DO DIFFERENCES IN PREFERENCE-BASED HRQOL SCORES FOR COMMON CHRONIC CONDITIONS IN THE U.S. DIFFER BY GENDER OR RACE?

PURPOSE: To examine if race or gender are associated with the impact of common chronic health conditions as measured by 3 preference-based health-related quality of life (HRQoL) indexes.
METHODS: We analyzed 20,748 non-proxy respondents ages 18 and older from 2003 Medical Expenditure Panel Survey and used sociodemographic data (gender, age, race, ethnicity, education, income) and 11 chronic conditions (diabetes, asthma, high blood pressure, coronary heart disease, angina, myocardial infarction, other heart disease, stroke, emphysema, joint pain, arthritis).  Three summary scores were derived from EQ-5D and SF-12v2^TM questionnaires (EQ-5D using U.K. and U.S. weights and SF-6D). Average differences in utility between people with and without health conditions (“disutility”) were estimated using weighed least square regression in SAS 9.1. Each of the 3 summary scores was regressed on one health condition at a time, gender and an interaction of gender with the health condition to produce unadjusted disutility and to test the impact of gender on disutility. The same models were estimated replacing gender with race (Black, White).  To test whether gender or race have an impact on adjusted distutilities, we estimated models for each summary score with all health conditions and sociodemographic variables included in the model and interactions added one at a time.
RESULTS: As measured by the interactions, men had significantly (p<0.05) smaller disutilities — i.e., have smaller utility decrements — than women for asthma, myocardial infarction and joint pain by 0.025, 0.055, 0.02 on average across HRQoL indexes, respectively. After adjustments, disutilities differed significantly by gender for high blood pressure and myocardial infarction (averaged estimates -0.009 and 0.038). Unadjusted disutilities were significantly smaller for whites than for blacks for asthma, high blood pressure, angina, other heart disease, stroke, joint pain and arthritis with averaged estimates ranging from -0.109 to -0.014. Adjusted disutilites differed significantly between blacks and whites for asthma, angina, other heart disease and stroke with averaged estimates of -0.025, -0.068, -0.028 and -0.088.  The magnitude and statistical significance of interaction effects varied by summary score.  
CONCLUSION: Disutilities differed significantly by gender and race for some conditions. Significant differences occurred for more conditions between blacks and whites than between men and women.  It may be important to use gender- and race-specific disutility values for certain health conditions when calculating QALYs for cost-effectiveness analyses.