All in the family practice; ClintonCare's medical-specialty quotas.

"As the American health care system has become more complex, specialized, and technical," says the official summary of the Clinton administration's health-care plan, "it has neglected some simpler and, ironically, less costly needs....If the American health care system is to provide high-quality care at affordable prices, it must strike a better balance between physicians, nurses, and other professionals who take care of basic needs and those who provide the most sophisticated and specialized treatment for serious illness."

To many health-care reformers, the need for "a better balance" is clear. About 30 percent of U.S. doctors are considered "general practitioners," while in some countries the percentage is almost 70 percent. In Canada, a single governmental organization controls the number of specialists trained. But most Americans still believe that patients acting as consumers, and medical students themselves, should have some say in the matter. Bill and Hillary Clinton have decided to guide medical education onto a more enlightened path. In reply to a question from a medical student worried about being forced into primary care, the First Lady told the annual meeting of the Association of Medical Colleges, "There will clearly still be opportunities to go into specialties and subspecialties. But you know, it's about time we start thinking about the common good and the national interest, instead of just individuals, in our country." If the Clintons have their way, the federal government will decide how medical students will be trained and what they will do when they graduate.

In all the hype over ClintonCare's more conspicuous components--employer mandates, new taxes, restrictions on patient choice, etc.--the plan's new rules regarding medical schools and teaching hospitals have received relatively little attention. Yet the attempt to achieve the "right" mix of specialists and general practitioners has far-reaching implications that go beyond the obvious limits on the freedom of medical students to shape their careers. It would threaten the economic viability of the nation's premier medical institutions and undermine the quality of health care without reducing costs over the long run.

Policy makers have for years been trying to change the mix of general practitioners and specialists in the belief that more preventive care will help control health-care costs. Several states have established financial incentives and funded G.P. residencies to encourage students to pursue primary care, particularly in rural areas. Bills have been passed in New York and New Jersey limiting the number and types of residency positions, and legislation almost passed in California that would have allocated half of all University of California positions to primary care. In 1992, the federal government began phasing in a new fee schedule for Medicare that cut payments to surgeons by between 8 percent and 14 percent while increasing payments to family practitioners by about 15 percent.

Under the Clinton administration's plan, within five years at least half of new U.S. physicians would be trained in primary care rather than specialty fields. Primary care, according to the administration, includes family medicine, general internal medicine, obstetrics/gynecology, and general pediatrics. During the five-year period, the plan would increase the number of primary-care residency positions in the nation's hospitals by 7 percent a year. The number of filled specialty training positions in which "excess supply exists" would drop by about 10 percent annually.

Each year, Secretary of Health and Human Services Donna Shalala would determine the number of training positions that should be available in each specialty. The secretary would appoint a National Council on Graduate Medical Education to advise her in this process. HHS would also appoint 10 regional councils, reporting directly to the department, to allocate residency slots to each teaching hospital.

The federal government would "encourage" hospitals and medical schools to go along with these rules by changing medical education funding. The plan would offer about $6 billion, pooled from Medicare funds and a surcharge on health-plan premiums, to support medical training. The money would be available only to residency programs approved by the federal government. This funding mechanism is intended to replace today's system for supporting medical training, which relies on a combination of federal research grants and cross-subsidization of medical schools and residency programs by hospital bills. The administration also promises "transition assistance" for both doctors and hospitals, so they can retool to supply more primary care, and a loan-forgiveness program for medical students who go into general practice rather than specialties.

It is true that medical schools and teaching hospitals have a strong incentive to encourage their students to work in specialty fields. Medical schools get roughly 21 percent of their revenue from federal research grants, 42 percent from treating patients, and just 4 percent from tuition and fees. The federal grants are frequently tailored to studies of specialized treatments or procedures. And medical schools' patients are generally those seeking treatment by specialists for conditions their family doctor or local hospital could not adequately treat. So when these centers are pressured to emphasize general practice rather than specialties, they face potential losses of both research grants and patient revenue.

Meanwhile, the health of academic medical centers is already threatened by the more familiar parts of the Clinton plan: The Regional Health Alliance monopolies will elevate cost considerations above all else, harming academic hospitals that usually compete on the basis of quality and charge patients a little more to subsidize research and medical education. These medical centers are some of the most prestigious and innovative parts of American medicine. More than half of the 15 hospitals U.S. News & World Report ranked as "America's Best" in 1993 were university medical centers.

Even if these institutions could be protected, there is a more fundamental problem with ClintonCare's plan to push physicians into general practice. The administration's rules on medical training are motivated by a widely held belief that the key to reducing burgeoning costs and improving medical services is expanding so-called preventive care. Generally speaking, preventive care is any measure--including tests, examinations, diet and exercise regimens, and other behavioral changes--designed to head off a potential medical problem. If an ounce of prevention really is worth a pound of cure, as the old saying goes, then preventive care would seem to be the solution to America's health-care needs. By making use of relatively cheap, simple procedures, patients can avoid medical conditions that could require more expensive treatment, including surgery, later on.

But preventive care as the Clinton administration envisions it is not a reliable recipe for cost savings. Indeed, overconsumption of tests and procedures--encouraged by tax-free, low-deductible insurance and the resulting sense that such services are "free"--is a major driving force behind health-care cost inflation. "The fact is, we can't afford all the 'preventive care' that it is possible to consume," says John Goodman, president of the National Center for Policy Analysis in Dallas and co-author of Patient Power: Solving America's Health Care Crisis.

While the wisdom of providing generous preventive care to the American population is widely assumed, research in medical economics suggests a murkier picture. Dr. Jane Sisk of Columbia University's Department of Health Policy and Management reports that, "based on health effects, costs, and equity considerations, some preventive measures, such as adult and childhood vaccinations, are clearly underused. Others, such as cholesterol screening for elderly people, may be overused."

The economics of preventive care often depends on a doctor's response when a test comes back positive. One potentially deadly disease in the news lately is prostate cancer. High-profile deaths from the disease, such as those of actor Bill Bixby and musician Frank Zappa, have intensified a campaign to encourage screening and early treatment. But since the disease isn't usually fatal, it's not clear how often men should be tested and whether they should undergo radical treatments such as surgery or radiation.

Take the case of Tom Wirth, a 52-year-old Massachusetts architect whose father died of prostate cancer. Wirth was diagnosed in 1992 with some cancerous cells in his prostate gland. He asked five different doctors what he should do next and got five different answers. Finally, he decided not to take drastic action but instead to seek acupuncture therapy. "Each expert feels that his option is the panacea," Wirth told Newsweek.

Wirth's predicament is not unique. Doctors have no unanimous verdict on early treatment of prostate cancer, and many counsel patients not to undergo any significant treatments unless the cancer begins to spread. A 1988 study of U.S. urologists found that 80 percent favored radical prostatectomy for a man in his 60s with a tumor confined to the gland. On the other hand, 92 percent of radiation oncologists recommended radiation rather than surgery.

When highly trained experts in the field differ so widely in their views on early treatment--and tend to see problems from their own, sometimes narrow point of view--it is misleading and even dangerous to suggest that preventive care is a panacea for our health-care woes. In 1986, when the federal government first approved prostate-specific antigen testing, the American Cancer Society and the test's manufacturer began to call for all men over 50 to undergo the test. By 1990, the number of radical prostatectomies performed in the United States had risen to 16,000, more than all the prostatectomies in 1984, 1985, and 1986 combined. No one knows how many of those patients suffered unnecessary surgery.

In her 1986 book, Is Prevention Better Than Cure?, published by the Brookings Institution, Louise B. Russell cites a study showing that, over the long run, a concerted effort to prevent certain illnesses can generate medical expenses four to six times greater than simply letting the illnesses run their course. Partly this reflects that fact that many people who receive preventive care will not actually develop the anticipated illness, or at least will not develop the most dangerous form of the illness.

And if we must don our green eyeshades to measure costs and benefits to "society" of preventive care, as the Clinton administration suggests, then we should be honest: Expending resources to keep someone from developing an illness may well cost more than letting the illness end in death. Smokers, for example, far from costing the total health-care system money (as Hillary Clinton alleges), may actually save the system a little bit because they tend to die off before they get old enough to start running up huge medical bills.

John J. Whyte, a medical intern at Duke University Medical Center in Durham, North Carolina, offers a perspective on preventive care and general practice from the front lines. Contrary to popular belief, he says, "health promotion and disease prevention usually increase, rather than decrease, medical expenditures." He provides the example of treating high blood pressure. When patients come into Whyte's clinic at Duke, their vital signs are routinely taken. If their blood pressure is high, Whyte will repeat the procedure during that visit and, if it remains high, he'll check them again in two to four weeks. If a patient consistently demonstrates high-blood pressure, he or she will probably go on a relatively expensive medication and visit the doctor often, particularly during the first few months of treatment.

In the late 1970s, researchers at Harvard studied whether such screening and treatment for high blood pressure would save money. They found that for people with moderate or severe high blood pressure, the costs of treatment were four times as large as any savings. "Preventing disease involves risks, which are rarely mentioned, as well as benefits," Whyte notes. "Due to the imperfection of medical science, we sometimes treat people who do not need to be treated and neglect to treat those who do need treatment."

Setting aside the issue of preventive care, there is still an argument about the relative efficiency of specialists. The evidence that specialists contribute a large amount of waste to the health-care system is not clear. Several studies of medical efficiency currently underway, including the much-ballyhooed Medical Outcomes Study by a national team of researchers, have found that specialists are responsible for more costs per patient than primary-care physicians.

But these studies don't adequately adjust the data to account for differences in the severity of patient illnesses. People with more serious, and thus more costly, conditions are more likely to visit specialists. According to Dr. Roz Diane Lasker and Dr. David W. Shapiro of the Physician Payment Review Commission (a federal agency), generalists in the Medical Outcomes Study "may have seen a substantial number of patients with upper respiratory infections or acute low back pain, whereas cardiologists may be seeing patients referred for evaluation of syncope or the new onset of chest pain." This kind of difference is often not captured in the studies the Clinton administration cites.

Moreover, generalists and specialists are not as easily differentiated as the Clinton administration assumes. Americans are spending a lot more time in specialists' offices than you might think. About 20 percent of U.S. patients receive most of their continuing medical care from specialists, and almost 40 percent make most of their visits to a specialist. Many of these specialists already provide basic care to their patients in addition to more-specific treatments. And many managed-care networks are encouraging this trend by including in their preferred-provider lists only those specialists who can treat multiple conditions and who can provide primary care on site.

Market pressures are working against physicians who overspecialize. A new category of doctors--so-called primary-care specialists--is meeting the needs of patients in a more efficient manner. David Meltzer of the University of Chicago's Pritzker School of Medicine argues that this trend will reduce the need for referrals and multiple workups, particularly for patients with chronic conditions such as hypertension, diabetes mellitus, angina, and osteoarthritis--who account for a substantial share of total doctor visits. As managed-care networks continue to spread, Meltzer contends, "many specialists will find primary-care responsibilities surprisingly attractive."

There's good reason to have a lot of doctors who can provide basic care but who also specialize in one area. "Given the uncertainty about the appropriateness of many medical interventions," Meltzer says, "it is potentially important that specialists can keep up with advances in their areas of specialization that the generalist simply cannot."

Take congestive heart failure. Michael Bristow, a top researcher in the field who has run several clinical trials of experimental heart medicines, told a convention of science writers in January that between 20 percent and 50 percent of American heart-failure patients are being treated with ACE-inhibitors, the most effective treatment available. "The number should be 90 percent," he said, "but the majority of heart-failure patients are managed by family practitioners and internists, and, for whatever reason, the message isn't getting out." Bristow noted that patients seeing cardiologists are twice as likely to get ACE-inhibitors as those seeing family doctors.

Of course, patients may not have enough information to know which "primary-care specialist" they need to see. That's why they've traditionally relied on doctors in general practice to act as gatekeepers, to route patients to specialists and monitor their overall health and drag intake. But these gatekeepers probably don't need to be M.D.s. (See "Market Medicine," March 1992.) Even the Clinton administration has granted the need to allow nurse practitioners, physician assistants, and other non-doctor providers of health care more freedom to serve their patients. In most cases, restrictions on such providers are enacted by states, with plenty of support from doctors' associations. Lifting them would improve the efficiency of the system by matching the services patients need (gatekeeping or specialty care) with appropriately trained people.

Dr. Stuart Bondurant, dean of the University of North Carolina School of Medicine and current chairman of the American Association of Medical Colleges, is surprisingly blunt: "If you had a lot more non-doctor providers, then you wouldn't need so many generalists." Another trend that may reduce the need for G.P. gatekeepers is the computerization of patient records. Once each patient has a "smart card" that contains his medical history, a specialist may not need to confer as often with that patient's primary-care physician.

The goal of health reform should be savvy and cost-conscious consumers and a freer market for are, not a new distribution of producers imposed by Washington. Today's insurance-heavy funding system has weakened the connection between patients and doctors, encouraging the former to consume and the latter to prescribe as much care as possible. The answer to this problem, however, is not government planning to encourage one purportedly efficient form of care over another. "I don't think anyone knows for sure what the specialist quota should be," says Tom Reardon, a trustee of the American Medical Association and a general practitioner in Portland, Oregon, for 33 years. Dr. Gregory Bulkley, the Ravitch Professor of Surgery at Johns Hopkins Hospital in Baltimore, is even more direct: "No central planner can predict the future, and that's just as true for medicine as it is for anything else."

Instead, we should change tax laws and regulations to make it easier for patients to self-insure with Medical Savings Accounts and pocket savings when they forego care. In that way, all doctors--generalists and specialists alike--will be dealing with consumers spending their own money. That will lead to more-efficient consumption at the point of sale, whether the door outside the doctor's office reads "Family Medicine" or "Otorhinolaryngology."

There is evidence to suggest that giving people a choice between consuming care or pocketing their savings will generate the best results, even for preventive and primary care: A study published in the Spring 1990 issue of the journal Inquiry found that a $5.00 office-visit copayment for preventive-care services in a Washington state HMO resulted in a 14-percent drop in examinations but did not significantly affect immunization rates for young children, cancer-screening tests received by women, or medication use by people with cardiovascular disease. The study concluded that "small copayments appear to have little impact on the most valuable types of preventive care services."

Yet the Clinton administration distrusts the decisions of patients, preferring decisions made "in the public good" by Donna Shalala and her successors in the federal bureaucracy. It is a frequent mistake of big-government liberals to assume that universal rules, prescribed from afar, are better than sometimes messy personal decisions. They believe that an ounce of prevention is always worth a pound of cure--even when you're comparing an ounce of gold and a pound of dross.

Contributing Editor John Hood is vice president of the John Locke Foundation, a state policy think tank in Raleigh, N.C., and a syndicated columnist.