Abstract
BACKGROUND: Many publicly-funded health systems apply cost-benefit frameworks in response to the moral dilemma of how best to allocate scarce healthcare resources. However, implementation of recommendations based on costs and benefit calculations and subsequent challenges have led to 'special cases' with certain types of health benefits considered more valuable than others. Recent debate and research has focused on the relative value of life extensions for people with terminal illnesses. This research investigates societal perspectives in relation to this issue, in the UK. METHODS: Q methodology was used to elicit societal perspectives from a purposively selected sample of data-rich respondents. Participants ranked 49 statements of opinion (developed for this study), onto a grid, according to level of agreement. These 'Q sorts' were followed by brief interviews. Factor analysis was used to identify shared points of view (patterns of similarity between individuals' Q sorts). RESULTS: Analysis produced a three factor solution. These rich, shared accounts can be broadly summarised as: i) 'A population perspective - value for money, no special cases', ii) 'Life is precious - valuing life-extension and patient choice', iii) 'Valuing wider benefits and opportunity cost - the quality of life and death'. From the factor descriptions it is clear that the main philosophical positions that have long dominated debates on the just allocation of resources have a basis in public opinion. CONCLUSIONS: The existence of certain moral positions in the views of society does not ethically imply, and pragmatically cannot mean, that all are translated into policy. Our findings highlight normative tensions and the importance of critically engaging with these normative issues (in addition to the current focus on a procedural justice approach to health policy). Future research should focus on i) the extent to which these perspectives are supported in society, ii) how respondents' perspectives relate to specific resource allocation questions, and iii) the characteristics of respondents associated with each perspective.