Individual quality of life: adaptive conjoint analysis as an alternative for direct weighting?

In the schedule for the evaluation of individual quality of life (SEIQoL) the weights for five individualized quality of life domains have been derived by judgment analysis and direct weighting (DW). We studied the feasibility and validity of adaptive conjoint analysis (ACA) as an alternative method to derive weights in 27 cancer patients and 20 patients with rheumatoid arthritis. Further, we assessed the convergence between direct weights and weights derived by ACA, and their correlation with global quality-of-life scores. All respondents finished the ACA task, but one in five respondents were upset about the ACA task. Further, the task was vulnerable to judgment 'errors', such as inconsistent answers. The agreement between the two weights was low. Both weighted index scores were strongly correlated to the unweighted index score. The relationships between the index score and scores on a visual analogue scale for global individual quality of life and global quality of life were similar whether or not the index score was calculated with DW weights, with ACA weights, or without using weights. We conclude that, because weights did not improve the correlation between the index score and global quality of life scores, it seems sufficient to use the unweighted index score as a measure for global individual quality of life.