Education And Debate Managing demand

Managing demand in general practice

BMJ 1998; 316 doi: (Published 20 June 1998) Cite this as: BMJ 1998;316:1895
  1. Stephen Gillam, director, primary care programme (cnsgil{at},
  2. David Pencheon, consultant in public health medicineb
  1. aKing's Fund, London W1M 0AN,
  2. bInstitute of Public Health, University Forvie Site, Cambridge CB2 2SR
  1. Correspondence to: Mr Gillam

    This is the third of five articles on ways of managing demand for health care.

    Increasing patient expectations are placing strains on general medical services. While empirical evidence for increasing demand is difficult to establish, the population's use of primary care services has changed and will continue to change in response to demographic and technological pressures. Consultation rates have increased slightly over the past 10 years but out of hours calls have escalated more dramatically.

    Most chronic diseases can be managed with little recourse to specialist support, but patients are increasingly aware of diagnostic and therapeutic advances. The information revolution has fuelled demands for user friendly services of constantly improving quality across the public sector. Users seek appropriate information skilfully communicated by the health professional of their choice. Increasingly, people want to be involved in decisions about their management. The sense of a growing gap between what their patients want and what the service can afford provides a stressful undercurrent in the working lives of general practitioners.

    The advent of fundholding heralded a major shift of control over the allocation of health service resources to general practitioners. These powers are set to increase and with them new budgetary responsibilities. This paper analyses approaches to demand management which reflect both the changing relationship between general practitioners and their patients and the changing interface between primary and secondary care. With the development of more graduated pathways of access through the health system, these interfaces are blurring. How general practitioners manage needs expressed at first contact can influence patients' future self care; how they manage the interface with secondary care adds to pressure in this sector.

    Summary points

    The advent of primary care groups will involve all general practitioners in resource management

    Multifaceted approaches are required that affect demand at all points along the path from first contact to possible referral

    Informed involvement of patients in decision making may reduce their subsequent use of health services

    Telephone triage systems (both in and out of hours) offer enormous opportunities to change existing workloads

    A more interactive dialogue between general practitioners and hospital specialists using new technologies should allow more patients to be managed in the community without direct access to specialists

    The scope

    The objectives of demand management are essentially utilitarian: the maximisation of total met need for the greatest number within available resources. The process is not simply about curtailing demand for ineffective services but may involve creating demand for underused services known to be cost effective. Attempts to manage demand can be directed in three ways: at patients, at health professionals, or at the health system as a whole. Both patients and doctors are influenced by their perceived need for care, by preferences for treatment, and by motives unrelated to health. Both patients and doctors respond to wider environmental influences such as economic incentives.

    In one sense the business of primary medical care is about nothing else but demand management at all points in the patient's passage through the system. The figure shows some of the strategies that can be more systematically applied in general practice. Underpinning all these activities is the desire to share information and make decision making more explicit. Increasingly, general practitioners and other members of their teams can influence demand beyond the surgery door through educational work (for example, in schools or work places), in new clinical roles (for example, in casualty departments), or through commissioning activities. However, this article will concentrate on decision making with patients from the point of first contact within a single treatment episode to the point of possible referral.


    Approaches to demand management in primary care

    The patient-practice interface

    Health education

    The previous article in this series described how provision of appropriate patient information can help patients manage minor conditions without recourse to professional help. Do healthier people who feel they have more control over their own health make less (expensive) use of health services1? The impact of health promotion programmes that involve people in decision making about their own care is unclear.2 It is possible that they generate a shortlived increase in demand before the benefits of longer term reduction in consultation rates are realised. However, self care handbooks have been shown to save two to three times their cost and decrease medical visits.2 Some health maintenance organisations in the United States are subsidising the cost of self help videos for their members. The goals of health promotion extend beyond managing demand. Likewise, traditional health promotion in the form of risk reduction provides only one component of comprehensive demand management.

    Greater understanding of how people react to their own ill health is a prerequisite to better management of the transition from self care to professional care. The likelihood of seeking care is related to belief in its effectiveness, how serious the symptoms are perceived to be, and previous experience of making contact in the same situation.4 The investment of extra advice during consultation for minor illness can encourage subsequent self care. For example, the appropriate deferral of antibiotic prescription reduces the likelihood of subsequent reattendance with upper respiratory tract infection.5

    “Contracts” with patients

    Many health professionals regard the Patient's Charter as overemphasising entitlements and thereby generating inappropriate demand. One way of better reconciling rights and responsibilities may be to develop more explicit “contracts” between users and their providers. Pietroni has described such an attempt to empower patients to make contact when it is important to do so while dissuading them in other circumstances. Further research is required to establish whether this limits superfluous demand and encourages earlier intervention where appropriate.6

    Telephone advice and triage

    In the US as much as a quarter of all primary medical care is conducted over the telephone. “Patient risk assessment” using iterative computer generated algorithms and protocols may (more swiftly) ensure that the right patient receives the right care from the right health professional than the traditional chain of referral.7 Evidence is increasing on the benefits of telephone triage. Evaluation of telephone advice software in south London suggested that many people requesting urgent appointments were able to look after themselves after receiving advice from specially trained nurses. The system is now being used by many general practitioner cooperatives8 and has been successful at a national level in other countries.9

    The importance of telephone helplines was affirmed in the white paper on the NHS in England, with the announcement of a nurse led help line (NHS Direct); pilots have already started. Further research is required to establish the diagnostic sensitivity of telephone advice software at a distance and whether or not some categories of patients are placed at increased risk. Compared with commercial deputising services, cooperatives appear to “convert” more home visits into telephone advice or consultations at an out of hours centre. Here again more evidence is required to judge the quality of the service they offer.10

    Other means

    General practices are beginning to make self help information available via the internet. The UK has one of the highest levels of home ownership of personal computers in the world. In future will every practice use its own home page to manage demand for its services? Such advances have major implications in terms of investment and training for different members of the primary healthcare team.

    Copayments remain beloved of policymakers in search of a quick budgetary fix. Evidence from both developed and developing countries shows, however, that charges are most likely to deter those in greatest need.11 As a result, the later presentation of ill health may place a bigger financial burden on the health service, though this remains speculative. The scale of professional support for such charges is unclear.

    Clinical decision making

    By providing test results more speedily, technological advances in electronic data interchange and near patient testing may offset the need for some urgent referrals. As yet, however, the cost effectiveness of much of the available haematological and biochemical near patient testing equipment is unclear.12 Similarly, the benefits of the recent growth in outreach clinics is unclear. Easing access to specialists in this way may simply lower referral thresholds and add to rather than substitute for existing referral activity. General practitioners seldom use the opportunity outreach clinics provide for educational exchange.13


    Prescribing is the area in which the most sophisticated demand management is currently undertaken. Mechanisms in use in Britain include one to one “academic detailing” from medical or pharmaceutical advisers, feedback of accurate data on prescribing from the PACT (prescribing activity and cost) database, formulary development, and downward pressure on prescribing budgets.

    Plenty of evidence suggests that doctors overestimate patients' expectations: about one fifth of patients leave general practice consultations with prescriptions they did not expect.14 Conversely, those patients who believe that antibiotics are the answer to their presenting problems may exert a powerful effect on doctors' prescribing behaviour. Dissatisfied patients are more likely to reconsult. When doctors' perceptions and patients' preferences are not concordant, inappropriate prescribing and drug wastage may result. It takes time to explore patients' beliefs, reasons for consulting, treatment preferences, attitudes to drugs, and previous experiences, and doctors commonly use prescriptions to close consultations. Ironically, by reinforcing patients' misconceptions, this may fuel inappropriate demands.

    Surprisingly little evidence exists on the impact of guidelines on professional practice. Guidelines that are properly designed, disseminated, and implemented can successfully reduce inappropriate care. They need to be based on valid evidence, tailored to the local setting, and disseminated by active educational events involving local opinion leaders. When there is much unmet need, demand may increase. The related development of integrated care pathways also promises to reduce inappropriate resource use.15 By providing desktop access to guidelines and evidence as well as key details of local providers (such as waiting times), innovative software such as WAX may help reinforce their effectiveness.16 Finally, guidelines can provide the basis of audit.

    The interface with secondary care

    Many factors influence general practitioners' referral rates to secondary care. Researchers have sought unsuccessfully to explain variations in referral activity. A variety of doctor related, patient related, and service related factors have been found to explain only a small proportion of the variation. Recent attention has focused on individual practitioners' “referral thresholds” and their capacity to cope with uncertainty. Variation in this threshold between and within general practitioners probably accounts for a large proportion of the variability in health service use between general practitioners in small area analyses.17

    Three interrelated questions are relevant to attempts to change professional behaviour at this interface.

    • What is the purpose of seeking the specialist's opinio?

    • What are the alternatives to referral to secondary care?

    • What incentives exist for general practitioners to change current practice?

    General practitioners often do not make clear why they are referring a patient. When reassurance or advice on management is being sought, a simple telephone call may provide the necessary support. While it may not always meet patients' expectations, general practitioners seldom cross refer to colleagues with specialist skills and interests before referring to hospital. The incentives for doing so may increase in primary care groups. In the US reimbursement may depend on a second clinical opinion. Teleconferencing consultations may in future provide an effective alternative to routine outpatient referrals.18

    Referral criteria may ensure that needier patients get referred to hospital as a priority, reduce the number of inappropriate referrals, and better prepare those who are being referred. Joint ownership of such explicit criteria helps reduce the burden of refusal falling on the general practitioner. The informed, active involvement of patients and providers is an important part of this process.19

    In the UK large fundholding practices and multifunds have led the way in both monitoring and sharing details of their doctors' referral activity and in dividing labour within large teams to better exploit existing skills. However, evidence on the impact of feeding back to practices their comparative referral rates is limited. Low referral rates are not necessarily a sign of desirable practice. Given the intrinsic difficulties of interpretation, data alone will not have a significant or sustained effect on outliers. It may be more useful to concentrate on validated markers of quality care. The use of certain diagnostic tests (such as radiography for back pain) or prescribing patterns (such as the ratio of inhaled steroids to bronchodilators) may provide markers of cost effective practice. Further research is also required to explore whether outcome measures such as admission rates for acute asthma reliably reflect the quality of care provided.

    Proponents of US style managed care have imported plenty of rhetoric but little else. This may change: approaches to utilisation management adopted by health maintenance organisations may shape the development of clinical governance in primary care groups. Some of the more innovative proposals contained in the personal medical services (PMS) pilot schemes presage the vertical integration of health care, with great scope for changing the place of care. One perverse consequence of unified budgets may, however, be the growth of more income generating, low priority activities. Such reservations surrounded the boom in minor surgery following the contractual changes of 1990.

    A strategic approach

    As primary care groups take responsibility for devolved budgets and become freestanding bodies accountable to their health authorities, the incentive to manage demand more systematically will grow. They should be able to reinvest savings in patient care. However, both these arrangements and the personal medical services pilots present complex ethical conflicts to the general practitioner as both purchaser and provider. Doctors remain understandably wary of being held to account for cash limited budgets for services available on demand. Patients will need to be involved individually and collectively in decision making.

    Given the plethora of different approaches to demand management, there is a danger of overemphasising single approaches in isolation. Multifaceted strategies will be required to change doctor and patient behaviour at practice level. These need to be grounded in solid evidence and owned and reinforced by other members of the primary healthcare team.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    14. 14.
    15. 15.
    16. 16.
    17. 17.
    18. 18.
    19. 19.
    View Abstract

    Log in

    Log in through your institution


    * For online subscription