In the past couple of days the health policy blogosphere is all a twitter in response to what appears to some to be a dishonest hit job on the front page of the NY Times against the so-called Dartmouth Atlas analysis.
Those who follow the health policy debate will recall that there are huge variations in the cost of hospital services in different hospitals and in different localities. And that some places "do more" and charge more for it, even where the amount of the underlying disease is the same and/or there are more doctors treating it.
Indeed, one of most basic big issues in health care economics is that simplistic "Economics 101" supply-demand does hold true. In what we might call "Health Economics 302", it turns out that if there are more orthopedic surgeons or invasive cardiologists in an area, neither unit cost nor total costs go down. Instead, the number of procedures done on a population, and total costs, goes up. Even cost per procedure or cost per doctor stay the same or go up. The basics of why health economics does not follow the simplistic supply-n-demand has been established since Arrow's foundational health economics article (.pdf) in 1963.
The folks who have pioneered the recent research into geographic variation in the amount, intensity and cost of care are based at Dartmouth, and it is important and mostly valid research. It has formed the basis for the widely promoted New Yorker article by Gawande and before that the somewhat tendentious Overtreated book by Brownlee.
Much of the medical establishment, especially the hospitals and the hospital-based medical and surgical specialties whose pockets will be gored, have criticized the Dartmouth Atlas, and are no doubt part of the organized attacks on it. As always, it is about power and money.
However, the Dartmouth academics are capable of explaining (.pdf) and defending (.pdf) themselves. And they have many advocates in the health policy blogosphere.
The short answer is that is it true that, even after accounting for actual higher need (e.g., more sick local population and more sick individuals) and quality of care, there is still excess cost in some areas. And this cost is partly due to unneeded care, inapropriate care, wrong care, poor care, excess care, local "professional cultural" practices not grounded in evidenced based care, unwarranted care (excess volume), unwarranted higher costs (excess unit-cost), etc. The bottom line is that there really is unwarranted and substantial variation in the cost of care for people of similar health depending where they happen to live and which institutions they go to. It is also the case that clinicians and institutions with the best results can have lower (e.g., Mayo Clinic), not higher, costs than average.
It should also be noted that there are legitimate critiques about the degree of controlling for quality, the degree of controlling for case-mix (amount and severity of illness), whether local cost of living ought to be factored in, the legitimate role of the higher cost of teaching (medical students, residents, fellows) and research institutions. But in their essentials, the Dartmouth research is valid and important.
However, there is also a progressive critique of the "over-treated" argument, partly due to the actual Dartmouth research focus, but especially due to some of the uses others (not the Dartmouth researchers themselves, for the most part) have put it to.
My main criticism of Dartmouth work and especially some of the use it has been put to is its mostly ignoring (and then being used to deny by others) the real problem of the UNDER-served.
Dartmouth research is mostly valid, but only for what they actually are studying. Their focus is on inpatient hospital care, and those who have already managed to get their way into such care. The data is primarily Medicare Part A, because that is where the data exists to be able to make systematic nationwide comparisons. But therefore the results are by definition limited to that (by definition) well insured population who have successfully achieved access to care.
It completely misses (but is still used by some to deny the existance or importance of) the larger problem (in my opinion) in the American health care system which is the lack of access to primary care; the widespread existence of medically underserved areas and populations and health professional shortage areas; the shortage of primary care clinicians.
It completely misses the problem of disparities in health status and disparities in health care.
I am not saying that many of the Dartmouth researchers and many of their liberal proponents are not aware of and in principle concerned about these issues. But it is not addressed by this research, and too littel attention seems to given to these issues in the health care debate.
The big problem, as a matter of policy and politics has been that, in the hands of many, the Dartmouth analysis gets over-generalized. Doctors and hospitals can generate excess and unwarranted demand. But this is primarily a specialist, procedure and hospital phenomenon. The discussion around the Dartmouth Atlas work mostly ignores primary care and underserved areas. Their classic scenario is take high specialist and well insured areas (most of the research uses Medicare data, so by definition have insurance) and show excess care in some hospitals and some areas. There is not a lot of analysis of outpatient diabetes management in the South Bronx or Appalachia.
As popularized in the Brownlee's "Overtreated" book this then gets generalized into the entirety of the U.S. health care system and from their turned into Republican (and alas some Democrats) talking points and health policy that the MAIN problem is excess care. It is part of the problem, but not the main or only one. It then becomes a convenient way to cut costs (deny care and access to care as unnecessary) while increasing the profitability of the insurance companies, and not addressing the reality of health disparities and the under-served. As always, it is about power and money.
Meanwhile, over-generalizing the importance of the over-treated problem ("oh those Americans, they are just all getting too much care too easily"), becomes a way for some to ignore or dismiss the real problems of too few doctors, clinics and hospitals in many areas; too little care, to hard to get, with too many hoops in the way; too many uninsured and too many UNDERinsured; too high out of pocket costs keeping people from getting primary care; too much paperwork for patients and doctors, too high out of pocket costs resulting in medical debt and bankruptcy if you do get sick; too much administrative costs and overhead and bureacracy by the truly unecessary insurance companies whose goal it is to not pay for health care. As always, it is about power and money.
Unfortunately it is also true that some of my fellow single payer advocates are too dismissive of both the Dartmouth Atlas work and the related research showing that so-called "Accountable Care Organizations" can (if they are not-for-profit and especially if they are physician managed) provide high quality comprhensice care at lower costs. Speaking for myself, not as official PNHP by any means, both the Dartmouth findings, an ACOs and Single Payer not only not contradictory or mutually exclusive. Just the opposite, they work together complimenting each other as the only way we are ultimately going to get health caer that is comprehensive, high quaioality and with cost control for both individuals and the system as a whole.
As Arnold Relman the distinguished former editor of the New England Journal of Medicin, among others, has pointed out, the other component of Single Payer, in addition the use of single universal "All America" insurance pool paid for by the same progressive taxation that we pay for other government services and the savings from both getting rid of the private insurance companies as a wasteful unecessary middleman and the benefit of monospony in paying for drugs and services, has always included the utility of global budgeting and strategic planning. It would include better control of number of medical student and residents; better control over the number going into different specialties (more into primary care; less plastic surgeons); the wider use of nurse practitioners and physician assistants and other health care workers; higher reimbursement rates for primary care in general and less for some specialty and procedure-based care and hospital care; higher reimbursement rates for practicing in inner city ghetto and rural and other "undesirable" places with high unmet need (and lower reimbursement for practicing in Bel Air and Park Avenue); enhanced reimbursement for quality of care, comprehensive care, patient centered medical homes and accountable care.
It is not either/or.
We are going to need all of these.
Many things are true at once.
Update (h/t SC Kitty):
Excellent detailed essays the past two days from Maggie Mahar on why the NY Times was a deliberate it job, misquoting sources, leaving out analyses actually done, and distorting analyses cited... and why the Dartmouth work is important and (mostly) valid (within its domain).
