THE increasing complexity of the challenges of medical practice demands new modes of organising practice. Both the processes of diagnosis and of clinical management are becoming more challenging as a result of ageing (and hence more prevalent biological deterioration) of populations, increasing co-morbidity (as a result, in part, of more effective treatment and, hence, longer survival of people with more problems), increasing iatrogenesis (as a result of more invasive treatment modalities), and increasing recognition of the myriad influences on health (and, hence, of myriad alternatives for prevention and management of ill health).1 Many countries, as well as the World Health Organisation, have recently adopted the strategy of improving primary care as a response to the challenges of improving the quality of care while minimising rapidly increasing costs. However, the burden of meeting these mounting challenges cannot be shouldered by primary care alone. Whereas health services in general have been responsible for a substantial proportion of the increase in life expectancy in the last half century,2 at least one-third of this is solely attributable to primary care (Bunker et al, unpublished data, 2003). The additional degree to which primary care (particularly in an increasingly specialised medical environment) has contributed, through its referral patterns, to the overall positive effect of health services is unclear. How much are the benefits of health services attributable to the relationship between primary care and specialty care? There is little information that addresses this question or, even, that considers the way in which the relationship should be envisaged. Little attention (at least in Anglophone countries) has been devoted to developing either consensus-based or evidence-based guidelines for referral from primary care to specialty care and, in fact, the need for evidence on relative degrees of benefit does not appear to have been recognised. This paper provides a general framework for starting a process of rationalisation, and some preliminary evidence that could inform strategies to proceeding further. The data come primarily from national samples in the United States (US) — the better organisation of general practice in the United Kingdom (UK) may be associated with far better communication between primary care physicians and other specialists in the UK. Nevertheless, the data presented herein raise questions that are more generally applicable in all countries, and the US data provide a basis for crosscountry comparisons.
Stored only when both papers are in the corpus — 49 citations exist upstream.