An age-period-cohort approach to analyzing trends in suicide in Quebec between 1950 and 2009.

Abstract

Suicide rates in Quebec over the second half of the 20th century show a wide range of variation depending on age and time period. However, few studies have verified the presence of a cohort effect affecting trends in Quebec suicide rates. This study is designed to evaluate the potential effects of age, period and cohort (APC) on trends in suicide between 1950 and 2009 in Quebec. For these APC analyses, we used a multiphase approach combining a graphical inspection followed by an analysis that isolates the cohort effect from age and period effects (linear regression of the residuals from a median polish of the rates). The graphical inspection of trends in rates points to combined effects of age, period and cohort among both men and women. However, the median polish analysis attributes primary importance to period effects, followed by age effects, but also shows weak cohort effects that are significant only among men born between 1950 and 1979. The variation in Quebec suicide rates appears to be primarily a reflection of period, age and, to a lesser degree, birth cohort. Thus, in addition to sex, selection of risk groups should be based more on age and time period than on birth cohort.