboriginal Peoples Survey indicated that slightly more than two-fifths (41%) of Inuit, and 34.5% of Native Indians on reserves, report that suicide is a problem in their community.<16>

Suicide rates in federal and provincial prisons vary from 8 to47 times rates in the general population. Suicide is the primary cause of death in Canadian penal institutions.<17> In addition to the high rate of completed suicide, the rate of nonfatal self-inflicted injuries among inmates is considerably higher than that for the general population (11.8:1 for males, and 45:1 for females). People with family member who committed suicide are nine times more likely than others to kill themselves.<18> Ethnicity has also been associated with suicide, with first generation immigrant females (2:1), particularly those of European and Asian origin being at higher risk.<19>

Maneuver

Physician involvement in suicide prevention is crucial. Out of every 10 persons who complete suicide, eight have previously signalled their intentions quite clearly. Distress signals about their suicidal impulses are often aimed at family physicians.<20> Studies show that many adolescents,<21> adults,<22> including physicians,<23> and elderly people<24> who commit suicide contact their family physician shortly before death. Successful intervention for suicide prevention depends on early detection, accurate assessment and diagnosis and appropriate treatment by the family physician. Physicians should routinely evaluate the risk of suicide among people in high-risk groups, particularly if there is evidence of psychiatric disorder. Assessment should include personal and sociodemographic risk factors, a thorough collection of clinical data, categorization of the problem and matching of the findings with well established risk factors.

Four maneuvers for treatment of suicidal patients have been identified: 1) medication (if indicated) targeted to the treatment of diagnosed mental illness. Prescriptions should be given for small amounts to reduce the risk of over dosage; 2) appropriate use of psychiatric consultations, referral and hospitalization where necessary;3) psychosocial and psychotherapeutic interventions such as provision of social support, counselling and close follow-up visits, especially for depressed and isolated patients; and 4) environmental interventions, such as educating the patient and family members about the illness, identifying coping strategies and stress management, and development and utilization of a network of community social supports. Additionally, the physician may take a community leadership role among health care professionals, policy-makers, civic leaders and the general public to enhance support of suicide research and prevention programs, and development of "At-Risk-Clinics".

Effectiveness of Prevention and Treatment

Although there is only poor evidence (level III) supporting the effectiveness of intervention by the family physician in preventing suicide, there is fair evidence that family physicians may have a low rate of recognition of psychiatric disorders and suicide risk.<24> In one study of completed suicides, more than two-thirds of the victims had made previous attempts, but only 39% of their physicians were aware of their history.<24> Of the more than 30,000 people world-wide who have taken the suicide-prevention program at the Suicide Information and Education Centre (SIEC) in Calgary, physicians have constituted less than 1%.<25>

One recent study<26> has shown that physicians’ knowledge of the risk and treatment of suicide patients improves after training in a suicide prevention program. Physicians’ knowledge was compared, in pre-post situations, for three groups: those with no training, a group receiving written information only, and a group receiving the written information and seminar training. The last group showed significantly more knowledge about suicide prevention than the two other groups. However long-term follow-up of retention of knowledge, and its effect on the suicide rate, were not reported. Another pre-post study<27>reported decreased suicide rates after a systematic post-graduate training program for general practitioners. The program was directed at the diagnosis and treatment of patients with affective disorders. The study location was a geographically defined catchment area. Suicide rates were compared, pre- and post-program, with the area rates for previous years, and the Swedish national rate. A 50% reduction in the community suicide rate was found in the year following the program, which was attributed to the effects of the program. However no comparisons were made for subsequent years, and no long-term follow-up was conducted. Replication of this type of program in a controlled trial, with long-term follow-up is needed. Overall, there is fair evidence for both the benefit of physician education programs and reduction of suicide, after physician education, in selected groups.

The increase in attention given to adolescent suicide has led to a proliferation of school-based prevention programs. These programs have been described as ineffective, inefficient and even potentially deleterious in their attempts to reduce suicide risk. Program goals are to: 1) raise awareness of the problem of adolescent suicide; 2) train participants to identify adolescents at risk; and 3) educate participants about community mental health resources. The mean duration of U.S. programs is two hours. No evaluation of a Canadian curriculum-based suicide prevention programs exists. One descriptive survey<28>examined 115 U.S. school-based suicide prevention programs. It found that short-term educational interventions were not effective in the prevention of suicide among self-identified adolescent suicide attempters. It was suggested that such programs might actually facilitate suicide, or suicide behaviour, by not allowing adequate time to deal with issues raised by program content. A more recent study<29> which compared pre-post attitudes of curriculum-based suicide prevention programs for 758 teenagers with those of680 control pupils also found little program impact. Both studies recommended caution in relation to prevention programs because of the possible stimulation of imitative behaviour in vulnerable youths. One recent study<30> advised physicians consulting to schools in the aftermath of a suicide, to resist the pressure to implement a curriculum-based prevention program that may have little or no impact. They suggested that physicians and other health care workers take a more active role in identifying children who may be at risk for suicidal behaviour, such as friends and relatives of the victim.

Suicide prevention centres differ from crisis centres and general help lines in the specificity of their focus on suicide. Evaluation of the effectiveness of the 97 Canadian suicide prevention and crisis intervention centres, listed in the 1993-1994 Handbook of the American Association of Suicidology, has not been carried out, due to non-comparability of data across centres. The consensus from evaluation studies in the United States suggests that suicide hotlines are minimally effective in reducing suicidal behaviour and community suicide rates. One descriptive survey<31> found that crisis centres do attract high-risk populations; centre clients were more likely to commit suicide than were members of the general population, and individuals who committed suicide were more likely to have been clients than were members of the general population. In another survey,<32> a slight reduction in risk in young white women, who were the most frequent users of such services, was found. Preliminary data from a descriptive evaluation survey of two suicide prevention centres in Quebec<33> showed that a significant number of callers feel less depressed at the end of a call, and that there was a reduction in suicidal urgency in a large number of callers. More work needs to be done in standardizing techniques used at different Canadian centres to achieve any meaningful evaluation of their reduction of suicide risk.

Medical treatment for the prevention of suicide involves mainly the treatment of depression and management of individuals who have attempted suicide previously. One early cohort study of hospitalized patients<34> admitted for self-inflicted injuries reported fewer subsequent suicide attempts in those who received psychiatric counselling, compared with controls who were discharged before seeing a psychiatrist. Another early survey of medical interventions<35> found that the risk of suicide among people with affective disorder was decreased when optimal pharmacotherapy was combined with routine psychiatric consultation. A more recent meta-analysis<36> which combined data from 17 double-blind clinical trials in patients with depressive disorder, showed that significantly fewer patients treated with fluoxetine suffered an increase in suicidal thoughts and actions, when compared to patients treated with placebo or a tricyclic antidepressant. Another descriptive survey of the use of antidepressants in the provocation or the prevention of suicide<37>concluded that antidepressants, with serotonin reuptake inhibitors, had a clear and consistent positive effect in reducing suicidal behaviour during treatment in depressed patients.

A review<38> of five British psychosocial intervention studies shows no statistically significant reduction in the risk of suicide, or episodes of deliberate self-harm in the active treatment groups. A study of parasuicides,<39> randomized to hospital admission or to discharge home, found no significant differences in psychological tests or further suicide attempts between the two groups at one week follow-up, but long term results were not evaluated. A more recent Canadian study,<40> which randomized suicide attempters to intensive follow-up or usual care, failed to reach its goal of halving the risk of repeat attempts in the intensive intervention group in the two years following the suicide attempt. Although not statistically significant, a 2% decrease of risk of a repeat suicide attempt was found in the intervention group. Of studies to date, none has shown a statistically significant benefit of psychosocial intervention on reducing suicide repetition rates. However, one limitation of these studies is that all except the Canadian study lacked sufficient power to assess whether an intervention reduced the risk of suicide repetition. That is, sample size was not adequate to detect clinically significant effects between groups. Also the results may not be generalizable to all persons attempting or completing suicide because of potential differences between persons attempting and persons completing suicide. Further evaluation of psychosocial interventions is required to assess their effectiveness.

Involuntary hospitalization may benefit persons with suspected suicidal intentions, and may be required for medical or legal reasons. However no data are available to assess the effectiveness of this intervention in reducing the risk of suicide.<41>

Recommendations of Others

In 1989, the U.S. Preventive Services Task Force recommended against routine screening for suicidal intent.<42> However, they also felt that clinicians should be alert to signs of suicidal ideation in persons with established risk factors; persons suspected of suicidal intent should be questioned regarding the extent of preparatory actions and referred for further evaluation if evidence of suicidal behaviour was detected.

Conclusions and Recommendations

There is poor evidence (based on expert opinion alone) to include or exclude routine evaluation of suicide risk in the periodic health examination (C Recommendation). Physicians should remain alert to the possibility of suicide in high-risk patients. They should routinely evaluate the risk of suicide, particularly if there is evidence of psychiatric disorder, especially psychosis, depression or substance abuse, or if the patient lives alone, recently attempted suicide, or a family member has committed suicide. Special attention should be paid to young Native and Aboriginal males. Physicians should be particularly alert when patients carry several known risk factors for suicide.

Physician education in the prevention of suicide, and thedetection and management of depressed patients are promising approaches to the reduction of suicide risk. There is fair evidence to support physician education programs on suicide prevention(B Recommendation). There is insufficient evidence to evaluate school-based or community-based programs or interventions for those who have previously attempted suicide (C Recommendation).However, there is fair evidence to use medical therapy in the treatment of suicidal ideation (B Recommendation) and, where appropriate, for diagnosed depression (A Recommendation).

Unanswered Questions (Research Agenda)

The following have been identified as research priorities:
  1. Evaluating the accuracy of family physicians in identifying psychiatric disorder and in determining suicide risk.
  2. Obtaining better information on the factors that predict suicide among persons in high-risk groups, such as those who have a psychiatric disorder or have attempted suicide, particularly Northern Native and Aboriginal youth, or among those who abuse alcohol and other drugs.
  3. Developing, monitoring and participation in education programs designed to increase physicians’ effectiveness in identifying psychiatric disorder and people at high risk of suicide.
  4. Evaluating the effectiveness of suicide prevention programs that combine identification of high-risk patients and subsequent intervention, with long-term follow-up.
  5. Greater collaboration between physicians and school personnel in developing curriculum-based programs aimed at intervention/postvention or prevention, and the research of the reduction of adolescent suicide risk.

Evidence

The literature was identified with a MEDLINE search from January 1967 to November 1993 using the following key words:suicide, attempted suicide, parasuicide, epidemiology, at-risk populations, prevention, intervention, postvention.

Review of this topic was initiated in November 1993 and updates a report published in 1990.<1,2> Recommendations were finalized by the Task Force in March 1994.

Full Citation

Link to Structured Abstract of this review

Link to Summary Table of this review

Link to Selected References list of this review

Top of Page

Home PageCTFPHC Home Page

Reprinted in modified format by the Canadian Task Force on Preventive Health Care
with permission.
For any technical issues please contact: webmaster@ctfphc.org
Original Copyright © 1994 Minister of Supply and Services Canada.
Last modified March 27, 1998.