Random regression models for multicenter clinical trials data.

A random-effects regression model is proposed for the analysis of data arising from multicenter clinical trials. Advantages of the random regression model (RRM) in this context include that it allows for varying numbers of subjects within the different centers, it can assess the influence of variables measured both at the level of the subject and at the level of the center on the subject's clinical outcome, and it controls for and estimates the amount of intracenter variation that is present in the data. An example utilizing data collected in the National Institute of Mental Health schizophrenia collaborative study, where subjects were clustered within nine centers, illustrates the usefulness of the statistical model. Other applications and extensions of RRM within a psychiatric framework are discussed.