Randomized controlled trials (RCTs), such as the Diabetes Control and Complications Trial (DCCT), usually evaluate the efficacy of a single treatment strategy. The DCCT, for example, evaluates intensive diabetes management aimed at achieving glucose levels as close to normal as possible to modify specific pathophysiological outcomes--specifically, the development or worsening of microvascular disease. In contrast, longitudinal observational studies, such as the type II diabetes Patient Outcome Research Team (PORT) study, address medical effectiveness; that is, how well prevailing treatments work in clinical practice settings. The PORT relies heavily on patient-reported measures of general and diabetes-specific health status, in addition to using complications as major study outcomes. In the type II diabetes PORT, 4,000 patients with type II diabetes and a wide range of socioeconomic, demographic, and disease characteristics, from three widely dispersed geographic settings and varying systems of care, are being followed for a 2.5-year period. Data are collected from periodic self-administered patient questionnaires and from administrative data bases. In the PORT study, nonmutable confounders, such as case-mix, and potentially mutable features, such as patients' preferences for treatment, health habits, regimen adherence, family support, and physician's interpersonal style, are carefully measured. The PORT study will examine the effectiveness of preventive care and established disease treatment in relation to eye, cardiovascular, and extremity disease, measuring and relating use of health-care services to patient outcomes. The results have the potential for maximizing quality of care and minimizing use of services in type II diabetes by matching physician-level profiles of patient outcomes with medical-care-process data and making this information accessible to practicing physicians.