General Practice

Primary care: core values Primary care in an imperfect market

BMJ 1998; 317 doi: (Published 18 July 1998) Cite this as: BMJ 1998;317:186
  1. John Roberts (johnroberts{at}, staff physician.
  1. Madrona Medical Group, 4465 Cordata Parkway, Bellingham, WA 98226, USA

    This is the fifth in a series of six articles reflecting on the core values that will underpin the development of primary care

    Series editor: Mike Pringle

    Doctors generally disdain the word “market” when it is applied to the work they do. A market is an encounter controlled by supply and demand.1 In most markets there is a purchaser, who pays for the specific goods he or she will receive, and a seller, who has the goods and will provide them to the purchaser.

    Medicine is an imperfect market. In health care the purchaser is usually not the consumer, and the goods provided by the seller are difficult to define and often contingent on other aspects of care such as results of tests and treatments.2 In addition, the medical marketplace does not follow the classic rules of supply and demand. Doctors (to a diminishing extent) set the demand of the care that they will provide and therefore can artificially increase demand for the goods they supply, as highlighted by Roemer's law: “The supply of beds creates the demand for those beds.”3

    Economic theory also assumes that the buyer-consumer will be knowledgeable about the goods to be purchased and can compare sellers' quality and prices. This is difficult in medicine. Firstly, seller-doctors have until recently controlled all the information about health care. Secondly, consumer-patients tend to avoid using medical services until they need them acutely, and by then, shopping is virtually impossible. Thirdly, even if payer-insurers or consumer-patients try hard to compare seller-doctors and their products, data are expensive to collect and complex to interpret. Finally, in the United Kingdom and in the non-urban and poorer parts of the United States, seller-doctors can set up monopolies or oligopolies in which neither purchasers nor consumer-patients can shop or even easily negotiate services or prices.

    Despite these exceptions, it has become increasingly clear over the past 20 years that medicine behaves as a market in many ways. Most obviously, doctors behave according to the rewards they are given. Insurers, whether private or public, can no longer afford unfettered inflation and have become active buyers willing to invest heavily in comparing costs and quality. In the United States, purchasers now have more information on the market than do the doctors who deliver the services. Consumer-patients, too, respond to market incentives. For example, requiring them to pay a small amount (say $10 at the time of medical service) decreases the use of emergency departments by 10-20%.4 5

    • Medicine is an imperfect market, and does not follow the classic rules of supply and demand

    • The primary care physician acts as a health broker

    • With the rise in managed care has come a parallel rise in the demand for primary care

    • Four market models—the integrationist, outreach, competitive, and managed market models—operate in the United States, and there is also the single payer model

    • The marketplace cannot eradicate the tensions between primary care and specialist doctors, nor can an imperfect model ensure highest quality medicine at the lowest costs

    primary care perspective

    Where does primary care fit into the discussion? Primary care directly accessible 24 hours a day is usually the patient's first point of contact with the medical system. The primary care physician should be the friend, philosopher, and guide of his or her patients, an advocate and protector and coordinator of appropriate specialist services. He or she should provide long term, continuing, comprehensive care. The primary care physician also acts as a health broker. Within the community, primary care can improve the health of the population through helping to remedy social pathologies; providing planned health promotion; screening for risk factors; preventing disease; collecting reliable data on the condition of a community; and helping the community to decide on priorities for health.6

    The term “health broker” describes how critical primary care is to any sane medical market. A short history of medicine in the United States shows why. In 1940, 90% of physicians in the United States declared themselves “generalists.” When the nation went to war, workers' pay was frozen, so medical insurance quickly became popular as a legal way to make jobs more attractive to workers in short supply. By the war's end, most workers were covered by indemnity insurance, which paid doctors for services rendered (fee for service). Not only did this new system reward doctors for testing and treating with little regard to costs, it also rewarded patients for seeing the most expensive doctors. These were usually specialists, who were trained to deal with patients whose probability of severe or unusual illness was greater and therefore required greater expenditures. Such unchecked consumerism led to massive cost inflation, the tremendous expansion of specialty medicine, and the near demise of primary care. By the 1980s only about a third of doctors in the United States called themselves generalists.2

    From consumerism to managed

    The inflation became so burdensome to employers paying for medical insurance that huge companies were being crippled. Chrysler, during the 1980s, was paying more for medical insurance than for the steel it used for its cars. One remedy was managed care, which had begun 50 years earlier, during the labour movements of the 1930s. In managed care, insurers (usually called health maintenance organisations) pay doctors a prepaid capitation amount for each patient that the doctor agrees to care for. In essence, payment is for people served, not for services delivered. The incentives are the converse of those of fee for service medicine, and they encourage doctors to spend less and patients to see specialists only rarely.

    The rise of capitated care has been slow, but in the past 20 years it has come to surpass the fee for service system in primary care. About 55% of Americans are now in some sort of managed care arrangement, and the number may jump to 75% as the government embraces managed care as the preferred public insurance mechanism for the poor (through the scheme Medicaid) and elderly (through Medicare).

    In general, primary care physicians in managed care behave much like general practitioners in Britain: they serve as doctors of first resort for nearly all medical problems and act as gatekeepers for patients' access to specialists. With the rise in managed care has come a parallel rise in the demand for primary care; specialists now find it difficult to find work in a nation oversupplied by doctors, while primary care doctors are still in great (though diminishing) demand.7.Doctors in training have recognised this new situation and, for the first time in several decades, have in the past four years been choosing the primary care disciplines (family medicine, internal medicine, and paediatrics) rather than specialty training.

    Five market models

    America, as its politicians are fond of boasting, is an experiment, and nowhere is this more true than in medical markets. Nothing about American medical systems is true throughout the country. Los Angeles and its managed care system is both ideologically and geographically a continent away from the southeast, where fee for service medicine still predominates. This somewhat chaotic nation of healthcare systems illustrates how primary care affects various medical markets and how they, in turn, affect the practice of primary care. There are at least four models in the United States, and another–the single payer (box).


    Embedded Image

    The integrationist model is the fee for service system that many doctors in America still cling to, particularly in many eastern states. It remains in place, in an attenuated form, for several reasons. Firstly, it is traditional, so doctors and patients (and even insurers, to some extent) are comfortable with it. Secondly, fee for service medicine can draw doctors to rural areas, where recruiting general practitioners is often difficult. Thirdly, managed care does best in urban areas, where people can travel short distances to various competing medical centres.

    Fee for service medicine has generally been detrimental for primary care because it rewards oversupply of services and allows patients to bypass the primary care doctor and go directly to the specialist. Obviously, development of community oriented primary care is impossible in such circumstances.


    The outreach model has become popular with academic and other tertiary centres, with their surpluses of specialists. The primary care doctor's surgery remains the centre of clinical activity, and specialists regularly attend sessions there. Such systems have become common in smaller cities, where there are relatively fewer specialists, and among overpopulated specialties such as gastroenterology, cardiology, and orthopaedic surgery. In the midwestern states, where distances between cities are great, this system has become popular. Major medical schools, such as those in Chicago and Minneapolis, send their faculty members out by aeroplane virtually every day of the week.

    For primary care doctors, this system usually works well. A concern is that the outreach system allows more patients with complicated or chronic diseases to bypass their primary care doctors. This trend has been tempered by those who pay for managed care, who believe that specialty care is costlier but rarely better for such patients.


    The competitive model is common in cities where competition for patients is high, not only between specialists but also between primary care doctors and specialists, and where there are too many doctors. The specialist might tell the asthmatic patient, “Next time you have an attack come and see me directly.” Such behaviour undermines the patient's relationship with the primary care doctor or, when the primary care doctor is meant to be acting as a gatekeeper to each specialist visit, creates outright animosity between the doctors.

    One response has been the creation of various types of multispecialty schemes, where ll doctors in a group (including some primary care doctors) share the risk of the costs incurred by the entire practice. If the specialist believes she or he can provide care of equal quality that is cheaper than care offered by the primary care doctor, so much the better, as long as the doctors are working closely together. However, most evidence suggests that specialists are more costly, even when the cost of the sicker patients they see is excluded.8 As specialists learn to become more cost effective this model will probably become more common in the United States.

    Managed market

    The managed market is where America has been heading and is where primary care becomes a true gatekeeper specialty. It is crucial to understand that in the United States at present, managed care is an extremely competitive and risky marketplace, with more than 1000 insurers trying to sell coverage at lower and lower rates while trying to get doctors to accept more financial risk for patients. The trend now is toward mergers among insurers, which, if carried far enough, will create a landscape that resembles a lot of mini-NHSs.

    Two trends are noteworthy. Firstly, the push towards forcing doctors to take on financial risk has caused the demise of the singlehanded primary care practitioner. Doctors are getting together in bigger groups, both formally and informally, to promote economies of scale in purchasing supplies and delivering care. Some are adding specialists, moving toward a variation of the outreach and competitive models.9

    Secondly, very few managed care organisations have convinced the clinical teams to take on all risks. So far, risk has been limited mainly to primary care doctors, who get about 10% of the premiums paid to the insurer (typically about $15 per member per month). Specialists and hospitals have continued to use mainly a fee for service system.

    Single payer

    The single payer system, as exemplified by the British NHS, is used throughout the world–except in the United States. Variations on the model are as numerous as the nations sponsoring these systems.

    The advantages and disadvantages of the single payer system are fairly obvious. Firstly, a nation can set a global budget and decide exactly what it will spend on health care each year. Proponents say that overall medical budgeting is a political issue; critics say this is rationing. Secondly, in systems such as the NHS that rely heavily on primary care, community oriented primary care becomes the norm, with each doctor responsible for maximising the health of all those patients on a defined, registered list. Thirdly, this list system creates some restriction in choice for patients, especially when, under reforms such as fundholding that require doctors to assume risk, there are incentives to underrefer. Fourthly, under global budgeting, some doctors are sure to suffer disproportionately: in Britain, specialists have to deal with long queues of patients awaiting appointments and elective procedures.

    Donald Light has congratulated the United Kingdom for its wisdom in creating its system of paying primary care doctors, pointing out that its three part system of paying capitation, operating costs, and bonuses for targets ensures that patients are neither overtreated (as in the American fee for service system) or undertreated (a potential risk of the for-profit managed care schemes in the United States).10 The single payer system does not, however, foster experimentation and entrepreneurship; if a better idea comes along it has to be implemented through regional or national bureaucracies.


    The marketplace cannot solve the problems of medicine, nor eradicate the tensions between primary care and specialist doctors. Nor can an imperfect model ensure highest quality medicine at the lowest costs.

    But in considering primary care medicine and the marketplace, it may be helpful to turn to a failed reform of the American system, that of President Clinton in the early 1990s. His task force, while realising that an imperfect market can never be made truly perfect, did list five criteria for an optimal medical market:

    • Universal medical insurance coverage

    • Costs that are affordable to society and to patients

    • Comprehensive medical benefits

    • Freedom of patients to choose their own doctors

    • Public accountability, both in cost and in quality of care.

    Unfortunately, these five statements are probably mutually exclusive in practice. But they remain a goal for all of us to consider as we continue to reform our own medical marketplaces.


    Funding: None.

    Conflict of interest: None.

    • This article has been adapted from Primary Care: Core Values, edited by Mike Pringle, which will by published by the BMJ Publishing Group in July.


    View Abstract