How to include relatives and productivity loss in a cost‐effectiveness analysis

• Main Author: Davidson, Thomas
• Format: Doctoral Thesis
• Language: English
• Published: Linköpings universitet, Utvärdering och hälsoekonomi 2009
• Subjects: Economics and Business; Ekonomi och näringsliv
• Online Access: http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-16999; http://nbn-resolving.de/urn:isbn:978-91-7393-693-4

Health economic evaluations are today commonly used in the decision‐making process in health care. Within the field of cost‐effectiveness analysis (CEA), there are several methodological and empirical issues that cause debate about what is included in the analysis. This thesis covers two such issues; costs and effects for relatives, and the valuation of individuals’ productivity loss due to morbidity. The objective of the thesis is to provide further knowledge about what should be included in CEAs which take a societal approach. The papers that the thesis is based on, four in total, examine the theoretical aspects of the studied issues and test these aspects empirically. Three different data materials were used. The CEA and the estimation of costs and effects are central in all the papers. The outcome measure used is quality‐adjusted life years (QALYs). The relatives of an individual with a disease or disability often provide informal care, and there may also be concomitant effect on their own well‐being. Nevertheless, the costs and effects for the relatives are generally excluded from CEAs, and there are few guidelines for how to include relatives’ effects. This thesis suggests the use of a new measure, R‐QALYs, which can be used both to visualise relatives’ effects and to include them in the analysis. We found that while the EQ‐5D instrument can be used to capture some of the relatives’ effects, it most likely misses a number of important attributes, for example altruistic preferences. Methods of eliciting R‐QALY weights include direct valuation methods and indirect methods, using existing relativerelated instruments. However, none of these methods are without difficulties, and there is a need for more studies on estimating valid relatives’ effects. Another possible approach with high potential is to use monetary measurements for both the costs and effects relevant to relatives. The results also show that income affects the QALY weights if the individuals include the utility generated by consumption within their QALY weights. The empirical tests showed that a majority of individuals do not consider their own income when they value health states. An explicit instruction to take income into account seemed to affect the valuation of those health states that were assumed to have consequences on the ability to perform daily activities. These findings give support for including the productivity costs caused by morbidity in the analysis; as these costs are not, or are only to a minor extent, implicitly incorporated in individuals’ QALY weights. The loss of leisure time, however, is captured in the QALY weight, and care must be taken to avoid double counting this loss in the analysis. The results of CEAs will only be partial if relatives’ costs and effects and the costs of individuals’ productivity loss are excluded for health interventions where they are assumed to be of significant importance.