COMPARISON OF POPULATION-AVERAGED AND SUBJECT-SPECIFIC APPROACHES FORANALYZING REPEATED BINARY OUTCOMES

Several approaches have been proposed to model binary outcomes that ar ise from longitudinal studies. Most of the approaches can be grouped i nto two classes: the population-averaged and subject-specific approach es. The generalized estimating equations (GEE) method is commonly used to estimate population-averaged effects, while random-effects logisti c models can be used to estimate subject-specific effects. However, it is not clear to many epidemiologists how these two methods relate to one another or how these methods relate to more traditional stratified analysis and standard logistic models. The authors address these issu es in the context of a longitudinal smoking prevention trial, the Midw estern Prevention Project. In particular, the authors compare results from stratified analysis, standard logistic models, conditional logist ic models, the GEE models, and random-effects models by analyzing a bi nary outcome from two and seven repeated measurements, respectively. I n the comparison, the authors focus on the interpretation of both time -varying and time-invariant covariates under different models. Implica tions of these methods for epidemiologic research are discussed.