Performance indicators for general practiceBMJ 1995; 311 doi: https://doi.org/10.1136/bmj.311.6999.209 (Published 22 July 1995) Cite this as: BMJ 1995;311:209
- F Azeem Majeed,
- Simon Voss
- Lecturer in public health medicine Department of Public Health Sciences, St George's Hospital Medical School, London SW17 0RE
- Consultant in public health medicine Southampton and South-West Hampshire Health Commission, Southampton SO16 4GX
Will lead to league tables of performance
Some family health services authorities are now producing performance indicators for the general practices they administer.1 2 3 With the move towards a primary care led NHS,4 5 these indicators will become an important management tool. League tables of practice performance are a possibility: for example, practices could be ranked by rates of uptake of cervical smear tests and the proportion of drugs prescribed generically. Many general practitioners, particularly those who work in deprived communities, will find this development threatening and may think that league tables will unfairly label their practice as performing poorly. Family health services authorities must therefore ensure that performance indicators are interpreted appropriately.6
Performance indicators may be used to identify and reward high performing practices with increased allocations for staff and premises. Conversely, if resources are allocated according to health need rather than performance, then less well developed practices (which are often located in areas with high need) may receive more resources. Because such practices may not have the capacity to use additional resources effectively this may lower the morale of the more innovative practices. Performance indicators should not therefore be used uncritically when resources are allocated to practices.7
General practitioners can benefit from performance indicators. They can use them to identify how their practice deviates from the norm and where scope for further investigation and audit may exist. For example, a practice with a high proportion of technically unsuitable smears may want to investigate this further. Performance indicators can also help practices to identify priorities for improvement and to monitor how well they address them over time. Finally, performance indicators can be used to carry out descriptive research into variations in medical practice in primary care.8
The most important limitation of performance indicators is that they measure only certain aspects of performance. For example, they can tell us what a practice's referral rate is but tell us nothing about the appropriateness of these referrals. Performance indicators also tell us nothing about what most general practitioners would consider to be their most important role: the clinical care of individual patients. Secondly, performance indicators could create perverse incentives, with general practitioners concentrating on improving the indicators rather than improving the quality of their care. Thirdly, performance indicators are constructed from routine data, and there are errors in these data, especially in age-sex registers (inflation of lists), census data (under enumeration), and data on referrals (inaccurate coding). Finally, indicators of prescribing are derived from prescribing analysis and cost (PACT) data; with the steadily increasing cost of NHS prescriptions, more drugs will either be prescribed on private prescriptions or be bought over the counter, making prescribing indicators derived from PACT data less useful.9 Family health services authorities need to be aware of these limitations when they use performance indicators to assess practices' performance.
To improve the limited information available to patients (for example, in practice leaflets) when they choose a practice, some family health services authorities may wish to make performance indicators available to the public. Although many general practitioners will oppose the publication of performance indicators, we already have league tables for schools and hospitals, and the publication of league tables for general practices may be inevitable. But, because of the controversy raised by league tables elsewhere and the lack of consensus between general practitioners and managers over what constitutes “good” performance, this is a development that family health services authorities should handle sensitively. General practitioners should therefore be involved at all stages in the development and implementation of performance indicators.1 Even if the indicators are not released to the public, the new health commissions will make much greater use of performance indicators in monitoring general practices.3 4 General practitioners should therefore collaborate with family health services authorities to improve the quality and usefulness of performance indicators, and they should start to discuss with their local health commissions how they intend to use performance indicators in the management of primary care services.