In general, examination of disease rates may be carried out on three time scales: age (of the individual), period (time of diagnosis) and cohort (time of birth). Given any two of age, period, cohort, however, determines the third, and so one cannot uniquely identify the three different components.
Despite numerous warnings (Clayton and Schifflers, 1987a,b, Carstensen, 2007, Smith and Wakefield, 2017) over interpretation continues to occur in the literature.
In this course, the identifiability will be examined, and approaches to inference will be described. In particular, what can and what cannot be deduced from the data alone will be discussed.
Both frequentist and Bayesian methods will be presented. Throughout, ideas and modeling will be illustrated with examples. The examples will use publicly-available data, with methods implemented in the R programming language, with code provided, so that participants in the course will be able to carry out analyses with their own data.
Carstensen, B. (2007). Age–period–cohort models for the Lexis diagram. Statistics in Medicine, 26:3018–3045.
Clayton, D. and Schifflers, E. (1987a). Models for temporal variation in cancer rates. I: age–period and age–cohort models. Statistics in Medicine, 6:449–467.
Clayton, D. and Schifflers, E. (1987b). Models for temporal variation in cancer rates. II: age–period–cohort models. Statistics in Medicine, 6:469–481.
Smith, T.R. and Wakefield, J. (2017). A review and comparison of age-period-cohort models for cancer incidence. Statistical Science. 31, 591–610.