Is the age-period-cohort model well suited to an epidemic context? The case of the French BSE epidemic
2012
Sala, Carole | Durand, Benoit | Costagliola, Dominique | Ducrot, Christian | Calavas, Didier
Age-Period-Cohort (APC) models are routinely used in time trend analyses of chronic diseases but few examples of their application to epidemic diseases are available. APC analyses of the French bovine spongiform encephalopathy (BSE) epidemic evidenced an unexpected period effect first attributed to the design of the study, in relation to low BSE prevalence and a short surveillance period. The aim of our study was to evaluate the relevance of our hypothesis, the behaviour of the APC model in an epidemic context such as BSE epidemic and the impact of maintaining a period effect in APC model on the estimate of birth cohorts effects. We simulated datasets mimicking the French BSE epidemic and its varying pattern, as well as duration and location in time of surveillance period, and analysed them with a categorical APC model. Results showed that a period effect was evidenced in 44% of analysed datasets while no period effect was introduced in the data simulation process. Such an artefactual period effect was detected when a sudden change of cohort prevalence occurred over a short period of time. Additionally, in the context of BSE, maintaining a period effect in the model could strongly disturb the estimation of cohort prevalence, depending on epidemic pattern and, as expected, duration and location of the surveillance period in relation to the year at which highly infected birth cohorts are detected as BSE cases
显示更多 [+] 显示较少 [-]