Assessing the performance of a new generic measure of health-related quality of life for children and refining it for use in health state valuation.

BACKGROUND Previous research to develop a new generic paediatric health-related quality of life (HR-QOL) measure generated 11 dimensions of HR-QOL, covering physical, emotional and social functioning. These dimensions and their response scales were developed from interviews with children. Some of these dimensions have alternative wording choices. The measure is intended to be preference based so that it can be used in paediatric economic evaluation. OBJECTIVES The aims of this research were to assess the performance of this new descriptive system in a general and clinical paediatric population, to determine the most appropriate wording for the dimensions and to refine the descriptive system to be amenable to health state valuation to make it suitable for use in economic evaluation. METHODS A sample of 247 children was recruited from general and clinical paediatric populations. Each child completed the descriptive system and data were collected to allow assessment of practicality (including response rates, completion rates and time to complete), content, face and construct validity, whether the child could self-complete and preferences for alternative wordings that could be used for dimensions. These data were used to inform the final choice of wording for dimensions, the scales used for each dimension and the reduction of dimensions to meet the constraints of health state valuation. RESULTS The descriptive system demonstrated good practicality and validity in both the general and clinical paediatric samples. The completion rates were excellent (>98%), the mean time to complete was low (3.8 minutes for the general and 5.3 minutes for the clinical sample) and there was evidence of face, content and construct validity. The descriptive system was able to demonstrate significant differences between the general and clinical samples and according to the level of health of children. 96% of the school sample and 85% of the clinical sample were able to self-complete. The final choice of wording for the 11 dimensions was determined by the preferences and comments of the children. To make it amenable for health state valuation, the number of dimensions was reduced from 11 to 9 by removing the dimensions 'jealous' and 'embarrassed'. CONCLUSIONS The descriptive system performed well in both the general and the clinical populations, and the final descriptive system generates health states that are feasible for health state valuation. Further research is needed to value the final descriptive system by obtaining preference weights for each health state, thereby making the measure suitable for use in paediatric economic evaluation.