Validity and responsiveness of the EuroQol as a measure of health-related quality of life in people enrolled in an AIDS clinical trial

Objective: Brief utility measures are needed in clinical trials in addition to existing descriptive measures of health-related quality of life (HRQOL). We examined the reliability and validity of the EuroQol (EQ-5D) and MOS-HIV and their responsiveness to HIV-related clinical events. Methods: Subjects with advanced HIV disease (CD4 < 100) were enrolled in a randomized trial for CMV prophylaxis (n = 990). The EQ-5D includes a weighted sum of five domains (EQ-5D Index) and a visual analog scale (EQ-VAS). The MOS-HIV has 10 subscales and physical (PHS) and mental health summary scores (MHS). Construct validity of the EQ-5D was tested based on hypothesized relationships to subscales of the MOS-HIV. Relative precision and responsiveness to adverse experiences and opportunistic infections (OIs) were compared for the two instruments. Results: Mean age of the patients was 38, 94% were male, 80% white, and 7% had injected drugs. Mean baseline scores for EQ-5D Index and EQ-VAS were 0.80 and 76.0, respectively, 28 and 4% reported maximum scores. Mean MOS-HIV subscales score ranged from 55 (role) to 84 (cognitive); mean PHS and MHS were 47.4 and 49.5, respectively. Correlations between MOS-HIV subscales and EQ-5D Index ranged from 0.45 (role) to 0.63 (pain); correlations with EQ-VAS ranged from 0.33 (cognitive) to 0.66 (health perceptions). Correlations between MOS-HIV PHS and MHS with EQ-5D Index were 0.61 and 0.58; and with EQ-VAS, 0.57 and 0.60, respectively. Responsiveness to adverse experiences was highest for MOS-HIV pain and PHS (effect sizes = 0.9 and 0.4); pain had the highest relative precision (2.4) for adverse experiences; EQ-VAS had the greatest relative precision (1.6) for developing an OI. Conclusion: In these patients with advanced HIV disease, EQ-5D showed good construct validity, but there may be a ceiling effect for its EQ-5D Index component. EQ-5D was less responsive to adverse events than the MOS-HIV. However, the EQ-VAS was most sensitive to developing an OI and is likely to be a useful measure of HRQOL for generating QALYs in cost-utility studies involving patients with advanced HIV disease.