Age-period-cohort analysis of suicide mortality rates in Spain, 1959-1991

Background. Although there is evidence that suicide rates may be increasing in Spain, formal epidemiological studies have been limited to specific cities or counties. The objective of this study was to investigate nationwide trends in suicide mortality from 1959 to 1991 in Spain, with emphasis on age, period, and cohort effects. Methods. Age- and sex-specific suicide mortality rates from 1959 until 1991 were obtained from official vital statistics tables from the Institute Nacional de Estadistica, the official registry of vital statistics in Spain. Poisson regression and graphical methods were used to model and estimate age, period and cohort effects. Results. Suicide mortality rates increased with age, with a proportional increment for each decade of life of 45% (95% confidence interval: 45-46%). In both males and females, age-adjusted suicide mortality rates decreased from 1959 until the late 1970s and early 1980s. In 1982, trends started to increase, returning to the levels of 1959 in less than 6 years. Cohort effects were small for cohorts born prior to 1940. For cohorts born after 1950, suicide rates increased markedly. Conclusions. The increase in suicide mortality in younger cohorts and the high rates of suicide in the elderly demand further investigation to establish causal mechanisms and preventive strategies.