Construct validity for the Health Utilities Index in a sleep center

Abstract

We utilized well-established and validated quality of life (QoL) measures to test the validity of the Health Utilities Index (HUI) in patients with untreated sleep disorders in a cross-sectional study. We submitted a 66-question survey to 86 patients presenting to the University of Maryland Sleep Disorders Center (68 with sleep apnea, 18 with other sleep disorders). HUI mark 2 (HUI2) and HUI mark 3 (HUI3) were compared with other indices of QoL: medical short form 12 (SF-12), Epworth Sleepiness Scale (ESS), and Functional Outcomes of Sleepiness Questionnaire (FOSQ). The participants of this study were eighty-six patients presenting with untreated sleep disorders to the University Medical Center. There were no interventions. The mean age in our patients was 50.7 +/- 14.2 years with a mean body mass index of 32.9 +/- 7.6. For patients with sleep apnea, the mean respiratory disturbance index was 32.6 +/- 29.1. For all patients, the mean HUI2 score was 0.73 + 0.23, and the mean HUI3 score was 0.6 +/- 0.35 (p = 0.0066 for difference from HUI2). The mean SF-12 physical components (PC) score was 43.6 +/- 12.1, and the mean SF-12 mental components (MC) score was 49.8 identical with 40.1. The mean ESS score was 10.5 + 5.2, and the mean FOSQ score was 16.4 +/- 3.5. Significant bivariate correlations were found between HUI2 and HUI3 as well as between both of these and age, SF-12, SF-12 PC score, ESS, and FOSQ. Stepwise multivariate regression revealed significant independent correlations between HUI2 and age, SF-12 PC and the FOSQ, and between HUI3, SF-12 PC, and the FOSQ. There were no differences in HUI scores between obstructive sleep apnea (OSA) and non-OSA patients: (1) HUI3 scores are lower than HUI2 scores, (2) both HUI2 and HUI3 are correlated with other indices of QoL in sleep patients, and (3) the HUI surveys should be considered in QoL and economic impact studies in patients with sleep disorders.