Social capital appears to be an important determinant of health production and health utilization and demand. However, there is limited evidence on the mechanisms underlying this relationship. In this article we draw on the evidence and insights reported in this special issue along with findings from the economic and other social science literature to develop a discussion on the explanations of the likely (behavioural) mechanisms that underpin the connection between social capital and health. An important and under-explored influence mediating the relationship between social capital and health (behaviour) lies in the ‘social formation of health preferences and constraints’ individuals face in determining their life-styles and in using health care. In particular, we point to the interdependence in how individuals in the first place perceive and also respond to common health risks and the role of cultural transmission and social identity as conveyors of this process. We argue that an emerging body of evidence suggesting that interdependent preferences influence health calls for further re-formulation of traditional demand for and production of health models. Additionally, methodological problems are highlighted and possible ways forward suggested.

This comment on Paul Dolan's essay, ‘Developing methods that really do value the “Q” in the QALY,’ seconds his critique of current preference-based methods of valuing health states but disputes both his assumption that health states should be appraised by their bearing on well-being and his conclusion that the bearing of health states on well-being should be measured in terms of subjective experience. This comment defends instead the view that health states should be valued in terms of the range of activities that they permit.

There are a number of substantial problems with using decision-based utility measures such as the time trade off and standard gamble to value improvements in health. Dolan (this issue) argues that because of these problems, it would be better to use measures of real experiences (subjective well being). We review evidence that supports criticisms of decision-based utility measures, whether provided by patients or non-patients. But we also review a number of substantial problems with currently used measures of subjective well-being, and point out that there is no definitive evidence that they represent any improvement over decision utility measures. We conclude with a call for expanded research into developing new tools for quantifying health-related quality of life that are more valid, more sensitive to changes in health status, and less biased.