Intended for healthcare professionals


After 12 years, where next for QOF?

BMJ 2016; 354 doi: (Published 04 August 2016) Cite this as: BMJ 2016;354:i4103
  1. Nicholas Steel, clinical professor in public health1,
  2. Paul Shekelle, director2
  1. 1Norwich Medical School, Norwich NR2 3DP, UK
  2. 2West Los Angeles VA Medical Center, CA, USA
  1. Correspondence to: N Steel n.steel{at}

With bold innovation, QOF has the potential to lead the charge on patient centred care

The recent EU referendum result reminds us that behaviour is not simply determined by economic self interest, yet the notion of paying physicians to perform remains remarkably popular. Medicare payments in the US are increasingly linked to the quality of services, and pay for the 32 000 UK general medical practitioners working for the National Health Service has been linked to quality of care through the Quality and Outcomes Framework (QOF) since 2004.1 2

Back then, quality indicators for general practice were new, and electronic medical records had become widespread in British general practice.3 The UK Department of Health wanted to improve population coverage of healthcare processes that had evidence of improved health outcomes and invested £8bn (€9.5bn; $10.5bn) over the first three years of QOF to reward performance in four areas: clinical, organisational, patient experience, and additional services. There were 80 clinical indicators in 2004 and 77 in 2016, covering cardiovascular disease, diabetes, respiratory disease, dementia, mental health, cancer, chronic kidney disease, epilepsy, learning disability, osteoporosis, rheumatoid arthritis, palliative care, obesity, smoking, cervical screening, and contraception.4

Evidence of reduced mortality existed for 25 of the original 80 indicators, and applying that evidence to the UK population showed that QOF had the potential to reduce mortality by 11 lives per 100 000 people if performance improved from baseline to the target for full payment.5 Twelve years later, Ryan et al measured the actual reduction in mortality, which turns out to be 12 per 100 000 before correction for reduced mortality in similar countries without QOF and just four after correction (not statistically significant).6 The authors point out that their analysis might be underpowered, but it is as good as we are likely to get and consistent with other findings.6 7 QOF was not primarily designed to reduce population mortality and seems to have achieved pretty much exactly what was expected.

What else have we learnt from the first 12 years of QOF? No trials were conducted and the evidence is generally weak, but clear benefits are the new source of publicly available population data on major conditions, the modest improvements in quality of care for chronic diseases in the framework, and reduced inequalities between deprived and less deprived areas. However, improvement for conditions not in QOF (such as osteoarthritis) may have stagnated, continuity of care may have declined, and quality improvement may have become too narrowly focused on QOF to the detriment of other initiatives.6 8 There are concerns that guidelines in general, and QOF in particular, risk subverting patient centred care, may inadvertently encourage unwanted polypharmacy and drug interactions in older patients with multimorbidity, and may have uncertain relevance to primary care patients.9 10

QOF will become a teenager in 2017 and changes are needed. The focus should move from manipulating risk factors such as blood glucose and lipid levels to improving health in a way that has meaning for individual patients, while acknowledging that healthcare is only one factor that contributes to improved health. Examples of patient centred quality measures include medication reconciliation after discharge from hospital and patient reported change in health status.1 Most patient centred quality measures will be less directly under the physician’s control than current QOF indicators, and caution will be needed if payments are to be attached.

Financial incentives in QOF are far larger than they need to be (one QOF point for 2016-17 is worth £165 (€197; $218)). Indeed, economists tell us that, although financial incentives work for simple repetitive tasks, they are ineffective for more complex tasks and may even be counterproductive.11 There is a case for abandoning QOF payments altogether as some areas already have, but payment for performance is likely to remain in mixed payment systems when there are clear gaps in quality.12 Transparent data reporting should also be maintained.13

Applying population evidence to diverse individuals in primary care is difficult and uncertain; it needs to be guided by patient preferences as much as evidence. The National Institute for Health and Care Excellence (NICE) is responsible for managing QOF and is developing guidance on multimorbidity and patient preferences that will be highly relevant. Good resources for sharing decisions with patients exist (eg,, but more are needed, together with adequate consultation times, before they become routinely used in practice. QOF indicators should be accompanied by brief and clear decision aids about the benefits and risks of treatment options and the time to net benefit in cases where (small) harm is immediate and (greater) benefit delayed, to help combine evidence with patient preferences. QOF has the potential to lead the charge in the widespread implementation of patient centred care to improve quality, if NICE has the will and courage to innovate and evaluate.


  • Analysis, doi: 10.1136/bmj.i4060
  • Competing interests: We have read and understood BMJ policy on declaration of interests and declare that NS was a member of the NICE primary care Quality and Outcomes Framework indicator advisory committee from 2009 to 2013.

  • Provenance and peer review: Commissioned; not externally peer reviewed.


View Abstract