Every few days, we will be publishing layman’s abstracts of new articles from our prestigious portfolio of journals in statistics. The aim is to highlight the latest research to a broader audience in an accessible format.
The article featured today is from Statistics in Medicine and the full article, published in issue 39.4, is available to read online here.
Zeng, L, Cook, RJ, Lee, J. Multistate analysis from cross‐sectional and auxiliary samples. Statistics in Medicine. 2020; 39: 387– 408. doi: 10.1002/sim.8411
Epidemiological studies routinely involve sampling individuals from a population at a particular calendar time and assessing the relationship between the binary disease status indicator and some explanatory variables.
The effects of the explanatory variables on disease status are often interpreted as if they were risk factors for disease occurrence but the effects are complex functions of calendar trends in disease incidence, disease-free and post-disease morality rates, and the effects of covariates on such transitions. We show that cross-sectional analyses the relationship between fixed explanatory variables and disease status from cross-sectional samples yield uninterpretable effect measures except under highly specialized models. We also demonstrate that data from auxiliary samples facilitate fitting models for the full underlying life history process to obtain meaningful measures of risk and effects of explanatory variables. An application is given to data from a national cross-sectional sample assessing marker effects on psoriatic arthritis among individuals with psoriasis.