Analysis of change in health-related quality of life in patients with COPD over 6 years, including information on dropouts: Results from the German COSYCONET cohort

Science and Research

INTRODUCTION: Long-term studies as well as adequate methods accounting for attrition are necessary to measure the longitudinal change in health-related quality of life (HRQoL) in patients with chronic obstructive pulmonary disease (COPD). The aim of this analysis was to evaluate the change in HRQoL, measured by the EQ-5D-3L, over 6 years, while adequately accounting for dropouts. METHODS: We examined 6-year mean changes in HRQoL for 2701 COPD patients from the COSYCONET cohort study, based on data from baseline and follow-up visits after 1.5, 3, 4.5 and 6 years. Patients who dropped out during the 6-year follow-up were included in the analysis by imputing missing values using multiple imputation. We incorporated information on dropout reason in the imputation model and used additional information on HRQoL for dropouts to assess the imputation model. RESULTS: Average EQ-5D-3L deteriorated significantly by -0.121 (95%-CI: -0.125; -0.117) over 6 years for all patients and by -0.047 (95%-CI: -0.049; -0.045) when patients who dropped out because of death were excluded. Patients with impaired forced expiratory volume in one second (FEV(1)) in percent of predicted normal values at baseline had a more pronounced 6-year mean decline in HRQoL. Inclusion of dropout information in the imputation model was informative. Assessment of the imputed data revealed that performing multiple imputation produced less biased results compared to complete case analysis. CONCLUSION: The HRQoL decreased significantly over time for COPD patients. We propose a method to include all dropouts as well as additional information on dropouts to reduce bias in the results.