The idea that positive consequences induce behaviours is based in theory, and supports the use of pay-for-performance for healthcare quality improvement. In practice, however, there is insufficient evidence to sustain financial incentives to improve the quality of primary health care. The use of pay-for-performance has increased but it has been controversial and has achieved little, raising questions about which mechanisms are involved, how they work and whether their design is at fault.

This issue presents the findings of Norman et al. (p. 55-67) about perceptions of general practitioners (GP) in relation to the British pay-for-performance based on the Quality and Outcomes Framework (QOF). Being one of the largest pay-for-performance schemes in the world, it raises passionate debates about its features and effects, and expectations about the exemplary role it can play for the design and implementation of pay-for-performance interventions in primary health care.

The QOF was introduced in the UK in 2004, and findings regarding its effects are still limited, with few studies providing pre-post comparisons with sufficient follow-up period. On the one hand, the existing evidence indicates that QOF has reduced impact on improving health outcomes due to its focus on process-based indicators and the indicators’ ceiling thresholds being inconsistent and lower than those stipulated in clinical guidelines. On the other hand, the QOF has been evolving, searching for more transparency and independency, and moving from looking at process to looking at outcome. Within this spirit, the National Institute for Health and Care Excellence (NICE) has assumed the leadership in the development of the clinical and public health indicators since 2009.

In contrast to Norman et al. position on the role of variability and standardisation in health care, I am of the view that some standardisation based on good evidence is necessary and inherent to healthcare quality improvement. Part of the variability in healthcare practices may be legitimately attributable to pertinent choices, but part is surely attributable to inadequate care. Another study (Lester H et al. Br J Gen Pract 2013; 63:e408-15) contradicts Norman et al., pointing to the acceptance of pay for performance as a routine part of primary care in England, the evidence-based nature of the indicators as a key factor to its success, and the view of previous more individualistic and less structured ways of working as poor practice. With regard to undesirable unintended consequences of the British pay-for-performance experience, it seems more consistent the perception of a negative impact on medical professionalism, with induction of consultations focused on “ticking the boxes”, some loss of autonomy, and the trend of a minority of GPs to prioritise their own pay rather than patients’ best interests. Claims for more indicators valuing preventive care, care continuity, and other aspects of a holistic view of health care are recurrent.

The British pay-per-performance scheme has been subject to much change and is still a “work in progress”, and certainly, in being followed up, will provide important lessons about pay-for-performance in primary health care. At the same time, it is fundamental that other interventions are designed more carefully, accounting for in-depth appreciation of theories able to support their components, and involving more tests in different contexts before scaling-up.

Publication Dates