Pay-for-performance and primary care physicians: lessons from the UK Quality and Outcomes Framework for local incentive schemes

Abstract

Primary care has long been recognised as being central to the delivery of effective, efficient, equitable and safe care necessary for improved population health; however, it is vulnerable to variation in performance across providers. In an attempt to reduce this variation a pioneering national pay-for-performance scheme was implemented in the UK in 2004. The aim of the Quality and Outcomes Framework was to incentivise practices to deliver structured care orientated to the achievement of evidence-based quality targets for patients with predominantly chronic conditions. The ‘Next steps towards primary care co-commissioning’ policy document signals a move towards greater devolution of the development of pay-for-performance schemes to Clinical Commissioning Groups by giving them the freedom to develop alternatives to the national Quality and Outcomes Framework. Any Clinical Commissioning Group-developed incentive schemes will need to ‘ . . . be able to demonstrate improved outcomes, reduced inequalities and value for money’ (p. 14). At this point, it is worth reflecting upon the extent to which the Quality and Outcomes Framework demonstrates these characteristics; and it also raises two critical questions. First, are there any key lessons from the national Quality and Outcomes Framework for local scheme developers? And second, what is the future of the Quality and Outcomes Framework as a national framework? First, the impact of the Quality and Outcomes Framework upon improved outcomes and quality of care. There is some observational evidence of modest short-term improvements in quality of care demonstrated through a reduction in the variation of practice performance. The effects have been on process of care indicators, and real hard outcome data on mortality are lacking even though one can argue that improved management or reduction of process indicators such as hypertension and smoking have contributed to reduced ischaemic heart disease mortality. There is also some evidence that the Quality and Outcomes Framework is having an impact upon patient outcomes in terms of reductions in emergency admissions related to incentivised conditions. Second, the impact upon inequalities. Observational data show that there has been a narrowing of achievement between practices in the most and least deprived quintiles which does not appear to be reliant upon exception reporting or the ability to exclude patients from achievement calculations. Despite the ongoing political concerns about the ‘gaming’ of exception reporting to maximise practice income, its use has not been widespread. Rates of exception reporting are also comparable across deprivation quintiles suggesting that the Quality and Outcomes Framework, to some extent, has been an equitable intervention. However, exception reporting rates are variable across both indicators and practices, so the possibility remains that this may impact disproportionately upon the more marginalised and vulnerable population groups. Third, is there evidence that the Quality and Outcomes Framework demonstrates value for money? The National Audit Office defines value for money as ‘the optimal use of resources to achieve the intended outcomes’. They use three criteria to assess this: spending less, spending well and spending wisely. While this is challenging to assess in relation to the Quality and Outcomes Framework, there is some evidence from a health economics perspective that, for those indicators amenable to evaluation, the Quality and Outcomes Framework incentive payments are likely to be a cost-effective use of resources for most primary care practices. However, there is variation between indicators in the extent of improvement (in terms of additional numbers of patients treated) required for the Quality and Outcomes Framework incentive payments to be cost-effective.