An Evidence Base For Primary Care: A Conversation with Barbara Starfield, MD, MPH

MANAGED CARE June 2008. ©MediMedia USA

A physician-turned-academic-researcher builds a framework for challenging the health system's focus on specialists

For someone who would like to see the American health care system change almost completely, from a system dominated by specialty care to one that has primary care with appropriate secondary care backup as its core, Barbara Starfield, MD, MPH, spends a lot of time overseas. After earning a bachelor’s degree from Swarthmore College, a medical degree from the State University of New York, and a master of public health from Johns Hopkins University, she spent much of her career in the U.S. health services system, but also learned from experiences abroad.

Today, after years of studying the U.S. health system and as director of the Johns Hopkins University Primary Care Policy Center, she gives talks internationally about the goals of health systems and about how the apparent unwillingness of the United States to focus attention on primary care is compromising the health of the population, contributing to rising health care costs, and leading to poorer overall outcomes and greater inequities in care. Her work has been used by governments in more than a dozen countries to pass legislation and to gain support for primary care. Starfield recently spent several days in Geneva reviewing a new report being issued by the World Health Organization on primary health care.

At home, employers and several groups of primary care practitioners are using the evidence she’s compiled to call for greater emphasis on primary care. Yet while she is encouraged by the attention primary care is receiving, she is concerned that the United States does not have a unified movement or the political support necessary to move to a system based on more appropriate and rational use of health care resources. MANAGED CARE Editor John Marcille spoke with Starfield about why that is and whether she is hopeful that it will change.

MANAGED CARE: What does it mean for our country that two thirds of physicians are specialists?

BARBARA STARFIELD: It not only raises the cost of health care enormously, but it does not contribute commensurately to the health of the population. And it does not contribute to the equity of health care services provided to different populations. When I started studying this with 1980s data, the United States was somewhere around tenth in the world in terms of health indicators. Now it’s between twentieth and thirtieth, with costs that are much higher than any other country and more than twice as high as most other developed countries. The United States has been falling progressively further behind in most health indicators for at least two decades. So we are doing something really, really wrong, and one of the things we are doing wrong is using specialists for services that should be provided by primary care.

MC: As a consumer, I would think that going to the most highly educated doctor with the best imaging devices and other equipment would be the best thing to do.

STARFIELD: The elderly in this country are much more likely to go to a specialist than a primary care doctor, but I once spoke to a group of about 100 senior citizens in Baltimore and everyone was able to understand that inappropriate care is not good care. They could see the point. And the point is this: Specialists are trained in hospitals. They don’t get a representative picture of the way illness presents in the community because the population they are trained on is unusual. The patients they see are people who are referred to academic medical centers because their conditions are complex and unusual. So when physicians are finished with their training and begin to evaluate symptoms on their own, they are much more likely to suspect something that is serious. However, someone trained in the community will know that most symptoms don’t mean a zebra is present. They generally mean horses are present. But the specialist, oversuspecting the likelihood of a serious illness, will do a very expensive and unnecessary workup. And all the things that are done in an unnecessary and expensive workup have a finite chance of an adverse side effect, including death. So people who go unnecessarily to specialists will have very high costs, will have unnecessary things done, and are placed at risk. That’s why, if you don’t have something that requires a specialist, it’s dangerous to see one. It’s not difficult to convince people of this; it just has to be put in the right context.

MC: Does your training as a physician give you a unique perspective on this issue?

STARFIELD: Yes, it does. When I went through my training, something in me recognized the fact that the experience I was getting in the hospital was not going to be very useful for practicing in the community. So I arranged with my professor to do something different in my last year, and I trained with outpatients rather than inpatients. These ideas have been with me for a long time.

MC: What are the root causes of the problem?

STARFIELD: It goes back about a century. We have never in this country had a strong primary care education model in medical school. In about 1910, Abraham Flexner published an influential report on medical education. It was a good report, but it had side effects. It said that medical education needed a science base, and it resulted in the closing of all the proprietary medical schools. It set in motion a focus on the biomedical in medical education, and from that time on, medical schools were built on a specialty basis. It grew worse after World War II. Physicians in the service had learned a lot during the war about orthopedics and other specialties, and when they returned, the government paid hospitals to train them as specialists. It has been downhill since then.

MC: What are the three best things we can do to turn this around?

STARFIELD: We need another Flexner Report. Flexner was a smart man, and he realized that medical education should take social phenomena into account. But he didn’t believe we had enough scientific knowledge about it to incorporate it into medical education at the time. But if we didn’t have scientific evidence for the social basis of medicine then, we have it now. So another Flexner Report would balance the specialty focus with the primary care focus. The second thing we can do is change our reimbursement policies. That has to start with the federal government because of the Medicare program. We have to start paying primary care physicians and specialists equally. The third thing is to start putting into practice what we know about primary care. We know exactly what primary care is and what it accomplishes, and we have to start using that evidence to change the way physicians practice and refer.

MC: How do you define the primary care mandate?

STARFIELD: Primary care has four functions, and you have to have all of them to have good primary care. The first is that people have to know that they can get access to their primary care doctor when they have a new problem. The second is that care has to be person-focused, not disease-focused. We tend to think about priority diseases, but we are making diagnoses earlier all the time, so even young people and middle-aged people have multiple morbidities. You need person-focused care over time — not disease-focused care, because most people have mixtures of different types of illnesses. The third function of primary care is comprehensiveness. That means primary care has to address all the common needs in the population and only refer to specialists what specialists are trained to do, which is not routine follow up. Half of what specialists in this country do today is simply routine follow up, and that’s a huge waste of expertise and money. And the fourth challenge is to integrate care. People will have to go to specialists when they have uncommon or unusual problems, and it’s a primary care physician’s job to integrate that into ongoing care.

MC: What should primary care physicians do that they are not doing today?

STARFIELD: Our biggest challenge is to make primary care more comprehensive. Primary care physicians here don’t provide care that primary care physicians in other countries do, such as minor surgery and minor orthopedics. In most countries, people are not referred to specialists for such procedures as suturing, joint aspirations, and routine casting. In this country, we now have more visits to specialists than we have to primary care physicians. Primary care is getting less and less comprehensive.

MC: Are primary care physicians equipped to provide those kinds of services?

STARFIELD: They could be. Family physicians are much more comprehensive in the care they provide than are general internists and general pediatricians. But they can be taught to do it.

MC: Does all of this tie into the medical home concept?

STARFIELD: The patient-centered medical home, the PCMH, that is being proposed by the four primary care societies — the American Academy of Pediatrics, the American Academy of Family Physicians, the American Osteopathic Association, and the American College of Physicians — relies on the evidence we have produced that primary care is useful. Everyone says to me, “You should feel terrific. Finally you’ve gotten your message across.” And I guess I do feel terrific that my work has been recognized. My worry is that it’s going to be misused.

MC: What is the potential for misuse?

STARFIELD: Focusing on the wrong things for primary care, like leaving comprehensiveness out of discussions and focusing on chosen diseases instead of on dealing with peoples’ problems. These physician groups agree on the four functions of primary care, and those four functions are in their mission statements. But as the medical home concept is being implemented, some things are getting lost in favor of bells and whistles that are not evidence-based.

MC: Bells and whistles?

STARFIELD: Electronic health records are an example of something that would be helpful if they contribute to the four functions of primary care, but no one is making an effort to provide evidence that they will. I am also concerned that the instruments being developed to certify physician practices as being a PCMH are completely missing the comprehensiveness function. They address access issues that are related to first contact, they have elements of patient-centeredness — although they sometimes define that as patient-focus in a disease context — and they have elements of integration, but they are completely missing the comprehensiveness side.

MC: This is the first cautionary discussion of the medical home concept that I’ve ever heard.

STARFIELD: Others are worried, too, especially in the business community. They are strongly in favor of primary care as we have defined it.

MC: Employers seem like a natural audience for your work, especially those that are self-insured.

STARFIELD: Last March, I received an award from the National Business Group on Health for excellence and innovation in value purchasing for promoting primary care. Can you imagine? They gave an academic an award for innovative purchasing. I’m not even an economist. I’m a physician and I understand primary care from practice experience and from research.

MC: That must be encouraging. Do you see other evidence that things will change?

STARFIELD: Many physician groups talk about reform, and we have seen a very large physician and allied health movement toward a single-payer system, but we don’t see a unified movement that is directed at changing the way we train physicians and the way we organize services. Rural areas have always cared about the shortage of primary care physicians, but now the shortage is everywhere. Still, there’s no unified movement that says, “Do something about it!” We see many splinter groups talking about it, but as long as the vested interests can divide and conquer and prevent a unified movement, we are not going to get it.

MC: Where do we chip away at this?

STARFIELD: Congress is one of the places you have to chip away. We need national health policies. You have to chip away at it with the Centers for Medicare & Medicaid Services. You have to chip away at it with the foundations, who could mount another Flexner Report. It was a foundation that did the Flexner Report to start with. The academic medical establishment also has to be forced to defend its notion that we need more and more specialists. They don’t defend it on the fact that we need it. They defend it on the grounds that people are demanding it. If they are going to go with demand, we will never control our costs, and we will never get a healthier population. So we chip away at those several places.

MC: The major presidential candidates use the word change quite a bit.

STARFIELD: The candidates are not talking about the health care system in the United States. They are only talking about insurance.

MC: But payment shapes everything, doesn’t it?

STARFIELD: Payment shouldn’t shape things. You have to have ideas that shape things, and then find a way to pay for them. Obviously, everybody needs to have access to care, but it’s not only financial. It has to be appropriate care. Nobody’s talking about appropriate care. They are only talking about ways to pay for care. And that’s why our health care costs are so high, because we are focusing only on paying without asking what we are paying for.

MC: You have written and talked about the health care systems in other countries. Can America learn from these other systems?

STARFIELD: I started my work with international comparisons, and then moved to looking at differences across states. We have so much evidence now from the United States that we don’t need to argue that we should do things the way the Brits or the Dutch do. It’s useful, though, to understand that our health is poorer than theirs. It’s useful because it forces us to think about why. And the answer is not in bad behaviors. Our population actually behaves better than populations of most countries in terms of smoking less and drinking less alcohol, for example. It is also not because we have major ethnic populations or racial populations, because even if you take out

   the data from those populations, we’re still worse off. We know that it has to do with our public policies toward health services.

MC: The Veterans Administration has moved to a primary-care model. Can we learn any lessons there?

STARFIELD: We could learn a huge amount from the VA. The fact that they made a change about 10 years ago that just completely reoriented the way they provide care suggests that this doesn’t have to be a slow process of change. You can do this relatively overnight. It won’t take a long time once you have the motivation to do it. The VA is a single-payer system, of course, and it is probably not generalizable because it’s not a cross section of the population, but I would like to see a commission formed to think about what from the VA experience can be expanded to the whole population.

MC: You seem to have a high-regard for the single-payer model.

STARFIELD: Yes, I do, because it’s going to focus attention on that single payer to do something. As long as you fragment the insurance system, nobody is going to take care of the delivery system It’s not that inherently the single-payer is better, it’s just that we don’t seem to be able to do it without the single payer.

MC: Politically, is it possible to move toward such a system?

STARFIELD: People like it when they have it. Medicare and the Federal Employees Health Benefits Program are single-payer systems. Unfortunately, the imperatives of the multiple-payer systems influence even what Medicare and the FEHBP do. But all the publicity for the last 50 years has been on what a great health system we have in the United States, and people believe that because they are never exposed to anything else.

MC: Did managed care’s use of the gatekeeper model harm or promote the idea of primary care?

STARFIELD: It should have been framed in the context of appropriate access to specialists, instead of keeping people away from specialists. We need specialists. Primary care physicians don’t get enough experience seeing unusual and rare things, and patients do have to be referred. But we’ve gotten to the point that we are way, way over-referring. The answer is not keeping people away from specialists, it is making specialist care appropriate.

Payment shouldn’t shape things. You have to have ideas that shape things, and then find a way to pay for them. That’s why our health care costs are so high, because we are focusing on paying.