Mapping EQ-5D Utility Scores from the Incontinence Quality of Life Questionnaire among Patients with Neurogenic and Idiopathic Overactive Bladder

• Published in: Value in health 2013-03, Vol.16 (2), p.394-402
• Main Authors: Kay, Stephen, MSc, Tolley, Keith, BA, MPhil, Colayco, Danielle, PharmD, MS, Khalaf, Kristin, PharmD, MS, Anderson, Peter, BSc, Globe, Denise, PhD
• Format: Article
• Language: English
• Subjects: Algorithms; Biological and medical sciences; Cross-Sectional Studies; cross-walk; EQ-5D; Female; France; Germany; Humans; I-QOL; Idiopathic; Internal Medicine; Male; Mapping; Medical sciences; Middle Aged; Miscellaneous; Nephrology. Urinary tract diseases; Nonlinear; Psychometrics - instrumentation; Public health. Hygiene; Public health. Hygiene-occupational medicine; Quality of life; Quality-Adjusted Life Years; real world; Reproducibility of Results; Sickness Impact Profile; Spain; Surveys and Questionnaires; United Kingdom; United States; Urinary Bladder, Overactive - psychology; Urinary Incontinence - psychology; Urinary system involvement in other diseases. Miscellaneous; Urinary tract. Prostate gland; utility values
• ISSN: 1098-3015; 1524-4733
• DOI: 10.1016/j.jval.2012.12.005

Abstract Objectives To provide a mapping algorithm for estimating EuroQol five-dimensional (EQ-5D) questionnaire index scores from the Incontinence-specific Quality of Life questionnaire (I-QOL) based on nationally representative samples of patients with idiopathic or neurogenic overactive bladder (OAB) using EQ-5D questionnaire preference valuations based on both the UK and US general populations. Methods Analyses were conducted for 2505 patients from the Adelphi Overactive Bladder Disease Specific Programme, a cross-sectional study of patients with idiopathic or neurogenic OAB, undertaken in the United States and Europe in 2010. A range of statistical modeling techniques was used. Tenfold cross-validation techniques were used to calculate mean absolute error (MAE) and root mean squared error (RMSE) goodness-of-fit statistics. Various predictor lists, together with a method combining stepwise selection with multivariable fractional polynomial techniques to allow nonlinear relationships to feature, were pursued. Results Choice of predictors was consistent for both the UK and US EQ-5D questionnaire tariffs. For idiopathic, the best model included the I-QOL total score and age (both modeled nonlinearly.) For neurogenic, the best model was the I-QOL social embarrassment domain score modeled linearly only. Best-fit results were better in the idiopathic (n = 2351; MAE = 0.10; RMSE = 0.14) than in the neurogenic sample (n = 254; MAE = 0.17; RMSE = 0.22). Conclusions This research provides algorithms for mapping EQ-5D questionnaire index scores from the I-QOL, allowing calculation of appropriate preference-based health-related quality-of-life scores for use in cost-effectiveness analyses when only I-QOL data are available. The strongest results were for idiopathic patients, but those for neurogenic are consistent with those of other published mapping studies.