The use of EQ-5D-5L as a patient-reported outcome measure in evaluating community rehabilitation services in Alberta, Canada

Abstract

Abstract Background: The purpose of this study was to evaluate the impact of community rehabilitation services on health-related quality of life (HRQL) in Alberta, Canada, using routinely collected EQ-5D-5L data, and explore factors associated with the impact of these services. Methods: A retrospective, longitudinal, observational design was used. Patients completed the EQ-5D-5L and demographic questions at intake and end of rehabilitation care. Change in EQ-5D-5L dimensions from intake until end of rehabilitation was examined using the Pareto Classification of Health Change. Change scores were calculated for the EQ-5D-5L index, VAS, and total sum scores. Change groups in the EQ-5D-5L index and VAS scores, were defined by minimally important differences of 0.04 and 7.0, respectively. One level change was considered important for the total sum score. Effect size of the change in index, VAS, and total sum scores was also examined. Chi-squared tests were conducted to examine whether change in EQ-5D-5L varied by age, gender, region, and having anxiety/depression at intake. Results: Three service programs were examined; pulmonary rehabilitation (n=542), group-based community exercise (n=463), and physiotherapy for bone and joint care (n=391). At intake, HRQL in all programs was lower than that of the general Alberta population norms and improved by end of rehabilitation. The mean (SD) change in index, VAS, and total sum scores were 0.02 (0.13), 6.0 (18.3), and -0.5 (2.4) in pulmonary rehabilitation, 0.06 (0.13), 6.6 (18.7), -1.2 (2.4) in community exercise, and 0.13 (0.16), 1.2 (0.9), and -2.8 (2.8) in physiotherapy, respectively. Based on change of the index score, 24% deteriorated, 38% improved, and 38% had no change in pulmonary rehabilitation; 17% deteriorated, 51% improved, and 32% had no change in community exercise; 5% deteriorated, 72% improved, and 23% had no change in physiotherapy. Similar trends were seen in the VAS and total sum scores. Age, region, and having anxiety/depression at intake were associated with change in EQ-5D-5L. Conclusions: The results of this study are intended to inform program/service level decisions by describing the characteristics and HRQL of patients accessing community rehabilitation, as well as the predictors of change in health status, which will help direct future program growth and service changes.