Singing from the QOF hymn sheet: Stairway to Heaven or Mephisto Waltz?

The general practice research community has lost no time in exploring the impact of the new general practice contract that was implemented in 2004, and this issue of the BJGP publishes a number of articles that offer food for thought. We find that QOF scores were related to practice size, but that smaller practices lost out through lower scores in the organisational rather than the clinical domain.¹ Wright et al,² in a study covering 8 569 practices in England showed that deprivation was inversely related to QOF score contradicting the findings of an earlier but much smaller study which showed that higher QOF scores were positively associated with deprivation.³ However their findings agree with Wang et al¹ that larger practices achieved higher QOF scores. On the other hand Wang et al¹ found no effect of deprivation on quality scores and also acknowledge that ‘quality, as measured by the QOF, may … reflect quality in data recording [as much as] quality in delivered care’. Guthrie et al⁴ demonstrate wide variation in financial reward per quality point, with the most affluent practices gaining substantially, and the most deprived losing out, thus institutionalising the inverse care law.

It is no surprise that larger, richer practices seem to score better in the QOF — they have more resources to record data systematically and perhaps they are historically more strongly motivated to maximise income through whatever means the contract allows. Many will have been fundholders in the past. We should perhaps not be overly concerned at income inequality between practices, since there will always be variation in the balance of income maximising compared to vocational motivation in the cultures of different practices. However Guthrie et al's⁴ findings raise the more worrying issue that the new contract actively reduces the resources available to practices servicing deprived populations. Morgan and Beerstecher's study takes a slightly different slant, relating QOF score to contract status, and concludes that GMS was the ‘most efficient contract status’, giving almost the same QOF score as Personal Medical Service (PMS) practices for less funding per patient per year (£62.51 compared with £87.38), and finding that Employed Medical Service (EMS) practices had much greater funding (£105.37 per patient) for much lower QOF scores (∼750 compared to <900 for PMS and GMS).⁵

The QOF reduces a variety of quantitative outcomes and processes in practice populations to a single number, but is not designed to measure or assess such important features of general practice as the quality of consultations, personal continuity of care or GPs' skills in managing uncertainty or coping with multiple problems often presented by patients. It would be unwise to interpret, for example, Beerstecher and Morgan's conclusion⁵ about the most ‘efficient’ practice contract relative to the QOF as a generalisable ‘fact’. Policy makers and primary care trust managers may be tempted to respond to it by attempting to change the contracts of EMS and PMS practices to GMS in order to achieve improved cost effectiveness. They should remember that the QOF and quantitative studies of it give only a partial picture of what practices do and what patients need. An EMS practice set up to care for an influx of asylum seekers is likely to underscore on the QOF but may nevertheless be an essential lifeline for its patients (and by taking on these often needy people with multiple problems, enable neighbouring GMS or PMS practices to achieve higher QOF scores). A PMS practice's high funding may merely indicate its historical success in playing successive governments at their perennial game of restructuring general practice.

There is evidence that standards of clinical care as measured by the QOF had already improved during the period 1998–2003, before the introduction of the new contract, and that predictors of these standards included practice size and low deprivation.⁶ An earlier study showed also that standards were higher where there was good teamwork.⁷ The importance of teamwork is reinforced by a qualitative study of the RCGP's Quality Team Development (QTD) programme.⁸ Teamwork is a vital component of quality that cannot easily be measured by the reductive instrument of the QOF. Moreover, teams are complex adaptive systems that adjust in unpredictable ways to the environment in which they operate.⁹

The big question is how these complex adaptive systems will respond to what is now becoming a protocol driven (rather than guideline informed) system of monitoring and rewarding clinical actions and outcomes, largely managed through computerised checklists in the consultation. Now that QOF is established, new protocols are being imposed by administrative diktat, without having gone through a process of educational dissemination and cultural acceptance. As Greenhalgh observes, ‘Traditional management theory often uses a mechanistic, cause and effect model of how organisations work’,⁹ and the QOF is a perfect example of this approach. This is a potential threat to intelligent and humane practice. In order to achieve maximum QOF points, GPs may adopt a more bureaucratic, less socially and psychologically responsive, style of consulting. Computerised checklists and reminders are often seen as an unwelcome intrusion into the consultation,¹⁰ but there is now a strong financial incentive to use them, so their use is likely to increase whatever reservations GPs or patients may have. There is therefore an obvious conflict between the core need to address patients' agendas and the need to satisfy ever increasing information and management demands to service the QOF.

Elwyn's¹¹ ‘reflective narrative’ of the patients he saw in one surgery just before Christmas 1996, with all their puzzling complexity, humanity and unpredictability, gives a vivid picture of traditional (and I think fundamental) aspects of general practice. These may not easily survive the unthinking implementation of expert guidance made concrete in QOF targets and lists of administrative tasks. By imposing a centrally driven, quantifiable disease management model of quality on primary care medicine, the softer, unquantifiable, but necessary human quality of caring for patients as individual people is inevitably downgraded. It may be argued that this is a price worth paying for the technical improvements in disease management that are claimed to be benefits of the QOF. However it is far from clear whether the QOF promotes ‘quality in practice’ or (at least thus far) merely reflects pre-existing secular trends in disease and risk management.

This first wave of research on the response to the QOF gives us some useful information by taking data that is in the public domain and processing it with quantitative methodology. There is nothing wrong with that, but it gives a necessarily incomplete picture, just as a street map gives an incomplete description of a city. It is worthy of note that patients' assessments of quality of care as measured by the general practice assessment survey, and technical measures of quality such as hypertension monitoring were poorly correlated.¹² Qualitative and social science researchers should investigate the effects of these changes in the contract on the social and psychological domains of general practice, both as regards the experience of patients and the behaviour and attitudes of clinicians and their teams. The Department of Health should reflect on the law of unintended consequences before pushing through further change. GPs should be careful not to neglect holistic personal and family medicine, which is their traditional strength, in pursuit of high QOF scores. And, before becoming too complacent about their increased financial rewards, GPs should perhaps reflect on the sad fate of Dr Faustus, who also believed at first that he had been offered a good bargain.