Establishing disability weights from pairwise comparisons for a US burden of disease study.

摘要: To determine valid and reliable disability weights for a U.S. burden of disease study, convenience sample 68 clinical experts was recruited, including representatives from over 20 NIH institutes Centers Disease Control Prevention. Experts were given various health state valuation tasks pairwise comparison, ranking, Person Trade Off. Materials consisted standardized descriptions 11 attributes per (Classification Measurement System Functional Health, CLAMES). Attributes comprised up to 5 ordinal levels disability. All states displayed either with or without labels. Health taken an existing comprehensive Canadian system. Conditional Logistic (CLR) Probit Regression (PR) used derive weights. CLR PR converged in yielding stable regression construct weights, correlation 0.816. The overall test-retest reliability amounted 92.5% identical decisions. No significant difference found the presentation A comparison expert valuations our study standard gamble based general population resulted agreement r = 0.61. chosen methodology yielded As it is on modularized set attributes, this will allow derivation basis using CLAMES.