Estimating importance weights for the IWQOL-Lite using conjoint analysis.

PURPOSE: Our objective was to estimate preference-based weights for the IWQOL-Lite that reflect the relative importance overweight and obese people place on the domains included in the instrument. METHODS: US residents, 18 years of age or older, who are overweight (BMI=25.0-29.9) or obese (BMI>or=30) completed an online survey instrument consisting of 12 choice questions. The survey included eight attributes: problems doing usual daily activities, physical symptoms, worry about health, low self-esteem, sexual problems, problems moving around or sitting in public places, teasing or discrimination, and problems at work. Ordered probit was used to estimate importance weights for the attributes and levels. RESULTS: Five hundred and two subjects completed the survey. Sexual problem was the most important attribute. The remaining attributes can be ranked on the order of importance as follows: low self-esteem, physical symptoms, daily activities, teasing or discrimination, moving around or sitting, problems at work, and worry about health. CONCLUSIONS: Our results confirm previous findings that weighting the individual items in the IWQOL-Lite by the importance of outcomes to overweight and obese subjects may provide a more meaningful evaluation of the effect of changes in weight on patient well-being than a nonpreference-based measure of HRQOL.

Duke Scholars

Author F. Reed Johnson Population Health Sciences