A Patient-based severity measure of chronic lung disease: evaluating functional health and predicting health services use.

Prior research has consistently shown that among patients with chronic lung disease (CLD), health-related quality of life (HRQOL) is tied more to respiratory symptoms than to physiologic measures. However, traditional methods to quantify the severity of CLD have been restricted to physiologic measures (eg, FEV1, FVC, etc) that are often poor predictors of HRQOL and utilization of health services. Using a patient-based measure of symptom severity for CLD developed in the Veterans Health Study (VHS), this article evaluated the impact of the severity of CLD on patients' self-reported HRQOL and future use of health services. We used data from the VHS, a prospective study of patients receiving ambulatory care services in 4 Veterans Affairs outpatient clinics in the greater Boston area. Three hundred fifty-two (14.5%) patients were identified as having CLD through self-report of having a physician's diagnosis of chronic bronchitis, emphysema, or asthma, and either using inhaled medications or having a productive cough for most days for 3 months. Ordinary least-square regressions were used to ascertain the effects of CLD severity on functional health and health services use. Compared with peak expiratory flow rate, which explained only 10% and 2%, respectively, of the variance in the SF-36 physical component summary (PCS) and in future doctor visits, the symptom severity explained 19% and 19%, respectively, of the variance in PCS and future doctor visits, after adjusting for age, education, and household income. The symptom severity measure is a strong predictor of future functional health (at 12 months of the VHS baseline) and health services use (within 6 months following the baseline). The study findings indicate that our measure of CLD severity is an efficient and easy-to-use approach that can be readily administered in ambulatory setting. It can be used as a case-mix adjustment in evaluating health outcomes and in predicting future utilization of health services.