According to the Royal Australian College of General Practitioners (RACGP):

In this paradigm, a condition such as chest pain may have its roots in the physical, social and/or psychological domains, and a robust general practice system is crucial to managing these complex clinical interactions. A GP is well placed to determine the relative contributions of these causes and to effectively manage the interface between primary care and the hospital. Starfield and colleagues have long argued this case, and few have offered a cogent counterargument.2 However, in the United Kingdom, and now in Australia, a formal system of GPs with special interests (GPwSIs) — that is, GPs having a subspecialisation within general practice — is touted as the answer to the growing demand for specialist services in this issue of the Journal (page 111)3 and elsewhere.4

There is clear evidence of the value of subspecialisation in some areas of general practice — for instance, in Indigenous health, palliative care, drug and alcohol services, and HIV management.5 The case for expanding these roles to include yet more “special interests” is that the policy will “develop careers through offering additional interest, personal development and heightened self-esteem”.6 This may be a persuasive argument for increasing the attractiveness of general practice, where there is a growing manpower crisis and an urgent need for enthusiastic new recruits. However, this approach is also associated with a call for more “locums” to make up for manpower shortages in general practice and the deskilling of the GP pool in some clinical areas. Thus, diverting the GP workforce will compound shortages in “core general practice”.

A key driving force behind the promotion of GPwSIs in the UK was the perceived need to reduce “inappropriate” referrals and hospital admissions, thereby improving the efficiency of health resource utilisation. However, referral processes in general practice are frequently complex and multifactorial, and are unlikely to respond to this approach.7,8 GPs may also be reluctant to refer to colleagues whom they perceive as generalists with much the same level of skill as their own.9 In other words, GPwSIs may offer an additional service rather than an alternative to specialist services.

If GPwSIs were to reduce the workload for specialists, their impact on primary care could be unhelpful or even detrimental. In a recent UK patient survey, it emerged that patients foresaw difficulty making appointments with their chosen GP in circumstances where a special-interest GP was working at the practice.10 In the Australian context, it may be difficult to predict the likely impact, given that general practice is a privatised business. However, if such difficulties with access were mirrored here, the resulting loss of continuity of care could diminish the key value of the GP as defined by the RACGP. This is of special concern given the growing burden of chronic disease in an ageing population. Indeed, many patients are still receiving suboptimal care for chronic disease.11 Furthermore, with the increasing policy emphasis on prevention, screening and surveillance, many patients will need more access to GPs. To achieve the long-term benefits of these policy directions, we need to protect the principle of equal access for all.

Patients in the UK, where the GPwSIs concept has been widely embraced, are impressed by the speed of access and personal aspects of the service, but not necessarily by its quality. In the UK, GPs-in-training have been reported to be naive about the potential complexities of accreditation and governance required for the roles of GPwSIs. If GPwSIs are to become a feature of the primary care landscape in Australia, appropriate training and accreditation will be critical to prevent them being regarded as a cheaper, second-class service.

Another danger with facilitating growth in the number of GPwSIs is the possibility that it will interfere with the fabric of primary care — namely, an accessible, generalised, integrated and coordinated approach to patient management. Furthermore, we should not underestimate the financial incentive to dabble in lucrative specialist procedures in a largely unregulated health care system in which many GPs feel undervalued. Indeed, a more appropriate remunerative structure, especially for cognitive (as opposed to procedural) practice, may well serve to change perceptions of what is attractive and rewarding. Thus, promoting GPwSIs in Australia may serve only to redistribute scarce GP resources while running the risk of exacerbating GP workforce shortages. This, in turn, will require more GP training positions to be established, a decision that is not under the control of the profession. Balancing the increasing role of GPwSIs in Australian general practice will require careful workforce modelling and planning.