Age–period–cohort analysis of Swiss suicide data, 1881–2000

Abstract  At the end of the 19th century, male suicide rates in Switzerland were as high as the respective rates in recent decades, whereas female suicide rates were distinctly lower. An age–period–cohort analysis was performed to provide more information about the genderspecific changes over the last century. Suicide mortality has been reported in Switzerland since 1876 when the standardised registration of mortality data began. The analysed data cover the period 1881–2000. The statistical analyses were based on log–linear models and data aggregated by 10–year age–intervals and 10–year periodintervals. The results indicate similar age and period effects in males and females. The estimates representing age–specific risk increase steadily with age, with intermediate plateaus in the 20s and the 50s. The period–specific estimates follow the economic cycles. The birth cohort effects are stronger in males and weaker in females. In the males' estimates, there is a peak in cohorts born around 1840 and a low in cohorts born some 60–100 years later. The estimates increased again in generations born after World War II. In females, the birth cohort estimates are low in cohorts born in the first half of the 19th century and increase until the first half of the 20th century. Birth cohort effects remain an intriguing topic in epidemiology of suicide. A better understanding of birth cohort effects might open new doors to suicide prevention.