Suicide prevention is one of the major goals today of the Public Health Service of the U.S. government; indeed, this has been the case since the 1960s when the National Institute of Mental Health established a center for the study and prevention of suicide. However, as noted a few years ago in a New England Journal of Medicine editorial (Hudgens, 1983), the accumulation of knowledge about suicidal behavior in the last quarter of a century has failed to bring about a reduction of suicide in America.
That editorial saw three promising strategies for prevention: improvements in diagnosing and treating depression, growth of suicide prevention and crisis intervention centers, and restriction of access to lethal means for committing suicide. In this book we argue that the latter approach has not been seriously considered hitherto and that reducing access to lethal agents such as guns, drugs, and carbon monoxide has great life-saving potential. As well as reviewing existing research, we present a number of supportive studies, which use data from America and elsewhere, concerning the relationship between rates of suicide and a variety of lethal agents.
The Nature of the Problem
The suicide rates for the United States (per 100,000 of the population) during the past fifty years, presented here, indicate that suicide rates have not changed much since the Second World War, although they are lower than those of 1933 (the first year in which all states reported):
In America in 1985, there were 29,543 suicides out of 2,086,440 deaths. It has long been known that suicide rates are higher in men than in women, in whites than in blacks, and in Native Americans than in others (Lester, 1983). However, two important features of suicide have called for special attention in the 1980s. First, rates of suicide are high among the elderly. For example in 1985, the suicide rates by age were as follows:
These figures clearly show that suicide is a major problem for those concerned with the psychological health and social well-being of the elderly, especially older men.
Second, there has been a marked increased in the suicide rates of young adults in recent years, especially among white males. The suicide rate of white males aged 15 to 19 years has risen from 9.3 in 1970 to 15.0 in 1980 (Centers for Disease Control, 1986). For white males aged 20 to 24, the rate has risen from 19.2 to 27.8 in the same period.
This increasing rate of suicide among the young has given rise to much concerned comment. Blum (1987) noted that adolescents are the only group who have not experienced improvement in their health status in the last 30 years. Death from violence has now replaced communicable diseases as the primary cause of adolescent mortality, and more than three-quarters of adolescent deaths are from suicide, homicide, and accidents. The young not only have much potential, but their recent improved material well-being makes their self-destructive tendencies hard to comprehend (Uhlenberg and Eggebeen, 1986).
Although suicide is considered a major mental health problem in America, the size of the problem is much greater in some other nations. In 1980, Hungary led the world with a rate of 45.0, almost four times the American rate. Sri Lanka had a rate of 29.2, Denmark 29.1, Austria 26.0, and Finland and Switzerland both 24.7. Furthermore, Lester (1988d) has shown that nations with the highest suicide rates in 1970 experienced the greatest increase from 1970 to 1980. So the problem is getting worse for these nations.
The rising youth suicide rates of the United States are also found in other nations of the world. The United States had a 40 percent increase in the youth suicide rate from 1970 to 1980; Norway experienced a 224 percent increase, Spain a 93 percent increase, Switzerland an 80 percent increase, and Thailand a 78 percent increase (Lester, 1988e).
Increased suicide rates with increasing age are also found in many other nations. For example, although the overall suicide rate in Hungary in 1980 was 45.0, the rate for males over the age of 75 was 202.2, more than four times higher.
However, figures for completed suicide are but the “tip of the iceberg,” for many more nonfatal suicide attempts occur than acts of completed suicide. These attempted suicides are sometimes called parasuicides, or acts of deliberate self-harm, to distinguish them from completed suicides. Many attempted suicides do not harm themselves sufficiently to come to the attention of the authorities, but it has been estimated that, for every completed suicide, there may be 8 (Farberow and Shneidman, 1961) or even 20 (Wells, 1981) attempted suicides. About 15 percent of these attempted suicides subsequently complete suicide compared with only 1 percent of the general population, so they constitute a high-risk population. One can see, therefore, that the problem of suicide is far greater than a cursory examination of rates of completed suicide might lead us to believe.
Research into the causation of suicide can be divided into psychologically inclined and sociologically inclined studies. In the psychological study of suicide, investigators have identified psychiatric disturbance, depression in particular, as the strongest indicator of a high suicidal risk. Beck and his associates have suggested that one of the cognitive components of depression, namely, hopelessness, is especially useful in predicting future suicidal actions (Beck et al., 1975).
In addition, it has frequently been noted that other types of self-destructive behaviors, such as alcoholism and drug abuse, are associated with an increased likelihood of suicide, and that stress, too (from loss of significant others, health, or status), appears to be especially great in the months before a suicide (Paykel, 1979).
Sociologically, weakened social integration and social regulation appear to be related to high rates of suicide, just as Durkheim (1897) first argued. Sociologists have found that high rates of divorce in a culture, high rates of interdistrict migration, and low rates of church attendance all are strongly associated with high rates of suicide.1
However, the New England Journal of Medicine editorial mentioned earlier was correct in arguing that the vast increase in our knowledge about suicide appears to have had few preventive benefits. And this is not through lack of communication. In the 1960s, about 1000 scholarly articles and books on suicide were published, in the 1970s about 2000, and in the 1980s about 3000 items will have appeared. Scholars are certainly writing about suicide!
Strategies for Preventing Suicide
Two major propositions are advanced in this book: that the availability of methods for committing suicide plays a causal role in suicide and that suicide can be prevented by reducing access to these methods. Before presenting the evidence for these propositions, we review the two main strategies that have been taken in the past to prevent suicide. This will facilitate comparison of our proposals with these earlier approaches, which consist of
establishing community suicide prevention centers and
searching for effective treatment of depressed and suicidal patients by the psychiatric and counseling professions.
Suicide Prevention Centers
Following the lead of the Los Angeles suicide clinic established in the 1950s by Edwin Shneidman and Norman Farberow, and stimulated by the National Institute of Mental Health’s center for suicide prevention, communities across America established suicide prevention centers in the following decades. In the United Kingdom, the Salvation Army started an early suicide prevention service in 1905, but more recently one central organization, the Samaritans, has organized centers throughout the country.
Suicide prevention centers are primarily oriented around a crisis model of the suicidal process. People who are suicidal are conceptualized as being in a time-limited crisis state. Immediate crisis counseling is intended to help the suicidal individual through the suicidal crisis, whereupon a normal life may resume.
The centers typically operate a 24-hour telephone service that people in distress can call to talk to a counselor. Counselors are typically para-professionals—ordinary people who have graduated from a brief training program and use crisis intervention as the mode of counseling (active listening, assessment of resources, and problem-solving). Some centers have walk-in clinics, and a few have set up “stores” in the poorer sections of cities and crisis teams who can visit distressed people in their own homes.
The centers are well equipped for early intervention with people on the verge of suicide, but, as noted (Lester, 1972b), their approach is essentially passive. The suicidal person has to contact the center. Active approaches, such as seeking out discharged psychiatric patients, elderly males living alone, and other high-risk groups are rarely pursued. In addition, community workers such as police officers, clergy, physicians, and even groups such as bartenders, prostitutes, and hair dressers who also come into contact with the public, have rarely been sensitized to the detection of depressed, disturbed, and suicidal people so that they could refer them to suicide prevention centers.
Several studies have been done on the effectiveness of the centers. In the first, Bagley (1968) compared English cities and found that those with a suicide prevention center did have a lower suicide rate compared with cities without such a center. In a more carefully controlled study, however, Barraclough and his colleagues (Barraclough et al., 1977; Jennings et al., 1978) compared another group of English cities, matching them for ecological similarity, and found no effect of the centers on the suicide rate. Furthermore, as a result of reanalyzing Bagley’s data, Lester (1980) concluded that Bagley had not in fact demonstrated the preventive effect of the centers he claimed.
Nor was much evidence of success found in a series of similar studies undertaken in the United States. Lester (1973a, 1973b, 1974a, 1974b) studied large samples of American cities, controlling for size of the city, and found no effect of centers on the suicide rate. For example, from 1960 to 1970, the suicide rate in cities without a suicide prevention center rose from 9.4 to 10.7 while the suicide rate in cities with centers rose from 12.1 to 13.6.
Bridge et al. (1977), who explored correlates of the suicide rates of 100 counties in North Carolina, found no association with the presence of a center. Because they did not examine changes in the suicide rate of the counties over time, however, their study was less than adequate. Indeed, all of the evaluations of suicide prevention centers were severely criticized by Auerbach and Kilmann (1977), who noted the impracticality of using suicide rates as a measure of effectiveness and the absence of work on treatment processes amd client behavior change.
A number of attempts were made in the 1970s to improve the work of suicide prevention centers (Resnick and Hathorne, 1973). The importance of identifying and locating high-risk groups in the population and fashioning specific programs for them was stressed. On the whole, the focus was on better ways of intervening rather than on programs to improve the social environment and thus reduce the forces leading to suicide.
One relevant study was conducted by Wold and Litman (1973), who had the excellent idea of examining the records of those who had killed themselves after calling a suicide prevention center. They found inadequate counseling on the part of some of the volunteers who handled the calls. But more important, they found that the crisis counseling approach was not a suitable means of dealing with chronic, high suicide risk callers who needed referral to other agencies, which typically the callers did not pursue. Wold and Litman discussed the possibility of establishing programs to deal more effectively with such callers.
In the most recent evaluation, Miller et al. (1984) have reported more favorable results. They compared a sample of counties in the United States with and without suicide prevention centers and found that the centers had a beneficial impact on the suicide rates for white females younger than 24 years of age (but for no other group). This result was replicated on a new sample of counties. Since young white females are among the more frequent callers to suicide prevention centers, this result makes sense.
Thus, at the present time it seems that the previously negative conclusions about the effectiveness of suicide prevention centers may need to be modified. If Miller’s results are replicated in other locales, we may conclude that suicide prevention centers do help prevent suicide in the sociodemographic groups that they most often serve.
Psychological Treatment of the Depressed and Suicidal Person
The psychiatric/psychological approach to suicide prevention is to take individual clients and identify the most effective ways of medicating or counseling them to reduce the risk of suicide. Since depression is both the most common psychiatric syndrome and the most common mood accompanying increased potential for suicide, its treatment has been the primary focus of this approach.
The majority of suicides are found to have been psychiatrically disturbed. Barraclough (1972) examined 100 cases of completed suicide and found that 64 had had depressive illnesses. Of these, 44 had previous depressive episodes, and 21 of these met strict criteria for diagnosing “recurrent affective illness.” Barraclough therefore argued for good diagnostic practices and effective treatment as a way of preventing suicide.
The provision of general psychiatric services ought to assist prevention efforts. Ratcliffe (1962) noted that the change of Dingleton mental hospital in patient management from a locked-ward to an open-door system was accompanied by a drop of about 60 percent in the number of suicides in the surrounding community during the following 10 years (while the suicide rate in Scotland, as a whole, stayed constant). Ratcliffe suggested that the open-door policy had induced more of the psychiatrically disturbed citizens in the community to use the psychiatric facility. However, no such drop in the suicide rate was noted by Walk (1967) when a community mental health center was opened in Chichester, England.
Medication is another staple of treatment. Barraclough, in his study reported earlier, argued for the use of lithium for those patients who have bipolar effective disorders. Montgomery and Montgomery (1984) have recently shown that administration of flupenthixol (a depot neuroleptic) resulted in a significant decrease in suicide attempts in suicidal patients diagnosed as having a personality disorder. In contrast, neither a placebo nor Mianserin, a less toxic oral antidepressant, had a significant effect on suicidal behavior.
Effective psychotherapy may also be of benefit for potential suicides. For example, Liberman and Eckman (1981) compared the effectiveness of behavior therapy and insight-oriented therapy for repeated suicide attempters. Each program included individual, group, and family therapy components. They found that the behavior therapy program (which included training in social skills, anxiety management, family negotiation, and contingency contracting) had a more positive outcome than the insight-oriented program after both nine months and two years. However, Montgomery and Montgomery (1982) reviewed six previous studies on the effects of counseling on suicidal behavior and found that only three were adequately designed and only one of these showed counseling had a significant impact on suicidal behavior.
Thus, while medication certainly has a place in the treatment of the depressed suicidal patient, we cannot be too optimistic about the effectiveness of the other psychiatric/psychological approaches at the present time. Furthermore, these approaches are useful only for those individuals who seek out the help of clinics or private psychotherapists for their personal problems.
Other Strategies for Preventing Suicide
Before coming to our own proposals, two other prevention strategies that have attracted little attention should be mentioned for completeness. One of these is to make suicide less likely by manipulating societal approval. As suicide becomes more common, the publici...
