Longitudinal designs are frequently encountered in epidemiologic research, particularly in the cardiopulmonary field. Many different models have been proposed for the analysis of longitudinal data in the statistical literature. These include the general linear model, autoregressive models, random effects models, and simple models based on an analysis of slopes over time. Complex models are not widely used in the epidemiologic literature, due mainly to a lack of understanding of their underlying utility and the types of questions that could be answered with complex models that cannot be addressed using simple models. An additional problem is a lack of software available for fitting complex models. We propose to perform a comparative study of these models on datasets from nine large epidemiologic studies in the cardiopulmonary field. The models will be compared as regards goodness of fit, ease of implementation, interpretability and robustness. The comparative study of these models will provide the opportunity to determine how the different models address given substantive research questions (e.g., how a change in exposure affects future values of an outcome variable). In addition, new statistical methods will be developed to model phenomena which seem poorly-fitted by currently existing methods, including (a) adult longitudinal pulmonary function data and (b) familial data collected in a longitudinal setting. The overall goal of our research is to develop tools for identifying appropriate classes of statistical models to use in analyzing longitudinal data. This has important public health implications, since longitudinal data continues to accumulate rapidly and no guidelines are available as to the appropriate methods of analysis for specific research questions. Furthermore, it is often only through the modelling of longitudinal data that processes pertaining to change can be understood.
Showing the most recent 10 out of 32 publications
