Estimating preference-based health utilities index mark 3 utility scores for childhood conditions in England and Scotland
Petrou S., Kupek E.
Background. A common feature of studies that have compiled lists or catalogues of preference-based health-related quality-of-life weights for inclusion within quality-adjusted life years (QALYs) is their focus upon adult populations. More generally, utility measurement in or on behalf of children has been constrained by a number of methodological concerns. Objective. To augment previous catalogues of preference-based health-related quality-of-life weights by estimating preference-based Health Utilities Index Mark 3 (HUI3) multiattribute utility scores associated with a wide range of childhood conditions. Methods. Data for 2236 children from the ''Disability Survey 2000: Survey of Young People With a Disability and Sport'' formed the basis of this investigation. Ordinary least squares (OLS), Tobit, and censored least absolute deviations (CLAD) regression methods were used to estimate adjusted marginal disutilities for each condition from 2 thresholds: 1) a threshold of 1.0 representing perfect health and 2) a normative childhood utility threshold. Results. Prespecified statistical tests indicated a preference for the OLS regression model over the Tobit and CLAD models. The unadjusted mean, median, 25th percentile and 75th percentile HUI3 multiattribute utility scores and adjusted marginal disutilities are presented for 43 conditions. Notably, based on the OLS estimator, the adjusted marginal disutilities for hydrocephalus; learning and physical disabilities; other syndromes and associations; meningitis, encephalitis, and other infections of the central nervous system; and microcephaly were estimated at -0.889 (95% confidence interval [CI]: -0.727, -1.000), -0.858 (95% CI: -0.727, -0.989), -0.838 (95% CI: -0.668, -1.000), -0.826 (95% CI: -0.677, -0.975), and -0.820 (95% CI: -0.670, -0.970), respectively, when a perfect health threshold was applied, and -0.814 (95% CI: -0.656, -0.979), -0.783 (95% CI: -0.656, -0.918), -0.763 (95% CI: -0.597, -0.937), -0.751 (95% CI: -0.606, -0.904), and -0.745 (95% CI: -0.598, -0.899), respectively, when a ormative childhood utility threshold was applied. Conclusion. Our estimates and their associated distributions can be used for the purposes of QALY estimation by analysts conducting economic evaluations within the childhood context.