Electronic supplementary material
The online version of this article (doi:10.1186/1472-6963-12-94) contains supplementary material, which is available to authorized users.
Some of the data used was from a study funded by the Department of Health, to which the authors contributed. The authors work for this paper was independent of the DH. RC was a member of NICE advisory committees from 2002-2009, RF has contributed as an expert advisor to a NICE methods review, and NS currently sits on a NICE advisory committee.
RF conceived the idea and designed the study, with supervision from RC and AH as part of his MD thesis. RC had a specific role in advising on the health economic and statistical aspects of the paper. RF wrote up the first draft of the study. All authors contributed to data analysis, contributed to revisions of the manuscript and approved the final draft. All authors read and approved the final manuscript.
The General Medical Services primary care contract for the United Kingdom financially rewards performance in 19 clinical areas, through the Quality and Outcomes Framework. Little is known about how best to determine the size of financial incentives in pay for performance schemes. Our aim was to test the hypothesis that performance indicators with larger population health benefits receive larger financial incentives.
We performed cross sectional analyses to quantify associations between the size of financial incentives and expected health gain in the 2004 and 2006 versions of the Quality and Outcomes Framework. We used non-parametric two-sided Spearman rank correlation tests. Health gain was measured in expected lives saved in one year and in quality adjusted life years. For each quality indicator in an average sized general practice we tested for associations first, between the marginal increase in payment and the health gain resulting from a one percent point improvement in performance and second, between total payment and the health gain at the performance threshold for maximum payment.
Evidence for lives saved or quality adjusted life years gained was found for 28 indicators accounting for 41% of the total incentive payments. No statistically significant associations were found between the expected health gain and incentive gained from a marginal 1% increase in performance in either the 2004 or 2006 version of the Quality and Outcomes Framework. In addition no associations were found between the size of financial payment for achievement of an indicator and the expected health gain at the performance threshold for maximum payment measured in lives saved or quality adjusted life years.
In this subgroup of indicators the financial incentives were not aligned to maximise health gain. This disconnection between incentive and expected health gain risks supporting clinical activities that are only marginally effective, at the expense of more effective activities receiving lower incentives. When designing pay for performance programmes decisions about the size of the financial incentive attached to an indicator should be informed by information on the health gain to be expected from that indicator.