Abstract

We assessed the contribution to measurement properties of weighting the items of a generic quality of life (QOL) index. Data were acquired from an observational study on QOL of 675 patients with end-stage renal disease (ESRD). As an outcome measure we used the Spitzer QL-index. We compared scores on this generic index to those constructed using weighting schemes based on personal rating of each of the items included in the generic index. We found that the weighted schemes were not superior to the nonweighted with regard to: (a) Their ability to discriminate between diseased and nondiseased patients with five major comorbidities, and among patients reporting five different categories of global assessment of QOL, (b) Their responsiveness, i.e., identifying changes in QOL over a year and (c) Their predictive ability for one year mortality. We conclude that weighting by personal rating added no marginal benefit to a generic, nonweighted QOL instrument in ESRD patients.