Social modeling in the transmission of suicidality.

Evidence from twin, adoption, and family studies suggests that there is strong aggregation of suicidal behaviors in some families. By comparison, the role of social modeling through peers has yet to be convincingly established. This paper uses data from four large studies (the WHO/EURO Multicentre Study on Suicidal Behaviour, the WHO/SUPRE-MISS, the CASE study, and the Queensland Suicide Register) to compare the effects of exposure to fatal and nonfatal suicidal behavior in family members and nonfamilial associates on the subsequent suicidal behavior of male and female respondents of different ages. Across all studies, we found that prior suicidal behaviors among respondents' social groups were more important predictors of suicidal behavior in the respondents themselves than previous research had indicated. Community-based suicide attempters in the WHO SUPRE-MISS had higher rates of exposure to prior suicide in nonfamilial associates than in family members. In an adolescent population, exposure to prior fatal suicidal behavior did not predict deliberate self-harm when exposure to nonfatal suicidal behavior (both familial and social) were controlled for, but exposure to nonfatal suicidal behaviors in family and friends was predictive of deliberate self-harm and suicide ideation, even after controlling for exposure to fatal suicidal behavior. The potential impact of "containment" of information regarding suicidal behaviors as a prevention initiative is discussed, in light of information behavior principles of social marketing.