Use of generic and condition-specific measures in NICE decision-making

Health-care resources are limited and there is a need to ensure these are used efficiently. The National Institute for Health and Clinical Excellence (NICE) issues guidance to the NHS on the prevention of ill health and the promotion of good health. Part of its work includes an assessment of whether improvements to health as a result of new health technologies (such as drugs) or programmes (such as to reduce the prevalence of smoking) are worth the additional resources they require. NICE?s preferred measure for improvements in health is the quality adjusted life year or ?QALY?. This is a measure of length of life weighted by a value representing the quality of that life. Assessments of the impact of treatments on quality of life are usually based on generic measures of health-related quality of life in the evidence provided to NICE. These generic measures show the impact of treatments on a broad range of dimensions of health which allows the same measure to be used for all health conditions. They also incorporate values from the general population about what different levels or severities of health would mean to them. However, these measures of health-related quality of life are not always included in clinical trials and they have also been criticised for not reflecting changes in health in some conditions. This project will draw upon existing research to use measures of health-related quality of life that have been developed for specific health conditions within the analyses recommended by NICE. It will establish whether the generic measures of health related quality of life are inappropriate for some conditions (such as cancer). It will also look at establishing the relationship between quality of life measures that are generic with those that are specific to conditions. The project will establish methods for extending the existing generic measures to better reflect specific health conditions or diseases.

The National Institute for Health and Clinical Excellence (NICE) recommends the use of generic-preference based measures for measuring health outcomes in its economic evaluations, specifically the EQ-5D. However, data using generic measures may not have been collected within clinical studies or the generic measures may not be suitable for measuring changes in health for specific conditions or treatments. The first stage of the study will be to critically review the use and performance of three widely used generic-preference based measures (EQ-5D, Health Utilities Index [HUI] 3, and Short-Form [SF]-6D) in four key areas of health: visual impairment, aural impairment, cancer and skin conditions. The assessment of the validity of preference-based measures of health is fraught with conceptual and empirical problems owing to the lack of a gold standard. The approach adopted here will distinguish the validity of the descriptive system from that of the preference-based index. It will examine the descriptive validity of the measures in terms of content, face and construct validity using both qualitative and quantitative evidence. The empirical validity of the index will be based on the convergence with other measures of stated preferences. The second stage of the study will be to estimate three or more mapping functions to predict EQ-5D preference scores from condition-specific and clinical measures. Mapping techniques have previously been used to estimate health-state utilities in NICE submissions; however the additional uncertainty in the predicted values arising from the mapping process is not incorporated in the analyses. Exploratory work into the appropriate methods for incorporating uncertainty in the predicted values into practical analyses will be conducted. The third stage of the project will be to develop new preference based measures of health-related quality of life by extending the existing EQ-5D measure. Based on the results of the review in stage 1, additional dimensions will be added to the EQ-5D and valued in a sample of the generic public using the time trade-off method. The new ?add-on? instrument that is found to add the most additional information will be selected for further study in order to develop a valuation system and to generate a set of methods that can be used by others when considering expanding the descriptive system of a generic instrument.