Suicide clusters have occurred throughout the world in communities of all types—rich and poor, big and small. Examples are even known from antiquity. While some clustering of suicides may be expected to occur due to chance, many of the clusters reflect contagious processes and are not due to chance, confounding factors, or homophily (among people who somehow have a prior inclination to kill themselves) [1]. …
The classic investigation of suicide contagion was published by sociologist David Phillips in 1974 [2]. He showed that during the period from 1947 to 1968, suicides increased nationally in the month after a front-page article appeared in the New York Times describing someone who had taken his own life. Phillips dubbed this “the Werther effect”, after Johann Wolfgang von Goethe’s novel The Sorrows of Young Werther published in 1774. The novel was read widely, and when some young men began committing suicide in a way that copied the protagonist, authorities in Italy, Germany, and Denmark banned the book.
There are two kinds of suicide cascades: those that work through media contagion, like Young Werther or the front page of Times (these can involve either fictional or factual accounts), and those that work through direction contagion among people who are connected to a person who has killed himself.
Concerns about media contagion have been sufficiently serious that they have led the Centers for Disease Control to suggest alternative ways of publicizing the occurrence of suicide. [3] …
The CDC guidelines recommend that news reports not explain the method of suicide or mention how “wonderful” the deceased teenager was; they should also refrain from suggesting that the suicide helped solve the teenager’s probelms, for example, by getting even with Jane (“When contacted, Jane sobbed as she reported how much she missed John”).
This works. When Vienna, Austria, finished its subway system in 1978, it was not long before people started using it for a purpose for which it was not intended: they flung themselves in front of the trains. Media reports were vivid, and suicide attempts (half of which were successful) numbered nearly forty per year. Viennese psychiatrists became concerned and began working with journalists. Changes in the reporting of suicides were implemented in 1987, and there was immediate and enormous drop in suicide attempts to roughly six per year thereafter. [4]
Since Phillips’s 1974 paper, the sophistication of suicide analysis has grown tremendously, and the geographic scale has narrowed to focus on localized outbreaks and those that occur by direct contagion. As in the case of MPI, the burden seems to fall especially heavily on schools and on small communities that are, as the saying goes, “tightly knit”. Moreover, suicide contagion occurs almost exclusively among the young. Adults older than twenty-four show little, if any, excess likelihood of killing themselves if someone they know has done so or if they simply read about a suicide in the paper. [5] But teenagers, who are especially impressionable and susceptible to peer pressure, are another matter. …
The average suicide rate in Manitoba, Canada, is 14.5 cases per 100,000, but in 1995, in a village of 1,500 people in the far north, the rate was 400 per 100,000. Six young people took their lives, mostly by hanging, in four months. A further nineteen attempted suicide. [6] …
Another well-documented outbreak took place in a high school of 1,496 students in Piitsburgh, Pennsylvania. Two students killed themselves within four days, apparently prompted by the suicide of a twenty-one-year-old former schoolmate; and during an eighteen-day period that included these two cases, an additional seven students attempted suicide, while another twenty-three reported thinking about killing themselves.[7]
Direct contagion can work the same way for suicide as it does for obesity, that is, via a spreading of ideas rather than by shared behaviors. Suicide in one person may lower the threshold for others to follow suit by changing attitudes and norms. It may increase the sense that it is something desirable to do (“look how all these people are so sad at the death of that person”). A case of suicide may make a person feel that the usual normative pressure to refrain from killing themselves is partially suspended. …
The most recent examinations of suicide cascades have merged network methods and very large data sets to further investigate and confirm direct contagion. A study of 13,465 adolescents in Add Health confirmed that having a friend who committed suicide increased the likelihood of suicidal ideation. Boys with a friend who had killed himself in the previous year were nearly three times as likely to think about killing themselves than they otherwise would be and nearly twice as likely to actually attempt it. Girls with a friend who had killed herself were roughly two-and-a-half times more likely to think about killing themselves and also nearly twice as likely to actually attempt it. [8] …
Suicide contagion is not entirely unknown in adults. One study of 1.2 million people living in Stockholm during the 1990s found that men (but not women) who had coworkers who killed themselves were 3.5 times more likely to commit suicide than they otherwise would have been. [9]
M.S.Gould, S.Wallenstein, and M.Kleinman, “Time-Space Clustering of Teenage Suicide,” American Journal of Epidemiology 131 (1990): 71-78.
D.P.Phillips, “The Influence of Suggestion on Suicide: Substantive and Theoretical Implications of the Werther Effect,” American Sociological Review 39 (1974): 340-54.
Centers for Disease Control: “Suicide Contagion and the Reporting of Suicide: Recommendations from a National Workshop,” Morbidity and Mortality Weekly Review 43, no. RR-6 (1994): 9-18.
E.Etzerdsorfer and G.Sonneck, “Preventing Suicide by Influencing Mass-Media Reporting: The Viennese Experience, 1980-1996,” Archives of Suicide Research 4 (1998): 67-74.
M.S.Gould and others, “Suicide Cluster: An Examination of Age-Specific Effects”, American Journal of Public Health 80 (1990): 211-12.
C.Wilkie, S.Macdonald, and K.Hildahl, “Community Case Study: Suicide Cluster in a Small Manitoba Community”, Canadian Journal of Psychiatry 28 (1989): 918-24.
D.A.Brent and others, “An Outbreak of Suicide and Suicidal Behavior in a High School,” Journal of the American Academy of Child and Adolescent Psychiatry 28 (1989), 918-24.
P.S.Bearman and J.Moody, “Suicide and Friendships Among American Adolescents,” American Journal of Public Health 94 (2004): 89-95.
P.Hedstrom, K.Y.Liu, and M.K.Nordvik, “Interaction Domains and Suicides: A Population-Based Panel Study of Suicides in Stockholm, 1991-1999,” Social Forces 87 (2008): 713-40.
Looking at those articles, it seems like the main criticism is that some of the authors’ analyses fail to distinguish between homophily and contagion. Several of the reports cited in my quote seem too extreme for homophily to be the main factor, and they’re much older and by other authors. So although Connected should be taken with a grain of salt, suicide clusters still seem like a real phenomenon.
From the book Connected:
M.S.Gould, S.Wallenstein, and M.Kleinman, “Time-Space Clustering of Teenage Suicide,” American Journal of Epidemiology 131 (1990): 71-78.
D.P.Phillips, “The Influence of Suggestion on Suicide: Substantive and Theoretical Implications of the Werther Effect,” American Sociological Review 39 (1974): 340-54.
Centers for Disease Control: “Suicide Contagion and the Reporting of Suicide: Recommendations from a National Workshop,” Morbidity and Mortality Weekly Review 43, no. RR-6 (1994): 9-18.
E.Etzerdsorfer and G.Sonneck, “Preventing Suicide by Influencing Mass-Media Reporting: The Viennese Experience, 1980-1996,” Archives of Suicide Research 4 (1998): 67-74.
M.S.Gould and others, “Suicide Cluster: An Examination of Age-Specific Effects”, American Journal of Public Health 80 (1990): 211-12.
C.Wilkie, S.Macdonald, and K.Hildahl, “Community Case Study: Suicide Cluster in a Small Manitoba Community”, Canadian Journal of Psychiatry 28 (1989): 918-24.
D.A.Brent and others, “An Outbreak of Suicide and Suicidal Behavior in a High School,” Journal of the American Academy of Child and Adolescent Psychiatry 28 (1989), 918-24.
P.S.Bearman and J.Moody, “Suicide and Friendships Among American Adolescents,” American Journal of Public Health 94 (2004): 89-95.
P.Hedstrom, K.Y.Liu, and M.K.Nordvik, “Interaction Domains and Suicides: A Population-Based Panel Study of Suicides in Stockholm, 1991-1999,” Social Forces 87 (2008): 713-40.
Debunking social contagion. I’m not going to say this is definitive, but it’s at least a story worth watching.
Many more links.
Interesting, thanks.
Looking at those articles, it seems like the main criticism is that some of the authors’ analyses fail to distinguish between homophily and contagion. Several of the reports cited in my quote seem too extreme for homophily to be the main factor, and they’re much older and by other authors. So although Connected should be taken with a grain of salt, suicide clusters still seem like a real phenomenon.