Longitudinal cohort studies.

Rheumatological disorders are complex conditions characterized by a variety of clinical manifestations and courses in different patients with the same condition. The best way to understand the course and prognosis of these patients is through longitudinal observational cohort studies. Such studies have been facilitated by computer technology that allows tracking of large numbers of patients and visits and analysis of a large amount of data. There are requirements for such datasets to be informative: internal and external consistency; clearly defined methods of observation and measurement; complete followup; consideration of potential confounders. We describe the usefulness of an observational cohort analysis using the University of Toronto Psoriatic Arthritis Clinic Database as a model. Analysis issues are highlighted and proposed mechanisms offered for keeping patients involved to ensure complete followup is maintained.