Age-period-cohort analysis of the incidence of schizophrenia in Scotland.
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.
Mots-clés Pascal : Schizophrénie, Psychose, Incidence, Epidémiologie, Ecosse, Grande Bretagne, Royaume Uni, Europe, Age, Etude cohorte, Homme, Etude longitudinale
Mots-clés Pascal anglais : Schizophrenia, Psychosis, Incidence, Epidemiology, Scotland, Great Britain, United Kingdom, Europe, Age, Cohort study, Human, Follow up study
Notice produite par :
Inist-CNRS - Institut de l'Information Scientifique et Technique
Cote : 96-0425073
Code Inist : 002B18C06A. Création : 10/04/1997.