A changing epidemiology of suicide? The influence of birth cohorts on suicide rates in the United States

Abstract

The increases in suicide among middle-aged baby boomers (born between 1946 and 1964) in the United States since 1999 suggest a changing epidemiology of suicide. Using data from 1935 to 2010, this paper conducts age-period-cohort analyses to determine the impact of cohorts in shaping temporal patterns of suicide in the United States. The analysis demonstrates that age, period and cohort effects are all important in determining suicide trends. Net of age and period effects, the cohort pattern of suicide rates is U-shaped, with cohorts born between 1915 and 1945 possessing among the very lowest suicide rates. Suicide rates begin to rise with boomers and subsequent cohorts exhibit increasingly higher rates of suicide. The general pattern exists for both men and women but is especially pronounced among males. The average suicide rate over the entire period for males is about 28 per 100,000, 95% CI [27.4, 28.7]. For males born in 1930–34, the suicide rate is estimated to be 17.4 per 100,000, 95% CI [15.9, 18.8]; for males born between 1955 and 1959, the rate is essentially the same as the average for the period while for males born between 1985 and 1989, the suicide rate is estimated to be 37.8 per 100,000, 95% CI [33.1, 43.4]. The results dispute popular claims that boomers exhibit an elevated suicide rate relative to other generations, but boomers do appear to have ushered in new cohort patterns of suicide rates over the life course. These patterns are interpreted within a Durkheimian framework that suggests weakened forms of social integration and regulation among postwar cohorts may be producing increased suicide rates.