Summary: Health care reformers have been promoting access to primary care as the answer to lifting the quality of care. If we had more primary care physicians, patients would be able to see them on a regular basis, and they would be less likely to wind up in the hospital–or so the theory goes. But it’s not quite that simple.
A report released today by Dartmouth’s Institute for Health Policy & Clinical Practice reveals that when it comes to managing chronic diseases Medicare beneficiaries who live in regions of the country where patients typically receive more primary care, fare no better than patients in other regions. This suggests that while primary care is important, it is effective only if it is part of a larger system of coordinated care.
More primary care is not necessarily better care. That’s the first surprise. The second is that there is no simple relationship between the number of primary care physicians in a given area and the share of patients who see an internist on a regular basis. In fact, the study shows “no correlation” between the “supply” of primary care doctors and access” to primary care. In cities such as Boston, where the number of internists per capita is high, patients still have a hard time making an appointment. This may be because physicians are seeing their regular patients more often—leaving little room in their appointment books for new patients.
If the goal is to improve the population’s health, while simultaneously reining in the cost of care, health care reformers should focus less on increasing the supply of primary care physicians, the report’s authors suggest, and pay more attention to turning what we euphemistically call a health care “system” into a coordinated delivery system where internists, nurses, specialists and hospitals collaborate to deliver more effective and efficient care.
~~~~~~~~~~~~~~~~~
The emphasis on primary care as the “key” to lifting the quality of U.S. healthcare may be exaggerated according to a report, released today, by Dartmouth’s Institute for Health Policy & Clinical Practice.
“Primary care forms the bedrock of a well-functioning, effective health care system,” the researchers observe. But– and this is an important caveat- “simply increasing access to primary care, either by boosting the number of primary care physicians in an area or by ensuring that most patients have better insurance coverage, may not be enough to improve the quality of care or lead to better outcomes.”
Wait a minute. In past reports, didn’t Dartmouth’s researchers tell us that patients fare better if they see fewer specialists and more internists?
No. Dartmouth’s earlier studies have shown that when patients see more specialists, care is more aggressive and more expensive, but, on average, outcomes are no better—and sometimes they are worse. This, however, doesn’t mean that primary care, by itself, ensures better care, even if a patient sees her PCP on a regular basis.
As the report points out: “Primary care is most effective when it is embedded in a high-functioning system, where care is coordinated, where physicians communicate with one another about their patients, and where feedback is available about performance that allows physicians and local hospitals to continually improve.”
Policy should “focus on improving the actual services primary care clinicians provide and making sure their efforts are coordinated with those of other providers, including specialists, nurses and hospitals,” says Dr. David C. Goodman, lead author and co-principal investigator for the Dartmouth Atlas Project.
That said, the study’s authors (Goodman, Brownlee, Chang and Fisher) agree that primary care is essential: “Primary care clinicians, whether they are general internists, family practice physicians, pediatricians, physician’s assistants or nurse practitioners, are trained to care for the whole patient. They can diagnose and treat a wide variety of illnesses, help patients avoid getting sick, and ensure that they get the specialty care they need. For chronically ill patients in particular, primary care clinicians serve a crucial role as coordinators of specialty care. They can also help patients control symptoms, slow the progression of their disease, and help manage acute and chronic conditions without resorting to hospitalization.”
But while primary care can do all of these things, this does not mean that it does.
Geographic Variation
This new study surveys access to primary care—and use of primary care— among the fee-for-service Medicare population in different regions of the country from 2003 to 2007, only to find, once again, that geography is destiny. As the map below reveals, Americans in some parts of the U.S. receive far more primary care than others. During the report period, the share of patients who saw a primary care physician on an annual basis ranged from roughly 60 percent of beneficiaries in the Bronx, N.Y. and Manhattan to nearly 90 percent in Wilmington, N.C. and Florence, S.C.—about a 50 percent difference.
(click map to enlarge)
More Primary Care Does Not Guarantee Better Management of Chronic Diseases
Yet, here is the first surprise: in those regions where patients have more access to primary care physicians, “this alone does not always keep people with chronic conditions out of the hospital, improve their chances of getting the optimal care recommended for their condition, or improve health outcomes.”
For example, when researchers looked at patients suffering from diabetes they found no relationship between rates of blood lipid testing and eye examinations and whether these beneficiaries with diabetes saw a primary care clinician at least once a year.
There also appeared to be no connection between rates of leg amputation, a serious complication of diabetes and peripheral vascular disease, and whether the beneficiary saw an internist at least once a year. But a patients’ risk of losing a leg did vary dramatically depending upon where he lived –the report reveals a tenfold difference in the rate of leg amputation, ranging from 0.33 per 1,000 beneficiaries in Provo, Utah to 3.29 per 1,000 in McAllen, Texas—the town made famous for over-treatment in Dr. Atul Gawande’s 2009 New Yorker story.
The report also found that having an annual primary care visit did not keep patients suffering from diabetes or congestive heart failure out of the hospital.
In this case, was a more than fourfold difference in the rate of hospitalizations among Medicare beneficiaries, ranging from 30.7 per 1,000 in Honolulu to 135.0 per 1,000 in Monroe, La. (This could be tied to the fact that Louisiana boasts more physician-owned hospitals and surgical centers than any state except Texas. Research shows that when doctors own hospitals, patients are more likely to find themselves in one of them.)
The researchers theorize that “perhaps primary care visits aren’t doing more to improve outcomes” because “the patients most in need of this care are not receiving it.”
But another possible explanation seems, to me, more persuasive: “primary care is most effective when it is embedded within a health care system that allows the coordination of primary care services with those delivered by specialists and hospitals,” the researchers observe. “Unfortunately, most health care providers in the U.S., including primary care physicians, are not organized to do this; many
physicians work in small practices, where there is little coordination of care, and communication among a chronically ill patient’s various physicians is often poor to non-existent. Large delivery systems can also fall short in these areas. The quality of the care provided by primary care physicians also varies widely. As a result, patients in regions of the country where they are more likely to have had a primary care visit are not necessarily receiving higher quality care—or enjoying better outcomes.”
Medical cultures vary widely around the nation. In some places, doctors are more likely to work in large mutli-specialty centers where collaboration is a top priority. In other towns, solo practitioners pride themselves on their autonomy. They may play phone tag, but most don’t use electronic medical records—and if they do, these records can’t “talk” to each other.
It’s Not about the Supply of Primary Care Physicians
More primary care does not necessarily mean better care. Perhaps that shouldn’t come as such a surprise. PCPs alone cannot solve the nation’s health care crisis. If we want to keep patients out of hospitals, PCPs and specialists must work together—and they must listen to patients and their caregivers.
If we want to reduce the number of diabetics who wind up losing a leg we should look at the larger problems that affect a diabetic’s ability to manage his disease, putting poverty at the top of the list. Indeed, the report points out that rates of leg amputation for all Medicare beneficiaries differed by a factor of 10. When researchers took a close look at 44 hospital service areas (HSAs) within a single
HRR, Atlanta, Georgia, they found a fourfold variation in leg amputation rates.
“Addressing these disparities in health outcomes will require attention to the full spectrum of health determinants,” they write, “ranging from lower levels of schooling and limited health literacy, to inadequate housing and lack of transportation, as well as lack of access to high-quality primary care that is well-coordinated with specialty care.”
That primary care is not a cure-all probably shouldn’t come as a shock. But the report’s second surprise is eye-opening: having more PCPs physicians –more general practitioners, internists and pediatricians–does not necessarily mean greater access to primary care: “Our findings suggest that the nation’s primary care deficit won’t be solved by simply increasing access to primary care, either by boosting the number of primary care physicians in an area or by ensuring that most patients have better insurance coverage,” says Goodman.
A shortage of PCPs “may contribute to the problem in some locations,” the report notes, but “there is no simple correlation between supply . . . and access,” says Dr. Elliott S. Fisher, a report author and co-principal investigator for the Dartmouth Atlas Project.
The study shows that, in some regions, a relatively high proportion of beneficiaries saw an internist at least once a year, even though overall primary care physician supply was low. This includes Wilmington, N.C., where there were 69.0 primary care physicians per 100,000 residents and 87.4 percent of patients had at least one annual primary care visit. Meanwhile, despite an abundant supply of PCPS in White Plains, N.Y. (101.4 per 100,000), less than 70 percent of beneficiaries saw a primary care clinician each year.
Here, I suspect that the residents of White Plains are simply more likely to go directly to a specialist, rather than consulting a primary care physician first. White Plains is a suburb of New York City, and New York boasts an embarrassment of specialists. (I know, from personal experience that I can usually get an appointment with a Park Avenue specialist within a few days—even if I am a new patient.) Patients in the New York area tend to believe that more expensive is always “better” and of course specialists are more expensive. And while some Americans are wary of “experts,” New Yorkers tend to like the idea of consulting someone at the very top of the food chain. (Many Manhattanites consider themselves experts of one kind or another.)
In addition, primary care doctors in New York, like PCPs in Boston, may keep their waiting rooms crowded by seeing regular patients more often than doctors in some other towns. Again, this is part of the medical culture in Manhattan. If you want to re-fill a regular prescription, your doctor will insist that you come in and see him every three months. If your insurance requires that you get a referral to see an eye doctor, your PCP will tell you that you must come in for an appointment first. Even though he’s not going to exam your eyes, he’s not going to be paid for making the referral unless you see him.
As a result, it’s difficult to squeeze in an emergency—or a new patient. If you call and say you’re experiencing chest pain, your internist will tell you to call an ambulance and go to the ER. (This happened to a friend recently, and stands as an example of the aggressive, expensive approach to health care that Manhattanites have come to expect. He was fine; it may have been indigestion.) Boston is much like New York. A friend there told about the time he cut his hand. His internist couldn’t fit him in. His wife’s primary care doctor couldn’t see him. He wound up having his sister sew up his hand on her kitchen table. My guess is that if my friend lived in Wilmington N.C., and he called his internist, the doctor’s receptionist would say “Come on in—we’ll stitch it up.”
I am, of course, speculating. And these are only anecdotes, but anecdotes can illustrate a medical culture. In addition, Dartmouth data confirms that New Yorkers see many more specialist than patients in Iowa—though our outcomes are no better. The problem in New York so much a dearth of internists as the fact that healthcare in New York City is so fragmented: most of our specialists work solo or in small practices. They value their independence. Many bristle if hospitals try to suggest “rules” or even “guidelines” for best practice. Thus, most surgeons don’t use checklists, even though there is ample evidence that a simple piece of paper can save lives.
Ultimately, this newest Dartmouth study suggests that healthcare, like real estate, is all about “location, location, location.” As the authors put it: “This report highlights the importance of understanding health care within a local context and underscores the need to address the underlying causes of . . . disparities both within and across regions.”
Where You Live Is Paramount
The report acknowledges that both lower-income Americans, and minorities receive less care. Put bluntly, the quality of care you receive varies, depending on who you are. “On average, blacks were less likely to see a primary care clinician than whites—70.4 percent had at least one annual visit in 2003-07 compared with 78.1 percent for whites.” But regional disparities are far greater: where a patient lives turns out to be even more important than the color of his skin. “In the U.S. health care system, it’s not only who you are that matters; it’s also where you get your care,” the authors report. “Regardless of race and income, patients receive care of widely varying quality depending upon where they live and the health system that provides their care.”
If you’re very lucky, you live in a place where the medical culture favors “collaboration” over “competition,” a town where general practitioners, specialists, and hospitals understand that medicine is a team sport. Too many primary care doctors labor alone—working long hours without sufficient support. This is one reason why being a primary care doctor is so difficult. I recall Dr. Donald Berwick , Medicare’s new director, once saying, “No doctor should be alone.” The job is too hard. Insofar as there is a single “key” to raising the quality of U.S. health care, “co-ordination” is, I think, the word to keep in mind.
unclear to me what’s surprising here. there was a hypothesis that care would be more efficient in areas where there were more primary care physicians. there has never been any evidence presented to support this hypothesis. which leads me again to pose my chronic question– what’s the justification for spending more $s to create more primary care docs or pay existing ones more? is there an iota of evidence this would be a better investment than building more minute clinics and increasing the supply of nurse practitioners while expanding their powers?
The real point is we need more primary care docs coordinating care among the various specialists: a director for the orchestra as it were.
If primary care docs are going to shoulder the burden of coordinating care, keeping track of what the specialists are doing, and nixing treatments that are counterproductive, ineffective, or whose risks outweigh the benefits, then primary care docs indeed need to be paid more for what they do.
For that to happen though, Medicare and insurers need to start paying primary docs for consultation visits; 30 minutes or more rather than reimburse based on tests/procedures.
Panacea & Jim
Panacea– you write: primary care docs need to be “a director for the orchestra”– exactly.
I also agree that the lowest paid primary care docs should be paid more—young primary care docs who usually have huge debts, and those working in the most difficult places–community clinics in inner cities and poor rural areas tend to be paid least.
I think staring salary for primary care should be much higher, and primary care docs working with poorer patients–who are more difficult and need more care– should earn more. Yet, they tend to be at the bottom of the income ladder.
That said, median income for primary care docs is now about $185,000, according to the AMA. This means that half of all primary care docs earn more than that. Most of those are doctors who are further along in their career, have paid off their med school bills, and work in affluent areas where working conditions are easier. Many don’t accept insurance and charge $200 to $300 for a 20-minute appointment–not a physical, just an appointment. (This is in Manhattan.) I don’t begrudge their $250,00 to $350,00 incomes–if people are willing to pay that, fine. (Though I think insurers should push back) But I don’t think we need to raise them.
But raises are definitely needed at the lower end of the income ladder–particuarly for the youngest doctors, trying to pay off loans, as well as those who see many more Medicaid and Medicare patients.
Jim–
Having more primary care docs does not, in and of itself, improve care.
There is, however, evidence that in areas where there are more specialists, they add greatly to the cost of care–without improving outomces.
“Minute clinics”– urgent care centers, or “retail healthcare” in Walmarts definitely are not the answer. No continuity of care, thus, no chronic disease management.
Chronic diseases account for more and more of health care spending. And as more baby-boomers dodge death by heart diease and death by cancer, more will
live long enough to die of long, lingering chronic disease (Alzheimers, etc.)
Very good primary care physicians, who are working in a place (this could be a town or a multi-specialty center) that emphasizes collaboration among PCPS, specialists, nurses and hospitala are able to have a huge impact on improving care.
This is one reason why outocomes are better (at a much lower cost) in many European countires where the ratio fo PCPS to specialists is significantly higher, and where they all work together, on salary.
These European countires also tend to make much better use of nurse practioners.
I toally agree with you that nurse practioners can help manage chronic diseases. But they cannot do it alone. If the specialist doesn’t call back–and doesn’t have the EMR that allows him to look at the same medical record that that the nurse practioner is looking at–he and she cannot collaborate.
The patient will end up in the hospital– where the tests that either the primary care doc or the specialist ordered are done all over again.
Why? The data doesn’t back it up why they should be paid more. If PCP’s aren’t going to take call or see patients in the hospital. Why do they need to be paid more? Why do we need more of them? If there is no benefit.And if they are just going to coordinate and not take the risk of treatment why do they need to be doctors?
Maggie….Thank you for discussing this important study.
Boy…I find this really discouraging. I hope the study design was flawed!
I have worked in both an integrated large medical group (all connected by one emr) and as a private practitioner. I have not found communication to be better in a big group.
Maggie,
You seem to have a strong opinion in regards to yor ideal model of care for which you are using this study to support. I think this is a gross misreading of what this data indicates or at minimum what conclusions can be drawn. Given the other variables for which there are no controls, including socioeconomic makeup of each region, the availibiity of specialty care, the number of hospital beds, etc., etc., I don’t see how you reach your conclusion that better coordination is the answer. Although institutions like Mayo are held up as the ideal health system, I have yet to see data comparing it to other economically well endowed health systems serving a similar demographic population. What does exist is showing Mayo provides quality for a cheaper price; not that the quality is superior to many other systems.
While intergrated medical records will indeed serve to better coordinate care by allowing the entire medical team access to records, this does not require that physicians be employed directly by health care systems to acheive this goal. Point in fact, health care systems can also be used in a manner that limits access to data for those who choose to oppose joining a given health care system, and I am indeed suprised that you are not concerned about the potential for market concentration and dominance that these systems may come to possess, possibly leading to monopolistic behavior and possibly further driving up the costs of health care.
You make several allusions to the problems in the study; some areas may simply be compensating for less less PCPs by patients going directly to specialists for instance. You also conveniently ignore the fact that many primary care docs have closed their practices to Medicare patients. Statistics show this tends to be the only area where access problems are developing for Medicare patients. Here again, the saftey valves may be the ER (are they primary care physicians?) or specialists that are seeing these patients.
The big difference in all this is not quality, but value. Many of these patients can be taken care of at a much lower cost if they are first consulting their primary care doctor rather than going to ERs or specialists. I do not suspect the quality measure will change much whether a diabetic is being treated by an endocrinologist or by a PCP, at least for the uncomplicated early stage diabetic. It may cost more with the specialist however. What we need is a shift away from highly compensated specialists to more PCPs (like in Europe)and a payment scheme that does not reward ordering more tests and performing more procedures. We have enough docs; they are just poorly distributed into predominantly specialty practices because this is what our current system has rewarded financially. It may not be a question of paying PCPs more, but maybe paying specialists less. But at the current wide chasm of compensation between PCPs and specialists, do not expect to get more young docs to go into this area of medicine. Better to spend alot less time in training and become a nurse practitioner!
Keith–
First of all, until I read this study, I too, thought that more primary care was a major part of the answer to better chronic disease management.
So the study isn’t corroborating a preconceived notion I had about an ideal model of care.
For a variety of reasons, it seems that PCPs aren’t managing chronic diseases as well as we hoped. This may well be because the typical 15 minute appointment just doesn’t provide enough time to adddress the issues. It may be because, these days, many PCPs describe themselves as “burned out” and extremely unhappy with working conditions.
Research shows that primary care docs who train and work at large mutli-specialty centers are happier in their jobs–presumably because they have more regular hours, don’t have to worry about paying rent, getting reimbursements from insurers, hiring staff, etc.
And, reserach shows that when speicalists and primary care docs are working together it is easier to co-orindate care.They are all looking at the same chart for a given patient.. Everyone knows what medications everyone else is prescribing. If one doctor makes a mistake, another doctor is likely to catch it. They are looking over each others shoulders. This is one explanation for the better outcomes that we see–not just at Mayo, but at Kaiser, particularly in N. California, Colorado and in the Northwest, at Geisinger, at Intermountain, etc.
Not all mutli-specialty centers produce better outcomes. Not all achieve the same level of co-ordination. But overall, studies show, outcomes are better and physician and patient satisfaction is higher, while costs are lower.
By the way, most of Mayo’s patients are not particularly affluent. Many are people who live in Minnesota, which has a large immigrant population from Southeast Asia and elsewhere. (Minnesota is the un-Arizona–it welcomes immigrants, even if they speak no English, many organizations help them find jobs, etc.)
Mayo also takes charity cases, Medicare patients and Medicaid patients.
A couple of years ago, I wrote a post about a doctor who had worked at Mayo for a number of years, then moved to NY and worked at Columbia-Presbyterian for a few years,(which serves many minority patients including a large community of Dominicans)and then went back to Mayo.
He said the patient populations were very similar. The major difference–co-ordination. It took three times as long to get anything done at Columbia Presbyterian (and he liked his colleagues there.)
If you look at the post I wrote about 10 communities in various parts of the nation that have significantly reduced the cost of care while maintaining or raising the quality of care, you find that co-and collaboration was key to their success. They also all moved away from fee-for-service payments to salaries, or some pooling of fees.
Yes, doctors in small private practices could be linked by EMRS– but who would pay for it? It’s not just a matter of purchasing the IT, staff and physiciains must learn how to use EMRS. During the first year, physicians report that their revenues and income drop signficantly because learning is time consuming. (Later, EMRS may well save time, but it’s a steep learning curve.)
Large, multi-specialty centers enjoy economies of scale that make EMRs much more affordable.
Whenever virtually anyone talks about the problems in our health care system you’ll hear the word “fragmented.” We’re trying to run healthcare as a cottage industry, and it just doesn’t work. The economic model is no longer viable except in rural areas where real estate and labor are much less expensive.
EMR purchase isn’t such a big problem as many places will buy them for you. I am in private practice and have an EMR hooked up to the hospitals that I have patient’s in. No problem.
In addition, it isn’t difficult to learn at all and my revenue has not decreased one bit. Furthermore coordinating care isn’t a problem for this private practice physicians. I think that is a bunch of bull. Maggie just wants all doctors on salary. She can’t stand it when doctors have a business.
Peter–
I really don’t care whether doctors run a business or work on salary.
I do care whether patients are getting better care. The research shows that fee-for-service does not encourage higher quality care.
Maggie,
Are we certain that the “cottage industry” is the problem? I don’t know that Mayo and Kaiser are providing quality care for less. Mayo is not accepting Medicare in Arizona and Kaiser premiums are in line with everybody else.
Like Keith, I am concerned with this push for consolidation. The original health insurers were all nice non-profits, but eventually evolved into something else. The same will happen if physicians are forced out of private practice and into salaried positions with corporations (they will not be non-profit for very long). I’m afraid the vision of happy doctors looking over each other’s shoulder taking good care of patients with no other care in the world is not very realistic.
The beauty of EMRs done right is that there is no need to work for the same organization, or be in the same building (or even country) in order to coordinate care, share charts and keep an eye on a colleague.
The main reason I am trying very hard to make EMRs work for small practices is precisely so that consolidation is avoided and our “cottage industry” remains alive and well.
I just don’t want to see Walmart style medicine around here.
Thanks for the dissection of the data Maggie. As much as I would like to think primary care access is key, it is not surprising that it’s more complicated than that. There are so many factors that go into the care someone receives that a simplistic explanation is never going to be sufficient. At my practice, patients don’t have much trouble being seen due to an open access scheduling system. They do, however, have trouble seeing the same doctor every time. They have trouble getting important lab results followed up because our practice lacks good systems to monitor things like that, and because their phone numbers and addresses change much more frequently than do those of people living in more stable situations. They don’t get timely specialty care (the wait for a specialist who accepts Medicaid can be over 6 months). When they need an urgent evaluation by a specialist I sometimes have to send my patients to the emergency room as I have no other option, which results in higher risk for the patient, and poor communication between me and the specialist.
These are just a few of hundreds of examples of ways access to primary care does not necessarily lead to good care. I am hopeful that we will start to identify what is necessary to provide good care; I imagine things like disease registries that send patients automatic reminders to come for flu shots, podiatry appointments for diabetics, etc, as well as more automated systems through EMR reminding us to follow up on someone’s cholesterol or abnormal liver tests may help. I also imagine that improved communication between specialist and primary doctor will be important, though I don’t know how to measure that.
I have a sinking feeling that many of the problems my patients have stem from poverty; they are unable to come to appointments because they can’t miss work; they run out of prescriptions and don’t refill them because they’re expensive; they can’t afford nutritious food or a gym membership; they have poor health literacy.
Margalit —
One quick correction:
Mayo is not “not accepting Medicare in Arizona.” Mayo accepts Medicare at most of its facilities in Arizona, including its main facilities there, but is experimenting with using a “concierge medicine” model at ONE of its smaller satellites in Arizona.
Pat,
I stand corrected. This is limited to only one clinic in Arizona, and only 5 primary care doctors. However, according to Mayo, this is not just an experiment. Mayo is losing money by treating Medicare patients, by its own admission
http://healthpolicyblog.mayoclinic.org/2010/01/05/medicare-and-mayo-clinic-in-arizona/
I find this a bit strange if indeed Mayo is so efficient. Other providers seem to do fine with Medicare payments, or at least not as bad as Mayo.
I also find it very interesting that the only services Mayo is not accepting assignments for are the notoriously underpaid office visits for PCPs. It must be nice to be able to pick and choose the profitable services, and dump the rest on the patient and/or non-Mayo PCPs in the region.
Margalit —
Actually, this IS an experiment. Here is the quote from the article you cite:
“Rather, a five-physician Mayo Clinic Arizona family practice clinic in Glendale, Ariz., has opted out of Medicare as part of a Mayo Clinic time-limited trial that will be reviewed at its conclusion.”
Like you, I am also somewhat dubious that Mayo is actually losing significant amounts of money on Medicare. When I worked at Mayo, many years ago, they always claimed they “lost money” on Medicare, but closer examination of the claim showed that they meant that they were paid less than by other third party payers — and Mayo does collect very large fees from private payers because of their ability to dictate prices to them — counting the difference in payment as “losses” even though they actually were in the black on the services. I am uncertain as to what the situation is now, but I do know (first hand) that many provider systems in Minnesota make a profit on the overwhelming majority of Medicare patients, although they do incur losses on some patients who fall outside Medicare hospital payment standards.
That said, as you have noted, office primary care is underpaid significantly by Medicare in many areas of the country.
My guess as to what is happening is that Mayo has a problem with whoever is the Medicare payment contractor in Arizona, and are not getting the deference they are used to in Minnesota. They are running a very high profile experiment, partly to see what happens — if they can make a go of refusing Medicare in a single affluent suburb — and also as a way of putting pressure on the Medicare service provider in Arizona with this insignificant but highly publicized ploy.
BTW, this all serves to answer the question by Jenga below as to why PCP’s should receive higher payments than they do now.
Margalit, thanks for your coomment
But I can’t agree on some points
– First -a small point: Mayo is Not losing money on Medicare. The chief executives at Mayo are reluctant to admit this in public, but the fact is, they do very well on Medicare.(see comment by Pat S.Sept. 20 . Other sources have confrmed wht he says. Undestandably, Mayo just doesn’t like to talk about this.. .
Mayo offers better care for less then the aveage medical center,not just in Rochester, but in all locations, though the flagship hospital in Minnesota is the most efficient. (see http://www.dartmouthatlas.org and search:”Mayo.”
As for Mayo in Arizona,see coment by Pat s.
Margalit, Jenga& Panacea, Sharon
Margalit– Yes, Kaiser’s premiums are in line with other insurers in Calif, but they provide better care. So patients are getting more for their dollars.
For example: “Heart attacks declined by 24 percent within a large, ethnically diverse, community-based population since 2000, and the relative incidence of serious heart attacks that do permanent damage declined by 62 percent, according to a . . . study in the current issue of the New England Journal of Medicine.”
This is remarkable, and just one example of how well Kaiser’s co-ordinated care works.
No one in the reform movement is talking about moving toward WalMart medicine. But the reform legislation does recognize that electronic medical records alone will not heal a fracturered delivery system.
It’s simply very expensive–and inefficient– to have solo practioners each paying for overhead and labor. The economies of scale that large multi-specialty practices enjoy reduce administrative costs. And the collegiality creates an environment where physians, nurses, physical theraptists and others can be constantly improving and learning from each other–as in “Hey, did you see that article in JAMA about . . .”
Jenga– First of all, see Panacea’s comment. Directing the orchestra can require more knowledge and skill than a single specialty.
Very good primary care doctors co-ordinate specialts’ care and also treat the patient sufferring from two or three chronic diseases. Their role is crucial.
But for a variety of reasons, many primary care docs are not able to do a good job– neither they nor the specialists the patient sees have EMRS, so communication is difficult. They wind up playing phone tag, waiting for faxes that never come . . Some primary care docs (both younger docs and those who take Medicaid) earn as little as $115,000 a year. Meanwhile, they are trying to pay off loans and keep a practice going. This means that they wind up spending 10 to 15 minutes with each patient. If the patient is in good health, and just needs antibiotics for strep throat, that may be fine. But if we’re talking about an older diabetic who also is suffering from congestive heart failure and depression, 15 minutes is not enough.
As Panacea suggests, he may need to be paid for 30 minutes–and a good share of that 30 minutes should be spent just talking to and listening to the patient.
We need to create a delivery system where primcary docs can do a good job. That’s what Panacea is talking about.
We know that at places like Kaiser or Geisinger or INtermountain, primary care is excellent–and there are small practices out there doing a good job. But it’s becoming harder and harder, especiallly in cities where overhead (real estate, etc.) plus the cost of labor is very high. Integrated systems offer much-needed economies of scale.
Sharon– You do a very good job of describing all of the obstacles that a primary care physician faces when trying to care for poorer patients.
Medicaid does not pay enough (though under reform they will begin paying primary care docs Medicare rates).
I think that, in many caes, well-run community clinics can provide the answer. Under reform, we’re doubling the capacity of these clinics. The funding is good, and we’re also offering full loan forgiveness and scholarsihps for docs and nurses willing to work in these clinics.
In New York City Dr. Neal Calman has made great strides in setting up clinics that are extremely efficient. They use EMRS, waiting times are 15 minutes or less, and patients have acccess to speicalists as well as primary care. I’m told that the systems management at these clinics is excellent.
Calman has been taking care of the poor for many years. I write about him in the book. And finally, he has gotten the funding to set up these clinics. They should serve as a model for what can be done. These are places where primary care docs have the support in terms of staff, EMRS, registries etc. to do exactly what you want to do.
Pat S.–
Thanks much for clarifying what is going on with Mayo in Arizona. I did look that situation at one point, and everything you say confirms what I learned.
Primary care is not compensated enough to do what is supposed to do. I just met a cardiac interventionalist who said it takes him about 40-50 min to put a stent in most cases , everything else is handled by nursing staff.
If a primary care gets to sees a person with 3-4 chronic illness and who complains of fatigue, it takes 45 min to check meds, get proper history and order only NEEDED test and this is not counting the follow up calls about results. In this scenario there is a 10 fold difference in reimbursement for the same time spent, nothing in the system currently encourages good primary care, those who do it are getting burnt because their time is under valued.
Ray–
I totally agree that the primary care doc needs 45 minutes in the situation you describe. And he should be paid for his time.
But median income for primary care docs is now around $180,000 (according to the AMA)–which means that half earn more than that.
The ones earning significantly less than $180,000 –docs earning $110,000 or $120,000 do need to be paid more. They are usually young docs struggling to pay off loans or docs who work with poor patients and take Medicaid.
But I’m not certain that the 50 percent of primary care docs earning more than roughly $180,000–, need to be paid significantly more–or anything close to what we pay the interventionist for 45 minutes of his time.
If the interventionist is earning 10 times as much per minute, maybe we’re over-paying the interventionist, while underpaying primary care docs on the bottom half of the ladder?
I’m not suggesting that we suddenly slash payments to the best paid specialists by 20 percent. But over time, we do need to redistribute those dollars, down the ladder, to docs who are not as well paid. At the top of the ladder, we’re looking at a bubble that we can’t afford.
I’m confused by two things Maggie has said.
1) this article seems to say that a greater number of PCP’s is not associated with higher quality. Are the authors saying that the finding by Chandra and Baicker in 4/7/2004 Health Affairs (exhibit 8) is wrong, In that study the authors reported that having more GP’s was associated with lower spending and higher quality. But this report seems to say the opposite.
2) I followed the link to dartmouth and searched for Mayo as Maggie suggested. But there is nothing about Mayo’s quality or their efficiency. All there are that i could find were reports for how many resources are used by Mayo for people who died. What does dying in a hospital have to do with the quality of care they give or how efficient they are? What if people are more likely to die when fewer resources are used to care for them? Isn’t medicine about preventnig death? Is dartmouth saying that we should gauge our system and pick the best providers based on who lets people die the most ‘efficiently’? What does that mean about what we aspire to in healthcare?
Epidoc,
Chandra and Baciker were writing about primary care in parts of the country where patients see more primary care docs, and fewer specialists.
In those parts of the country (upper middle-west, Northwest, Northern California, Maine, New Hamphsire and Vermont) ,primary care is generally much better than it is in places like NYC, Southern California, Florida etc., etc. etc.
This is for two reasons. First in the areas where people see more primary care docs, more of those physicians are working in integrated multi-specialty organizations where primary care doctors have much, much better suppport and are not under the financial pressure of trying to run a business.
Secondly, in some of these regions (upper New England, for instance, and Iowa), a primary care doc’s overhead is not nearly as high– real estate and the cost of labor are much lower. This means he doesn’t have to keep up a hectic pace, seeing patients every 15 minutes, just to keep the lights on.
This most recent study that I’m writing about was looking at primary care, in general, nationwide.
Overall, primary care nationwide isn’t doing the job most primary care docs would like to do–because they don’t have enough support (the electronic medical records and easy access to specialists that a Kaiser or a Mayo Clinic provides), and as very small businessses with high overhead, they are under financial pressure which forces them to see a patient every 15 minutes.
As to whether care at the Mayo clinic is better . . .
Have you ever talked to anyone who had been treated at Mayo?
I don’t think anyone questions whether treatment at Mayo is outstanding.
And if you look at the Dartmouth research, you will find that patient satisfaction and doctor satisfaction are much higher at Mayo. Outcomes also are better–even at Mayo “outposts” (outside Minnesota.)
And Mayo is not, by and large, treating “easier” or “more affluent” patients.
Mayo attracts the most difficult cases in the country, does a lot of charity care, takes Medicaid patients, etc.
Yes, some (not all) of Dartmouth’s studies compare patients who died. Everyone dies. But where you die is terribly important.
AT NYU medical center, chances are very high that you’ll die in the ICU–perhaps strapped to your bed, very likely with a feeding tube jammed down your throat–and limited hours for anyone to visit.
If you die at Mayo, you are far more likely to die in a hospice or at home with palliative care or hospice workers taking care of you, making sure you’re not in pain, still treating your disease (if you have chosen palliative care.) We now have much evidence that patients who have the opportunity to choose palliative or hospice care (as opposed to dying in a hospital ICU) live longer.
Maggie,
You’re kidding right? The Chandra/Baicker article analyzed data for all 50 states. But you are claiming a distinction between that and the new study which you say is different because it is nationwide? What is the difference between nationwide and ‘all 50 states’? Also, Chandra/Baicker exhibit 8 that i pointed to looked at quality against the number of pcp’s, not the ratio of pcp’s to specialists as you misstate. They did a separate analysis looking at specialists. As they state in their abstract, their major finding is: “States with more general practitioners use more effective care and have lower spending”. contrast that to your summary of the recent dartmouth report: “If the goal is to improve the population’s health, while simultaneously reining in the cost of care, health care reformers should focus less on increasing the supply of primary care physicians.” So, Chandra/Baicker say more pcp’s means higher quality and lower cost. The new report tells policy makers that if they aim to improve quality and lower costs they should not focus on increasing the number of PCP’s. Honestly, you can’t see that the two conclusions differ?
As for mayo, in your earlier post you claimed that mayo was a particularly efficient provider (you said ‘better care for less [money]) and said people could find those data on the atlas website. i said i couldn’t find it (and still can’t). I’d expect a link to the website to prove me wrong, but your proof that i’m wrong is to say “talk to someone who went to mayo, it is great’. Is it the case, or not, that there is somewhere on the Atlas website that shows data involving both the ‘better care’ and the ‘less money’ parts of your assertion for the same hospital with respect to other hospitals? Saying ‘you know it’s true’ is really not proof, even if i think the Mayo clinic is great.
Epidoc–
First, the Chandra Baicker study you refer to is from 2004.
Both their work, and Dartmouth’s work in general, has evolved quite a bit since then.
There are many ways of measuring quality. As Chandra and Baicker make clear, they are using One way: “Recent research has found large and persistent differences across states in the quality of care that Medicare beneficiaries receive. ONE WAY To Measure these differences is through differences in the use of effective, high-quality care, such as the administration of beta-blockers after heart attacks, mammograms for older women, influenza vaccines, or eye exams for diabetics. These procedures are relatively inexpensive, are known to have desirable medical benefits, and are rarely contraindicated. It is therefore puzzling that the use of these procedures varies so widely between states . . ”
In addition to use of beta-blockers, etc., they also looked at patient satisfactoin. These were hte markers to define “quality” in this particular study. (In hundreds of different studies, Darmouth resreachers have sliced into the question of “quality” from many different angles. The overall result: in areas where Medicare spends more, “quality” (defined in many different ways) is not better, and sometimes is worse.
In this particular study, Baciker and Chandra investigated “What causes some states to be high spenders and provide lower-quality care, while others are low spenders and provide higher-quality care? One Possibility is the composition of the medical workforce. Exhibits 6–11 examine this Hypothesis, illustrating the relationship between the medical workforce, spending, and quality. The exhibits adjust for the total number of physicians in a state and study the effect of specialists (Exhibits 6 and 7), general practitioners (Exhibits 8 and 9), and nurses (Exhibits 10 and 11) per capita on overall quality rank and Medicare spending. We are thus examining the effect of changing the composition of the medical workforce, , , . Together, these workforce measures can explain 42 PERCENT of state-level variation in Medicare spending per beneficiary.”
Note: First, they make it clear that there are examining the possiblity that more pimary care docs lead to better quality, as definited in this paper. It’s a hypothesis. And secondly, they conclude that the composition of workforce accounts for just 42 PERCENT of variation in spending.
This suggests it’s a major factor, but not the only factor that determines spending. .
They also note that:
“It is possible that although areas with more specialists do not provide higher-quality care along these dimensions,[giving beta-blockers, eye exams for diabetics, etc.] they may be better at the treatment of more acute conditions.”
Finally, Chandra and Baicker conclude: “inferences about causal mechanisms should be made with great caution.” Corrrelation is one thing, causality is another.
In 2010 this newest report goes at step further to show that: “neither delivering a greater amount of primary care, nor making sure patients routinely see a primary care clinician IS, BY ITSELF A GUARANTEE itself, a guarantee that patients will get recommended care or experience better outcomes.”
In other words, while better access to primary care may improve care–at least along some dimensions (getting hte beta-blockers etc.)– good access to primary care is not, in and of itself, a guarantee of better quality and better outcomes.
This study begins by acknowledging that what Chandra & Baicker said in 2004 is still true: “Access to high-quality medical care, and primary care in particular, is known to be
a key factor in preventive care and chronic disease management, such as reducing
complications from diabetes.” (the eye exams)
But, this new report goes a step further to say that while “Improving the care delivered by primary care clinicians holds great promise for better
patient health and well-being, but the value of primary care can be eroded by episodic
delivery that is uncoordinated with specialists and hospitals. Thus, simply increasing
access to primary care, either by boosting the number of primary care physicians in
an area or by ensuring that most patients have better insurance coverage, may not be
enough to improve the quality of care or health outcomes; nor is it likely to eliminate
racial disparities. Primary care is most effective when it is embedded in a high-functioning
system, where care is coordinated, where physicians communicate with one
another about their patients, and where feedback is available about performance that
allows physicians and local hospitals to continually improve . . ”
[This is what I was referring to when I noted that primary care leads to better overall outcomes only if it is good primary care that is co-ordinated with good specialty care. Primary care docs can’t do it all by themsleves. In states where patients see more primary care docs (in the Upper Midwest for instance), care tends to be more organized and better co-ordinated. So you see better results–both because patients are getting mroe primary care and because it is embedded in a system of care. In NYC, by contrast, outcomes are no better (even though spending is much higher) and often they are worse because care here is quite fragmented and disorganized.)
Note that this 2010 report defines “quality” in different ways. While the first report focused on whether or not patients received certain recommended tests (mammograms for lder women, eye tests for diabetics, etc.)
this later report also focuses on OUTCOMES. (As we learn more about how to measure quality, resrachers are focusing more and more on outcomes–and they’re getting better at defining good and bad outcomes. It’s not just “did the patient live or die?”
For instance, in this newest report, they look at how many patients underwent amputation of a leg –“a devastating devastating complication of
peripheral vascular and diabetes. Inadequate blood supply and
nerve damage predispose patients to injury and to infection that can fail
to heal.” Patients who have one amuptation tend to have a second amputation . . .
This 2010 report also measures quality by looking at how many patients were hospitalized because they wewre suffering from such as diabetes,
pneumonia and congestive heart failure. Ideally, doctors would be able to manage these disesase without hospitalizing hte patient. The authors write: “Spending time in a hospital poses
risks to patients, including infection and error, and a substantial cost to their families
and to society.”
When look at all of these measures of quality adn outcomes, the report concludes that access to Primary Care is not, BY ITSELF, a a guarantee of higher quality or better outcomes.
It appears that you only read the first 3 or 4 sentences of the abstract of the Baicker Chandra report.
The next time, you might want to read the entire study before saying things like “You must be kidding, right?” I realize you’re new to HealthBeat, but we try to avoid statements like that.
As for the Mayo Clinic, you’ll find the info here.
I really don’t have time to go through it with you line by line . . .
But when you got to “www.DartmouthAtlas.Org”you’ll see a button lableed “INformatoin by Hospital” at the top of the page.
Click on it. You’ll have an opportunity to search name of hospital. Type in “St. Mary’s” (that’s the Mayo clinic in Rochester, Minnn) You’ll get 40-some St. Mary’s hopsitals– scroll down till you get the one in Rochester, Minn.
Click on it. At the top of the page, you’ll see “patient expeirence”–click on it.
This will give you info on how many days patients getting their care at ST. Mary’s spent in the hospital during the final six months of life, how many died in an ICU, how many saw more than 10 specialists during hte last six months, when compared to the national average.
One reason patient satisfaction is so much higher at St. Mary’s is because most people don’t want to die in a hospital or an ICU. They would prefer to die in a hospice, or at home. (90% report they want to die at home.) They also don’t really want 15 or 20 different specialists poking and prodding them during the final six months. (More specialists tends to mean “too many cooks”– all consulting on one patient but not consulting with each other. If a patient is seeing 15 or 20 specialists it’s also a sign that the hospital hasn’t entirely figured out what his main problem is. Eight specialists might make sense. 17? Not really.
Most patients also don’t want to spend most of their final six months bouncing in and out of the hospital–( or just in the hospital.) These days, most of us die of chronic diseases that can be treated in the home or in hospice.
Here, the key to quality is “managing” the disease–avoiding crisis. The outcomes for all of the patients in this study were the same: they were within six months of dying (i.e. they were all very sick) so the question about the outcome is “was it a good death?’
Not only is patient satisifaction higher at Mayo, physician satisfactoin is higher– physiciains report little trouble communicating and co-ordinating with other docs,etc.
Again, this is only one way of measuring quality.
Dartmouth also has done studies on patients who did’t die but suffered one of 5 diseases (hip fracture was one of them), finding better outcomes at
lower-spendindg medical centers like Mayo where care is better co-ordinated.
Finally, you’ll find a description of Mayo’s systems, and medical culture, and what makes it better (by a doctor who has worked there and elsewhere) here
http://takingnote.tcf.org/2008/10/what-makes-the.html
oxxon–
There is far more than an iota of evidence.
For starters, Minute clinics provide no continuity of care –which means no chronic disease managment.
And cdm is our biggest problem.