This is the second in a two-part series. To read the first part, click here.

In another 20 years, there may be no more primary care physicians, only midlevels. They will be the new general practitioners. For most people, who are healthy most of the time, that will work. But not for the patient with multiple chronic health problems—congestive heart failure, diabetes, arthritis, depression—someone like my father. Yes, the patient appears to be stable, but all it takes is one dietary indiscretion—a couple of cans of soup for a man who doesn’t cook—and he can’t breathe. Or he skips a meal and forgets to adjust his insulin. A cold in someone with emphysema, the kind of cold a healthy person might not even notice. Or someone who has no spouse to supervise her medications. Someone living alone with mild dementia.

When, as often happens, these patients end up in the hospital, disaster ensues.

Yes, telemedicine can help.  Gadgets that transmit a patient’s weight, or his blood pressure, to his doctor every day (provided there is someone at the office to look at it in real time),  and frequent telephone calls from a nurse or case manager are helpful. And the social supports recommended by this blog can make all the difference.

But there also must be a physician who knows and understands the diseases the patient has and how they interact, all the medications he has been prescribed, which ones he actually takes, and, most important, the patient himself—his usual state of health and his routines. The doctor who knows when another dose of diuretic at home will do the trick, or when the patient must come in, because another diuretic pill could tank his blood pressure. A doctor who can accurately describe an acutely ill patient’s problem to an ER physician or a hospitalist, and explain the purpose of an admission.

Sensible mid-levels will avoid these patients like the plague, not because they don’t care but because they are prudent. Like all good medical professionals, they know their competencies and their limits.

Such a patient is a job for an internist, or a geriatrician. There is no substitute.

Today, an internist should be the hospitalist outside the hospital, a trusted specialist who does not just coordinate care, but leads the team, integrates all the information from the specialists and the tests, and makes the important decisions with the patient and the family. As hospitals become more expensive, the system is moving sicker and sicker patients into the ambulatory setting, where they need closer attention, more sophisticated care and more time with the physician.

These patients are expensive and difficult to care for. They do not fit in our one-size-fits-all outpatient coding system, developed in the 1950s for acute, isolated, mild illness. (Back then, most seriously sick patients were hospitalized). Indeed, for many of the services today’s complex patients require—telephone calls, consultation with specialists and visiting nurses, family meetings, review of reports and records—the physician is not reimbursed at all. So the services do not happen, or the costs are made up by other patients’ routine physicals and colds. No physician can afford take on more than a few complex patients.

This is a terrible state of affairs as the Boomers begin to retire. For many like my father, there will be no one in charge.

A Solution: Technically Easy, Politically Not So Much

One approach would be to update the coding system to reflect modern medicine and today’s complex patients. In 1993, with doctors’ charges to Medicare soaring out of control, Congress designed a system whereby the Centers for Medicare & Medicaid Services, the agency that administers Medicare, sets the relative value of various physician services. For Medicare patients, these values are then adjusted by various factors, such as local cost of living, to arrive at the actual fees. Today, almost all private insurers have adopted the basic relative value scale, making their own adjustments. Of note, since 1993 the process has been controlled by an American Medical Association committee, called the Relative Value Update Committee, or RUC. Committee members are selected by the AMA; its proceedings are secret. After the RUC has evaluated or re-evaluated a physician service, it tells CMS how many “relative value units” the service is worth. CMS adopts 90 percent of the RUC’s recommendations unchanged. Thoughtfully, the AMA does not charge CMS for its advice.

Since the majority of the RUC’s members are subspecialists who do procedures, the RUC has always valued procedures much more highly than advice provided in the office, thought to be easy and unimportant. As we have seen, this has become a self-fulfilling prophecy.

Our problem is not the fee-for-service system, though: it is what fees we pay for what services. In a free market economy like ours, everyone knows that if we pay too much for a service, we get too much of it; if we pay too little, we don’t get enough. Because of the RUC, we pay too much for some procedures, so we get too many of them; we pay too little for complicated consultations, so we get too few.

Evaluation and management services for complex, chronically ill or frail patients are now difficult and extraordinarily important—along with complementary social supports, they keep such a patient alive (hopefully with a good quality of life), out of the hospital and free of expensive complications. If we reimburse internists and geriatricians for the true value of their analysis and counsel, they will be able to reduce the size of their patient panels, and spend more time with the difficult patients who need the sophisticated care that only they are qualified to provide. We can start by paying for services, like emails, calls, and consultations, that are currently ‘free’: the reasons for these anomalies have long since faded into history.

We could also risk-adjust the patients. This is not difficult—in fact, doctors do it automatically every time they look at a problem list, a procedure history or list of medications. For example, CMS could make a rule that if a patient has five of a list of 40 chronic conditions, the reimbursement should be one and a half times baseline; 10 conditions, twice baseline, and so on. There could be a premium for an after-hours visit, or a visit within two weeks of hospital discharge. Internists who wish can move from low value, high volume to the high value, low volume work for which they are trained.

To treat the complex problems of very sick patients, we need to be paid more.

This is not to say we should pay primary care providers more, and specialists less. It means we should pay more for analysis and advice that requires more training, and has more serious consequences for patients. As we have seen, much of what we call “primary care” today (eat less, exercise more) can be competently provided by mid-level professionals: primary care providers are compensated appropriately for these services now. However, following CMS, many payers reimburse all physician office visits within the same narrow range. A geriatrician who sees a moderately demented patient with many problems for 30 minutes makes little more than I do for seeing a basically healthy patient in the same time for the flu. If we want internists and geriatricians to properly care for our most difficult patients, this is what we have to change.

But that is not the fix the policy elite—Republicans and Democrats alike—have in mind.

Instead, they propose a per-member-per-month “coordination fee” of around $20, a modest step towards the capitation they have always favored. (Capitation is an alternative to fee-for-service: In a capitated system, insurers disburse one annual payment per patient to cover whatever services the patient needs—the doctor keeps whatever he doesn’t have to use for the patient).

But instead of focusing doctors’ attention on the patients who need the most care, a per-member-per-month coordination fee benefits the physician who attracts healthy customers, those who need the least care. Coordination fees encourage less scrupulous providers to cut corners, to take chances, to skimp on patient care. If they get away with it, they are rewarded for productivity, efficiency, for providing “value-based” care—nothing bad happened, the doctor made money, the insurer saved money.

Financing the Fix

Where will the extra money to care for complex outpatients come from?  For sure, there will be no new money for health care in the U.S.; voters are hearing from all sides that our health care system is on an unsustainable course and must cut back, not expand.

Policymakers like to think that the money for new initiatives will come from savings from lower hospitalization rates in a reformed system. Maybe. Those savings could just as easily go to fatten insurers’ profits.

More realistically, the money needs to come from lower fees for procedures, equipment, labs, drugs and other overpriced medical services. Because these areas are so much more profitable than office-based medical practice, they continue to attract disproportionate resources. If a medical student knew he could make, say, only 40 percent more—instead of twice as much—as a cardiologist than as an internist (as in some European countries), would it still be worth doing the extra four years of training?  If a manufacturer of MRIs knew it would only get 20 percent more for a slight but very expensive improvement in resolution, would it make the investment?  Suppliers in ordinary competitive markets make these calculations all the time. This is the real way to introduce market discipline to medicine.

And this is the reason the Europeans can insure all their citizens, and get better health outcomes, for a fraction of what we spend. In principle, it is easy; in practice, given politics in the U.S. today, maybe impossible.

But the alternative—ever more sick, frail people falling through the cracks, today my father, our parents, tomorrow us—isn’t that even worse?