Happy with your capabilities? Valuing ICECAP-O and ICECAP-A sets based on experienced utility using subjective well-being data
The allocation of scarce health care resources is an important and difficult task for health care decision makers. In that context, the costs and benefits of competing health care interventions are increasingly compared with each other. Typically, such comparisons are supported by health technology assessment, with an important role for economic evaluations. In the health care decision-making context, the latter often takes the form of a cost-utility analysis in which costs are expressed in monetary terms whereas benefits are expressed in terms of quality-adjusted life-years (QALYs). Health-related quality of life is commonly measured by generic multidimensional instruments such as the EQ-5D. However, it has been questioned whether maximizing health, as captured in QALYs, is an appropriate representation of society’s values concerning health care or the appropriate objective in all areas of health care. The benefits of health care in many situations are not limited to health alone. In palliative and elderly care, for example, health improvement might not even represent the (primary) aim of the interventions. Interventions in these areas may be targeted at increasing well-being rather than health. This implies that (part of) the benefits of interventions may not be appropriately captured when using traditional health-related quality-of-life measures. This article describes the development of value sets for ICECAP-O and ICECAP-A based on experienced utility and compares them with current decision utility weights. Data from 2 cross-sectional samples corresponding to the target groups of ICECAP-O and ICECAP-A were used in 2 separate analyses. The utility impacts of ICECAP-O and ICECAP-A levels were assessed through regression models using a composite measure of subjective well-being as a proxy for experienced utility. The observed utility impacts were rescaled to match the 0 to 1 range of the existing value set. The capability instruments ICECAP-O and ICECAP-A have the potential to broaden the evaluative space of economic evaluations of health care interventions. Levels and dimensions of instruments such as the ICECAP-O and ICECAP-A have to be weighted to determine a single utility score that can be used as a measure of benefit in cost-utility analyses. These weights should ideally reflect what matters most to people and can be based on decision utility or experienced utility. This choice is not neutral, as resulting values sets can differ, as they do here. Although tariffs based on decision utility are available for ICECAP-O and ICECAP-A, this was not yet the case for experienced utility. Therefore, we developed these by directly assessing well-being capability values based on their impact on SWB using regression analysis, interpreting life satisfaction as (a proxy of) experienced utility. This is different from approaches often taken in the related literature on self-rated experienced health because of the broader nature of the ICECAP-instruments. To our knowledge, our study was the first to analyze the differences between valuations of capability states based on ex ante decision utility and experienced utility. The existing literature on using the latter to value health states, namely, of the EQ-5D-3L and SF-6D, shows that the estimates of the impact of specific dimensions can differ substantially between the 2 approaches, especially for mental health problems (e.g. EQ-5D dimensions anxiety and depression). Our analysis showed that calculating value sets for the ICECAP-O and ICECAP-A instruments based on experienced utility using SWB data is feasible and that the obtained weights to some extent differ from the weights previously obtained based on decision utility. This difference generates insights for policy makers in the context of the application of ICECAP-O and ICECAP-A as well as experienced and decision utility in economic evaluations.
Keywords
health, ICECAP-O, ICECAP-A