General practice and the rest of the primary care team, rather than specialist or hospital care, deliver the lion’s share (90%) of healthcare.1 They also provide the anticipatory care necessary for early and better management of the chronic diseases that characterise modern industrial societies. A strong, self-reliant primary care workforce increases quality as well as cost-effectiveness.2 Thus, the way a country remunerates its primary care workforce is vital. This is a good time to debate the options in Australia — a federal election year in which healthcare is likely to be a central issue. Changes to the administrative system can have enormous implications for primary care. So far, planned changes in Australia have been tentative, consisting of “add-on” improvements such as payments to general practitioners (GPs) in addition to the traditional fee-for-service arrangements. These include Practice Incentive Payments (PIPs), which pay GPs who can demonstrate using set protocols for managing some chronic diseases (eg, asthma, diabetes), and Service Incentive Payments (SIPs), which are specific payments for certain services such as mental health care and vaccination. The complexity of administering these programs has prompted complaints from GPs, and in response a Red Tape Task Force has been convened.3

Healthcare systems differ hugely from country to country, as we see from the articles that follow. Reimbursement is perhaps its most emotionally highly charged aspect and, however contentious an issue, some system has to be chosen. Strong primary-healthcare-led systems like those in the United Kingdom4 (page 109) and The Netherlands5 (page 110) use capitation systems as the basis of paying GPs: they contract to assume the obligation to provide care for a group of patients, and their financial rewards are independent of the actual service and care delivered. In the former Yugoslavia, with its socialist origins, remuneration took the form of a salary, accompanied by planning (and restriction in the number) of GPs. A more market-driven way of paying GPs is for GPs to “deliver” before payment (fee-for-service), and in competition with each other, as in Australia and the United States (page 113).6 Canada (page 111) has opted for a combination of methods,7 and New Zealand (page 106) is experimenting with a variety of interesting models in quick succession.8

Financial incentives have a direct influence on GPs’ behaviour. For example, in Belgium, 46% of GP–patient encounters are home visits,9 whereas in The Netherlands this is only a small proportion of GPs’ clinical activities. The population health status and infrastructures of the two countries hardly differ, so the difference can only be explained by incentives: under Belgian fee-for-service, a home visit is chargeable. It also strengthens patient satisfaction with the GP in a competitive environment. On the other hand, the Dutch GP receives a capitation fee irrespective of whether the patient is seen at home or at the practice (or not at all).

One problem with fee-for-service payment is that the way GPs are funded is confounded by other innovations that Australia should be considering, such as patient registration. Although theoretically this could be separated from how doctors are paid (capitation, for example), nowhere does this occur. A second problem is that fee-for-service can be inflexible about who is remunerated. This has held back the proper utilisation of nursing in general practice in Australia simply because nearly all services in general practice are ineligible for a Medicare rebate if provided by nurses, even though for many services (eg, preventive10 or protocol-driven chronic care11) nurses may be better suited.

A third problem is the need for a business mind with fee-for-service general practice. Many doctors want to practise unencumbered by a “small shopkeeper” role. One consequence was the evolution in the 1980s of “entrepreneurial practices” (those whose owners were more interested in making a profit than serving their communities). These offered greater flexibility for the increasing numbers of doctors who, wanting “just to practise medicine” and happy to abdicate their commercial role, flocked there.

“Perverse incentives” reward some clinical activities better than more valuable ones.12 For example, a GP who delivers many short consultations will earn more than one who has fewer and longer consultations — even though longer consultations are associated with better attention to preventive healthcare and psychosocial problems.13 Attempts to address this by providing less reliance on fee-for-service (to the fury of the Australian Medical Association)14 with additional alternative payment systems (so-called blended payments) such as PIPs and SIPs — funded by what might have otherwise gone to increased fees — are probably only partly successful. It is too early to tell if the complexity of administering them is any better than the fee restrictions that arose to discourage entrepreneurial practices. Do we need more fundamental reforms of the GP system? What are the alternatives?

Capitation payment and its associated patient registration (the “list”) feels like clinical freedom for many GPs in the UK and The Netherlands: the GP accepts an obligation to provide care for the patients on the list and do what is in their best interests. It has offered GPs a level platform to counter medicalisation, overdiagnosis, over-referral and spurious prescribing, without the tilt of having to please the patient (something usually miscalculated in any case15). But there are disadvantages. A capitation system can be a haven for laziness, because payment comes irrespective of the quantity (let alone the quality) of care. GP-initiated activities — like anticipatory (chronic disease management) and preventive care — are particularly sensitive to this. Perhaps this is the basis for the reforms currently under way in the UK NHS.16 Disciplinary hearings against GPs in the UK and The Netherlands are, to a large degree, occupied with GP failure to provide enough care — in particular, failure to visit patients in a (perceived) emergency at home. In Australia, the focus of disciplinary hearings is on overservicing. The issue for capitation systems is deciding what is enough care; for fee-for-service, deciding what is too much.

Blended payments (a mixture of fee-for-service and payments for good practice) sound sensible (the best of both worlds), but there is little evidence17 to reassure us we might not get the worst of both: entrepreneurial GPs learning which mix of activities yields the highest earnings, and government reacting by over-regulating the system to avoid this.

GPs are at the forefront of evidence-based patient care. It would be good if GPs’ financing systems were established by good evidence, but little exists regarding the effects on service of different payment systems (Box). Thus, we need to trial different systems, not simply enact the latest political ideology. The current flux in the Australian healthcare system is surely an ideal environment for such experiments. Possible alternatives would be payment systems that allow for patient registration, that include the option of salaries for doctors uninterested in running a business, and that encourage doctors to collect and analyse clinical data about the services they provide. There is no doubt that such trials would be hard to conduct politically,17 and perhaps randomised controlled trials would have to give way to the pragmatics of quasi-experiments. But we need changes to the system that are fundamental, rather than the lean-to sheds propping Australia’s current archaic system.