International prostate symptom score should be considered a complement rather than a substitute to generic preference-based measures for measuring lower urinary tract symptoms within economic evaluation

Abstract

Objective: To determine the suitability of using the International Prostate Symptom Score (IPSS), a condition-specific instrument, within economic evaluation of lower urinary tract symptom (LUTS) interventions. Methods: Data were obtained from a cohort of 2236 community-dwelling Australian men. Using correlations and a modified Bland–Altman plot, we investigated the convergent validity of the IPSS with two generic quality of life (GQol) instruments: the Assessment of Quality of Life 4 Dimensions (AQoL-4D) and the Short-Form 36 (SF-36). Discriminant validity was investigated using Wilcoxon–Mann–Whitney and Kruskal–Wallis tests, comparing instrument scores and utilities between subgroups varying in age, marital status, history of illness, smoking status and self-assessed general health. Results: Discriminant validity was confirmed for all instruments by finding statistically significant differences in summary scores and utilities between nearly all subgroups. Convergent validity between the summary scores/utilities of the IPSS and the GQoL instrumentswas weak (absolute correlation value range, 0.11–0.23) but differed considerably between dimensions of the instruments (absolute correlation value range, 0.01–0.24). Weak to moderate correlation between the GQoL instruments was seen (absolute correlation value range, 0.01–0.49). Conclusion: Our findings suggest that the IPSS has comparable discriminant validity to the GQoL instruments and therefore useful for assessing subgroup differences related to urinary symptoms. The weak convergence between the IPSS and the GQoL instruments however suggests that, within economic evaluation, the IPSS should be viewed as a complement rather than a substitute to the GQoL instruments because it captures different quality of life constructs.