More residency slots: Good, but not going to solve the primary care shortage...

The Center for Medicare and Medicaid Services, CMS, recently announced that it will be funding an additional 440 residency positions in 135 hospitals in 37 states. This is a good thing, and a step in the direction of reducing the glaring shortage of physicians in the US. CMS already funds most of the residency positions, something which is probably not intuitively obvious to those not involved in post-graduate medical education, but it is so even if it is not clear why (as opposed to, say, Congress directly appropriating such funding). This is important because while you are a doctor (physician) on graduation from medical school, you can’t obtain a license to practice anywhere without residency training (e.g., family medicine, pediatrics, surgery, psychiatry, radiology, etc.). Thus, past efforts to increase the number of physicians by increasing the number of medical schools (or the class size in existing medical schools) fail, because the residency pipeline has been static. It has only, perhaps, decreased the number of international medical graduates filling those positions. 

CMS further states, per this piece in MedPage Today, that 2/3 of those slots will be in primary care and psychiatry, two areas in which the shortage of physicians and other clinicians is particularly acute. This is also good, but will continue to have limited impact, at least until we see exactly which specialties are getting those positions. “Primary care” is not a specialty in which there is a residency; usually the term is used to include family physicians, general pediatricians, and general internists (as well as those general practitioners who completed only one year of post-graduate training, ie., internship, when that was sufficient to receive a license). The problem is mostly with internal medicine. After completion of a 3-year internal medicine residency, the large majority (about 80%) of graduates go on to complete a fellowship in an internal medicine subspecialty (cardiology, nephrology, pulmonary medicine, endocrinology, etc.). They do not practice primary care. In addition, about half the remainder become hospitalists, caring for people in the hospital only, leaving only a small number of internal medicine residency graduates to practice primary care. While many pediatrics graduates also sub-specialize or become hospitalists, the percents are much lower. On the other hand, about 90% of family medicine graduates practice outpatient primary care. In addition, it likely includes OB/Gyn which is not primary care.*

It would be good if we could get an accurate accounting. There are “general medicine” or “primary care medicine” residency programs among the internal medicine programs which produce higher percentages of primary care outpatient practitioners -- but still not all, or even close to the 90% of family medicine. We need to know if the new CMS positions that are designated “primary care” include internal medicine; if they do, obviously the output, the number of new outpatient primary care doctors, will be significantly lower. Why should they be unclear about this? It is in the interests of many institutions to keep it fuzzy. Medical schools have long reported on the percent of their graduates entering primary care, and they have included all those entering internal medicine (and sometimes OB/Gyn or even Emergency Medicine) as well as pediatrics and family medicine, inflating the percent that is ostensibly primary care. This practice has been called “the Deans’ lie”. The Association of American Medical Colleges (AAMC), quoted in the MedPage Today piece, has a similar reason to obfuscate the truth; their members are those same medical schools (and the teaching hospitals associated with them) and their leaders, both in AAMC and in the individual schools, are overwhelmingly non-primary-care subspecialists.

One of the main reasons for the shortage of primary care physicians, and thus this purposely-inaccurate effort to paper over the dearth of them, is that, while physicians overall make much more than the average person, there is a great disparity in income between the specialties, often being 4-5 times as much for certain specialists as others. The shortage of primary care physicians and psychiatrists that this change is making a small effort to rectify is very likely because pediatricians, family physicians and psychiatrists are on the low end of the physician income scale. The table below, from “PhysiciansThrive.com”, is one example, although it may actually understate the income of the top specialties.

 

 

The lowest-paid physicians are doing well compared to most Americans, but they often come out of school with $250K + debt, which is, of course, owed with compound interest. The income gap between specialties has to be narrowed, and some studies suggest that if primary care physicians made 70% of what other specialists do, money would largely cease to be an issue in student specialty choice. But how could we do that? Isn’t it really complicated?

Well, not as much as you might think. A recent installment of the New York Times feature “The Ethicist” responds to the question from a reader “Should I Feel Bad About Joining a Concierge Medical Practice?”. This in itself is a complex, but separate, question. However, the Ethicist notes in their response that one of the reasons it is hard to find a primary care physician is that 

‘Medicare, which effectively anchors compensation levels throughout the health-care system, reimburses physicians according to “relative value units,” and those are largely determined by an advisory committee dominated by specialists. Procedures are valued more than conversations.’

I was surprised to see this addressed in this column; while it is absolutely true, it is rarely discussed. The comment makes 3 points which together drive the income disparity:

1.     CMS sets reimbursement rates for Medicare, but it essentially drives all reimbursement as insurance companies use Medicare rates (or multiples of them) to determine their own payments.

2.   These Medicare rates are set by “relative value units”, or RVUs, so that some sort of equivalence can be made. For example, how many comprehensive examinations by a primary care doctor or assessments by a psychiatrist or chest x-ray interpretations by a radiologist is worth one gall bladder removal by a surgeon or joint replacement by an orthopedist? This is essentially dividing up a pie of set size; when one “gains” another “loses”.

3.  While CMS sets these RVU ratios, they usually rely upon the recommendations of a little-known group called the “RUC”, appointed by the AMA. As the Ethicist notes, this committee is overwhelmingly specialists. And, so, not surprisingly, the distribution of the pieces of the pie tend to favor those specialists, and, indeed, procedures are valued more than conversations. And “conversations” kind of minimizes the communication between people (patients) and their primary care doctors, as people’s stories greatly inform both the diagnosis and the best treatment (and the one the patient is going to accept and follow through on), and are essential for the patients to know what is going on. And presumably interactions with psychiatrists are more than simple “conversations”. (I have written about the RUC several times, including Doctors' incomes and patient coverage: both need to be more equal, Jan 26, 2014 and Pay primary care more: Kennedy may be getting this one right!, July 23, 2025.) 

So there is a clear-cut way to fix these disparities: CMS simply needs to adjust the RVU basis, increasing the value of “conversations” and really thinking and assessment and decision making relative to procedures. Of course, this means the highest-paid subpspecialists would make less than they do now, and so they are going to fight it, as are their organizations and groups like the AAMC. But really, even if they take a significant cut, they are a long way from the Food Bank!

The question is: Will CMS do this? Pressure from you, the people who can’t get into a doctor, is definitely going to help!

 

*Primary care is comprehensive care of the needs of the patient, with referral to specialists when needed; internists and geriatricians and pediatricians limit the ages of their patients but they provide comprehensive care for them. While some women only see an OB/Gyn, they only provide care for a woman’s reproductive tract, not comprehensive care.

 

Who is gaming the system? Surprise, it's the corporations!

I recently participated in a panel discussion following a presentation on the impact of the Affordable Care Act (“Obamacare”) by UC-Berkeley economist J. Bradford DeLong. Unsurprisingly, Dr. DeLong, who worked in the federal government as Deputy Assistant Secretary of the Treasury for Economic Policy in the early years of the Clinton administration, during an earlier attempt to pass comprehensive health reform, took an economic point of view. He described the economic theories behind each of the approaches to health reform, how the ACA was put together, how it most resembled the “RomneyCare” model implemented in Massachusetts and endorsed by Hillary Clinton but abandoned by the Republicans; he also showed that the Obama administration miscalculated the near-unanimity of Republican opposition. He also looked at how the implementation of the ACA has been more successful than many feared (or hoped) and how the economic analysis behind it was distorted by the Supreme Court decision to allow states to not expand Medicaid, which has resulted in an enormous transfer of wealth from the “red” states that have not to the “blue” states that have done so. Apparently, the ideological commitment by many states (including my own, Kansas, and neighboring Missouri) to harm its people and give away money is puzzling from a purely economic, as well as a human, point of view.

One of the themes Dr. DeLong notes coming especially from “conservative” economists and Republican politicians (and we hear a lot) is the need for people to have “skin in the game”, by which they mean co-pays, deductibles, and other ways of people paying out of their pockets. As Dr. DeLong noted, the only large, well-designed, and meticulous study of the impact of such “skin in the game”, the 1983 RAND experiment (which I have previously discussed; see Insurance company profits up and patient care down, May 17, 2011*)  showed that even small out-of-pocket payments discourage people from seeking care for both minor and major conditions, ultimately cost much more to care for, and harm the health of those people. As noted by one of the audience, current requirements in many “high-deductible” plans for “skin in the game” cost-sharing are far greater than those studied by RAND (and can be 45% of a person’s income!) and are thus even more likely to have a major negative health impact.

Another common “game” meme, mentioned by one of the other panelists, is concern with people “gaming the system”. If this conjures up images of elegant gamblers in formal wear playing roulette with James Bond in a posh casino on the Riviera, that is the intent. Like Ronald Reagan’s “welfare queens” and Kansas Governor Sam Brownback’s “able-bodied adults who refuse to work”, it is meant to divide people by creating a “them” who are taking it easy while the hard-working “us” pay the price. Of course, this is nonsense; most of those individuals so “gaming the system” are merely trying their best not to break their budgets paying for health insurance until they get so sick that they need it. Yes, this is certainly contrary to the concept of insurance (everyone pays and only some people benefit, but you never know when it will be you), and is a big reason that most countries have gone to a “social insurance” system that just covers everyone.

In fact, despite all the fooforaw about it, there is no data suggesting that there is massive “gaming of the system” by regular people. Michael Hiltzik’s Los Angeles Times column of February 27, 2016 makes this clear, focusing on “Special Enrollment Periods” (SEPs), times when people can enroll in or change their insurance outside of the usual ACA annual period. Huge insurance companies like Aetna and Anthem have asserted, without much evidence, that people are using these SEPs (mostly designed to allow changes when you get married, divorced, have a baby, move to a different state where your current plan wouldn’t cover you) to “buy to use”, in Anthem’s words, meaning you wait until you’re sick to get insurance. Hiltzik presents data that shows this is not significantly the case, and that it is absurd; he writes “Aetna must think the entire country consists of people plotting how to get a quickie marriage or divorce or have a baby just in case they get sick. The vast majority of SEPs cover a relatively trivial number of cases, unless you think there are hordes of people applying to become members of a Native American tribe after they get sick.”

Of course, people do game the system. But the big gaming is by the insurance companies themselves and the providers of care. These corporations, big insurance companies, health systems, drug makers, who have the clout to “game the system” do so all the time as part of their core business models. It is convenient for them to divert our attention to regular people, middle-class people, and especially poor people, as the ones gaming the system. In fact these are of course the folks who suffer the most harm, whose health is most affected, whose access to care is most limited, and who are stuck with crummy health coverage because this is all they can afford.

The insurers work every legal angle (and perhaps some not-so-legal) to figure out how to mostly insure relatively healthy, low-cost people (after all, 5% of people account for 50% of health costs and 1% for 20%, see my Red, Blue, and Purple: The Math of Health Care Spending, October 20, 2009, and Kaiser Health News report 2013),  while the high-cost patients are covered by Someone Else. Providers, especially health systems and hospitals, figure out how they can “upcode” to get maximum reimbursement from insurers, attract people who have high-profit-margin conditions to themselves, and encourage high-cost, low-reimbursement poor and poorly-insured people to find their care from Someone Else. Insurers blame providers for charging too much, providers blame insurers for paying too little. One of the other panelists, a hospital executive, complained about how insurers seek transfer risk, which is part of the definition of insurance, to the providers. Neither is blameless, and the other big players, pharmaceutical companies (and device manufacturers) make out like bandits, with no major candidate having a real plan to address this according to a report by Julie Rovner of Kaiser Health, (cited here by Medpage Today). Of course, this equates all plans to “negotiate prices” and it is obvious that a single-payer health plan, such as that advocated by Bernie Sanders, will have a lot more negotiating clout than the multiple-insurer mess that other candidates support and exists today.

What did I say as a panelist? Basically, that the goal of the system should be to maintain and improve the health of people, and that the economic design of the system should be designed to achieve that goal, rather than having competing economic theory be the driver, and people the incidental victims. I said that spending money on providing health care to people was not a bad thing, but spending huge amounts on “health care” when more than half was going to profit was. I said that all this spending on medical care (and profit) limited what was left to be spent on creating the conditions that allowed people to benefit from medical care – like housing, food, education – the social determinants of health.

I think that this resonates with people, both at the event and in the world. Or maybe I’m one of those “hopeless idealists”. If the alternative is being cynically corrupt, I wouldn’t want to be anything else.

-----------------------------------------------------------

* Citations from that blog post: “RAND Health Insurance Experiment (as cited in Freedom abroad, health at home: experiments in preventive health care, February 13, 2011; the study was published in the New England Journal of Medicine in 1983[1]; and it is summarized in an article by Joseph P. Newhouse, "Free for all?:  lessons from the RAND Health Insurance Experiment", RAND 1993.”