Mapping the Paediatric Quality of Life Inventory (PedsQL TM ) Generic Core Scales onto the Child Health Utility Index –9 Dimension (CHU-9D) Score for Economic Evaluation in Children Tosin Lambe ' • Emma Frew ' • Natalie J. Ives 3 • Rebecca L. Woolley 3 • Carole Cummins 4 • Elizabeth A. Brettell 3 • Emma N. Barsoum 3 • Nicholas J. A. Webb 2 • On behalf of the PREDNOS Trial Team 1 Health Economics Unit, Institute of Applied Health Research, University of Birmingham, Edgbaston, Birmingham, UK 2 Department of Paediatric Nephrology and NIHR Manchester Clinical Research Facility, Manchester Academic Health Science Centre, Royal Manchester Children’s Hospital, Manchester, UK 3 Birmingham Clinical Trials Unit, University of Birmingham, Birmingham, UK 4 Institute of Applied Health Research, University of Birmingham, Birmingham, UK Abstract Background The Paediatric Quality of Life Inventory (PedsQL TM ) questionnaire is a widely used, generic instrument designed for measuring health-related quality of life (HRQoL); however, it is not preference-based and therefore not suitable for cost–utility analysis. The Child Health Utility Index–9 Dimension (CHU-9D), however, is a preference-based instrument that has been primarily developed to support cost–utility analysis. Objective This paper presents a method for estimating CHU-9D index scores from responses to the PedsQL TM using data from a randomised controlled trial of prednisolone therapy for treatment of childhood corticosteroidsensitive nephrotic syndrome. Methods HRQoL data were collected from children at randomisation, week 16, and months 12, 18, 24, 36 and 48. Observations on children aged 5 years and older were pooled across all data collection timepoints and were then randomised into an estimation (n = 279) and validation (n = 284) sample. A number of models were developed using the estimation data before internal validation. The best model was chosen using multi-stage selection criteria. Results Most of the models developed accurately predicted the CHU-9D mean index score. The best performing model was a generalised linear model (mean absolute error = 0.0408; mean square error = 0.0035). The proportion of index scores deviating from the observed scores by< 0.03 was 53%. Conclusions The mapping algorithm provides an empirical tool for estimating CHU-9D index scores and for conducting cost –utility analyses within clinical studies that have only collected PedsQL TM data. It is valid for children aged 5 years or older. Caution should be exercised when using this with children younger than 5 years, older adolescents (> 13 years) or patient groups with particularly poor quality of life.