OBJECTIVES : Interstitial cystitis/bladder pain syndrome (IC/BPS) is a chronic, often debilitating, condition of urinary frequency, urgency, and pelvic pain and has a major impact on quality of life. Many Health Technology Assessment bodies require preference-based health-related quality of life (HRQoL) data to calculate quality adjusted life years (QALYs) and assess cost-effectiveness. Given a lack of utility data for these patients, we aimed to develop a mapping algorithm to predict health-related utility values from more widely used condition-specific questionnaires.

METHODS : An online survey was developed and administered to BPS patients in the UK, through two charities. The survey included EQ-5D and condition-specific questionnaires, including O’Leary Sant Interstitial Cystitis Symptom Index (ICSI) and Problem Index (ICPI), validated symptom scales used in BPS. The study was designed to predict EQ-5D values from ICSI and ICPI responses and sociodemographic/disease-related variables. Direct and indirect mapping approaches were used and a range of models tested. Model selection was based on model fit. Cross-validation was performed by comparing predicted and observed values. K-fold and Leave-one-out cross-validations were conducted to test the model stability to resampling.

RESULTS : 252 patients completed the survey. The best performing model was the two-part model using overall ICSI scores, age and recent experience of bladder instillation. The first part of the model focussed on the probability of full health based on ICSI score, bladder instillation dummies and age groups using logistic regression, and the second part estimated the utility function using OLS for those not in full health.

CONCLUSIONS : A mapping function to predict EQ-5D utility values from ICSI and ICPI scores in patients with BPS was successfully developed. This can be used to generate EQ-5D values for economic evaluations of BPS treatments when EQ-5D has not been directly collected in trials. Further validation in external datasets would be beneficial.