Psychological Medicine

Original Articles

Age–period–cohort analysis of the incidence of schizophrenia in Scotland

Noriyoshi Takeia1 c1, Glyn Lewisa1, Pak C. Shama1 and Robin M. Murraya1

a1 Department of Psychological Medicine, Institute of Psychiatry and King's College Hospital, London


Studies examining a possible decline in the incidence of schizophrenia over the last two to three decades have paid little attention to the possible role of birth cohort effects. We collected data on a Scottish national sample of all schizophrenic patients, admitted for the first time between 1966 and 1990 (N = 11348; male = 6301). In an Age–Period–Cohort analysis, a full model, incorporating three factors, had a substantially better fit to the data than other models (especially, an Age–Period model), providing clear evidence of the presence of a cohort effect. After adjustment for the effects of age and period, there was a 55% reduction in the rate of schizophrenia in men and a 39% fall in the number of women over the 50-year birth period from 1923 to 1973. The marked decline in the first admission rates observed in Scotland cannot, however, be attributed entirely to this cohort effect. Rather, a greater proportion of the declining first admission rates (88%) is ascribed to the period effect (i.e. artefactual or causally related cross-sectional effects). Nevertheless, the fact that a birth-cohort effect accounts for part of the declining incidence, suggests that causal environmental factors operating early in life have been diminishing in intensity.


c1 Address for correspondence: Dr Noriyoshi Takei, Genetics Section Department of Psychological Medicine Institute of Psychiatry and King's College Hospital De Crespigny Park London SE5 8AF.