Modeling Suicide Contagion as an Epidemic
Suicide contagion is a serious and real phenomenon in society in which a single suicide can inadvertently cause multiple other suicides. As opposed to the spread of disease, there are no physical media, such as bacteria or viruses, that are being transferred. Instead, “publicity and news coverage” around one particular suicide allows suicides to spread like a physical disease. In The New York Times article on suicide contagion, they brought up an example regarding the suicide of Marilyn Monroe in 1962. The extensive news coverage on this incident prompted a series of suicides in the nation, resulting in a “12 percent increase in suicide rate…compared with the same months in the previous year.” In a sense, Cornell also experienced a period of suicide contagion in the past, which led to increase in funding and support for the Gannett Health Center.
To model this form of epidemic, one can use the model involving the susceptible, infected, and removed labels for physical diseases. However, the actual models of the suicide contagion would be slightly different from that of a physical disease. The “removed” label would not apply here because, technically speaking, everyone is susceptible to suicidal thoughts. The “infected” label can represent people who are in deep depression or have already committed suicide. As a result, the S-I-S model could be used to model suicide contagion, which describes a person being susceptible to suicide, becoming very depressed and contemplating suicidal thoughts, and breaking out of depression and becoming susceptible again. This is an adequate model; however, it is not a complete model for suicide contagion. There are certain people who fall under the S-I model, in which the person unfortunately does not make it out of depression and commits suicide. A more complete model of suicide contagion involves modeling the general population as a mix of S-I-S and S-I models for a certain period of time, and then switching to just S-I-S model in which the contagion eventually dies off and stops spreading. Whether the person gets S-I-S or S-I in the initial period is determined by a certain probability influenced by the surroundings. The contagion is initiated by an “initial adopter”, who causes an “information cascade” to all the people who witnessed or are affected by the suicide. The contagion is spread by people with S-I model, in which every person who committed suicide acts as another “initial adopter” for another contagion. Modeling suicide contagion through these methods allow experts to contain and prevent the spread of the contagion by focusing on the initial period of the mix of S-I-S and S-I models. According to the article, suicide prevention advocates created guidelines for media coverage of suicides that involve avoiding over-emphasizing or glamorizing suicides, which has actually been proven to drop suicide risk in the general population. This basically increases the probability for the general people to follow the S-I-S model rather than the S-I model in the initial period of the contagion.
Sources: http://www.nytimes.com/2014/08/14/upshot/the-science-behind-suicide-contagion.html