Mode of Administration Is Important in US National Estimates of Health-Related Quality of Life

Background: It is unknown if different national surveys that vary in mode of administration yield similar national averages for healthrelated quality of life (HRQoL). Purpose: Examine HRQoL scores from 4 surveys representative of the noninstitutionalized US adult population for patterns related to age, gender, and mode of administration. Methods: We use data from the Joint Canada/United States Survey of Health (JCUSH; telephone survey), 2002 Medical Expenditure Panel Survey (MEPS; mail survey), National Health Measurement Study (NHMS; telephone survey), and US Valuation of the EuroQol EQ-5D Health States Survey (USVEQ; self-administered with interviewer present). We compare estimates from the EQ-5D, Visual Analog Scale, Health Utilities Index Mark 3, and general self-rated health stratified by age and gender. Scores were also regressed on age and gender within each survey and in a pooled analysis. Results: We used 4939 subjects from JCUSH, 23,006 from MEPS, 3844 from NHMS, and 3878 from USVEQ. The majority of age and gender strata had instrument completion rates above 85%. Age- and gender-stratified estimates of HRQoL scores tended to be consistent when mode of administration (self- or interviewer-administered) was the same. Telephone administration yielded more positive HRQoL estimates than self-administration in older age groups. Older age groups and females reported lower HRQoL than younger age groups and males regardless of mode of administration. Conclusions: When choosing survey-collected HRQoL scores for comparative purposes, analysts need to take mode of administration into account.