The value of thinly spread QALYs.

A number of recent findings from the literature imply that the value of a QALY varies depending on the concentration or dispersion of that QALY over treated individuals. Given that funding decisions are currently made under either the assumption of distributive-neutrality or some combination of explicit decision criteria and implicit adjustment for distributional concerns, it is likely that substantial social welfare gains are available if distributional objectives could be more accurately reflected in funding decisions. This paper considers three alternative approaches to explicitly adjust for distributional concerns with regard to the concentration or dispersion of individual health gains. Including non-health arguments in the objective function by 'weighting' QALYs for distributional effects or imposing differential funding thresholds for interventions with different distributional effects might be considered first- and second-best solutions, and would likely deliver the greatest social welfare gains. However, there is some doubt that first- or second-best solutions would be: (i) feasible given current data gaps; and (ii) politically acceptable. Rather, a simple and transparent approach is suggested wherein the sponsors of interventions that deliver health gains that are of questionable 'welfare-significance' for the treated individual would be required to provide decision-makers with an estimate of willingness to pay for the QALYs in question and would only be eligible for funding in the event that the positive net present value criterion is met.