Diagnostic delays and referral management schemes: how ‘integrated’ primary care might damage your health
International Journal of Integrated Care, 17 December 2008 - ISSN 1568-4156
Diagnostic delays and referral management schemes: how ‘integrated’ primary care might damage your health

For many years I was a disciple of the Starfield hypothesis—that health care systems with a stronger orientation to health promotion, disease prevention, and the provision of accessible, universal and integrated primary care-based services led by generalist physicians will achieve better health outcomes and at lower cost [1]. Moreover, I have been persuaded by the chronic-care advocates that a seismic shift is occurring in the burden of care that now requires a systemic move away from episodic care undertaken in specialist hospital institutions to long-term and integrated care management and coordination undertaken in the community [2].

The two agendas imply the need for a reinvigoration in primary care-led services and they sound complimentary strategies, but on closer inspection the two models differ greatly. For example, countries such as England and Denmark that are ‘blessed’ with a network of GP-led surgeries and health care centres have systems of primary care that are becoming less appropriate in meeting today's chronic care agenda.

There are a number of reasons for this. Firstly, there is growing evidence that a generalist physician's ability at the ‘problem recognition’ stage is limited when it comes to making an effective diagnosis of a chronic condition. Part of the problem for this is that generalists cannot keep pace with the assimilation of knowledge and diagnostic skill required to recognise all of the ever increasing and nuanced categories of chronic illness being presented by today's patients. In the time-frame of a normal consultation (<10 minutes in England) the ability to differentiate, say, an early-stage neurological problem or suspected cancer from the every day aches and pains of an older person is difficult. A forthcoming report on Rheumatoid Arthritis (RA) care in England, for example, concluded that GPs find it difficult to identify a person with RA given the generic nature of the symptoms; the possible range of associated inflammatory conditions; and a lack of recognition of those key presenting indicators that may suggest prevalence of RA [3].

Secondly, in countries where generalists act as gatekeepers to specialist care, this potential lack of diagnostic capability potentially has a knock-on effect in the form of delays in access to appropriate specialist consultations and treatment. For example, RA sufferers require rapid referral to specialist diagnosis and treatment to prevent irreparable damage to joints, but the RA study found unacceptable diagnostic delays in many parts of the country. Indeed, diagnostic delays to specialist care is one of the reasons put forward for the poor cancer survival rates in England and Denmark compared to countries like France.

To overcome the issues of problem-recognition and diagnostic delays an ‘integrated’ primary care system almost certainly needs to develop an appropriate (probably lower) threshold of referral for specialist diagnosis for those presenting with a suspected urgent or complex care need. However, this tends to go against the current zeitgeist of reforms in countries such as England where primary care organisations are establishing stronger referral management processes in order to reduce ‘unnecessary’ hospital-based activity and use the savings that accrue to enable a shift of care into primary and community care-based alternatives. This is where ‘integrated’ primary care potentially becomes damaging as systemic incentives create a disconnect, even competition, between providers of primary and specialist care.

To provide one example, a recent scheme put forward for implementation in one English Primary Care Trust—an ‘interface dermatology and skin surgery service'—aims to triage all dermatology referrals from GPs in an attempt to divert the majority away from the hospital to a new community-based service alternative. The service seeks to stimulate primary care-based provision despite the knowledge that it has less specialist capacity and will cost approximately 75% more on a cost-per-case basis. Nonetheless, the PCT expects that post-implementation no more than 12 per cent of referrals (an arbitrary figure for the most complex cases) should now reach the hospital sector—a decision that almost certainly puts some patients at risk by being ‘inappropriately’ referred to the community service whilst not providing value for money. In another part of England, Torbay Care Trust has created a points-based system of incentives for its GPs that would see them paid approximately £1200 per practice if they hit an 8% in-year reduction in the numbers of referrals to hospitals—justified on the basis that recent increases in GP referrals had ‘put pressure’ on the financial health of the local system [4].

Professor Frede Olesen of the Research Unit for General Practice at Aarhus University in Denmark calls this the problem of the ‘inverted T'—the situation in which a disconnect has occurred between the horizontally ‘integrated’ primary and community care services (the horizontal line) with the vertically integrated care pathway that takes a patient from first-contact to specialist to ongoing care (the vertical line). The English system seems a classic case, since much emphasis has been placed on streamlining and integrating services to both elements but at the expense of facilitating an effective integration of the two.

The conclusion to be reached is that, to avoid adverse outcomes, integrated chronic care needs to be developed from a systemic perspective rather than for isolated horizontal and vertical components. In this respect, there may be benefits to integrating generalists and specialists into care networks, perhaps even to co-locate them in primary care teams or polyclinics. Additionally, much needs to be done to assess the quality of diagnosis and referral practices since no formal quality assessment approach exists that includes the critical feature of problem-recognition despite the evidence that patients are often more likely to improve when they and their practitioner agree on what the problem is.

Whilst the 2008 World Health Report is undoubtedly right to suggest that the better integration of primary care services is a goal to be welcomed—‘now more than ever’ [5]—the pursuit of such goals needs to avoid perverse systemic incentives that may lead to worse rather than better health care outcomes.

Nick Goodwin, PhD, Chair International Network of Integrated Care and Senior Fellow, Health Policy, King's Fund, London, UK. E-mail: ngoodwin@kingsfund.org.uk

Starfield B. Primary care: balancing health needs, services and technology. Oxford, UK: Oxford University Press; 1998.
Goodwin N. National Health Systems: overview. In: Heggenhougen K, Quah S, editors. International Encyclopedia of Public Health. Vol.4 San Diego: Academic Press; 2008. p. 497-512.
Steward K, Land M. A postcode lottery of care? Report on the perceptions of patients and professionals on rheumatoid arthritis care. London: King's Fund; forthcoming;.
Torbay Care Trust. Practice-based commissioning—incentive scheme proposals for 2008–9. Torbay Care Trust, NHS; 2008. Available from: http://www.torbaycaretrust.nhs.uk/about-us/our-plans/practice-based-commissioning-1/2008-09-commissioning-plans/2%20-%202008-9%20incentives.doc.
World Health Organization. Primary health-care—now more than ever. Geneva: World Health Organization; 2008. (The World Health Report 2008). Available from: http://www.who.int/whr/2008/whr08_en.pdf.