Regional Variation Revisited: Price Differences Not A Significant Factor


Dartmouth scholars have revisited their analysis of regional variation in health  care spending and found contrary to the assertions of some critics that cost-of-living differentials do not account for much of the difference. However, they confirmed that some big cities with high poverty concentrations that also serve as training grounds for future physicians may have been unfairly lumped in with areas that overuse health care services.

The new study in Health Affairs showed after adjusting for price differences that Miami, Florida and McAllen, Texas still led the pack in terms of how much Medicare spent on each beneficiary. Both areas still spent nearly three times more than the lowest spending regions of the country, which remained Honolulu, Hawaii and LaCrosse, Wisconsin.

There were a few areas of the country where the adjustments made a big difference, and they were mostly big cities. The Bronx and Manhattan in New York City fell 39 and 33 percent, respectively, from the adjustments. But price was only a minor factor, according to the researchers, who were led by Daniel Gottlieb of the Dartmouth Institute for Health Policy and Clinical Practice.

Much of the reason why the New York metropolitan area is so costly is not because of the wage index per se (what we usually think of as “cost-of-living” adjustments), but because the CMS pays hospitals in the New York area so much to reimburse them for graduate medical education and caring for disproportionate shares of low-income patients.

Other high-spending areas frequently targeted by critics did not do so well under the adjustments. Los Angeles, for instance, dropped just 14 percent after adjusting for cost-of-living, graduate education and disproportionate share payments for low-income residents.

The Medicare Payment Advisory Commission issued a report late last year that suggested regional variation in use patterns were less than the Dartmouth Atlas of Health scholars had previously estimated. This latest study says regional variation still matter — a lot. The debate clearly isn’t over.

Here’s the list of the ten highest and ten lowest spending areas in the country both before and after adjustments for price, graduate medical education and disproportionate share payments:

10 high-spending hospital regions:

BEFORE        AFTER    % CHG.

FL-Miami               $15,909     $14,966      6%
TX-McAllen              13,633       13,881      -2
NY-Bronx                12,004         8,653      39
NY-Manhattan          11,744         8,861      33
TX-Harlingen            11,489       11,324      1
CA-Los Angeles        10,674         9,325      14
NY-Long Island        10,608         8,740      21
MI-Dearborn            10,460         9,791        7
LA-Monroe               10,226       11,385     -10
MI-Detroit                 9,954         9,541       4

10 low-spending hospital regions:
ND-Minot                   6,033         6,711     -10
VA-Lynchburg             6,022         6,524      -8
CO-Grand Junction      5,983         6,075      -2
OR-Eugene                5,968         5,798       3
IA-Iowa City               5,902         6,254      -6
SD-Rapid City             5,854         6,176      -5
OR-Salem                  5,810         5,642       3
IA-Dubuque               5,799         6,219      -7
WI-La Crosse             5,715         5,757      -1
HI-Honolulu               5,293         5,212       2

5 hospital regions with biggest drop due to price and other factors:

NY-Bronx                   12,004         8,653     39
NY-Manhattan            11,744         8,861     33
CA-Alameda County     9,251         7,094     30
CA-San Francisco         8,140          6,278    30
CA-San Mateo County  7,878          6,104     29

Categories: Uncategorized

Tagged as:

20 replies »

  1. “Um, hm, and by warding off diabetes and obesity-related problems, we preserve people for cancer, dementia and whatnot, at the same time extending the drain on social security resources.”
    Ok, just don’t treat the diseases that are man made and people will die sooner, far more efficient. Of course you’d have to stop the health industry from inventing ways to keep those people alive and mining the disease for many profit producing years.
    “Do you really believe Conagra controls my diet?”
    Maybe not much, I don’t know what you eat but they don’t control mine much. However they and other large corporate food producers/manufacturers profit from Ag subsidies that make unhealthy food cheap and that’s partly why we have $1 Big Mac value meals. HFCS, fat, salt all go into most processed foods. It’s hard to find foods without some subsidized corn based product in it. A healthy diet is out of reach cost wise for many, many people – you know, the ones that can’t afford health insurance but get those lifestyle diseases. I propose subsidies be switched to fresh fruits and vegetables while other foods get a calorie tax to help pay for future/present health costs. Of course you may oppose any tax that forces personal responsibility and people to pay up front for future health problems.
    “I’m not terribly familiar with the “death panel” controversy, but I don’t think the reformistas made it sufficiently clear that, in no event, would people be restrained from using their own resources any way they please, wrong-headed though they may seem.”
    I read that part of the bill and found no such stipulation, in fact the bill just “encouraged” people to write a living will and keep it up to date so that docs, relatives, government, could not override their decision, whatever that may be. How would the government stop people from spending their own money anyway? That would not be an issue for many people anyway given the cost of hospital care.

  2. It’s important to not generalize the Dartmouth Atlas assessment of costs to the overall cost drivers in a region. The Atlas is focused on Medicare costs which are based on utilization and Medicare-determined rates.
    Commercial prices are based on negotiated rates which, as has recently been reported in the Boston Globe http://www.boston.com/news/local/massachusetts/articles/2010/01/29/attorney_general_says_clout_drives_up_health_costs/?page=1, vary substantially base on market clout — aka leverage. It is the combination of the utilization patterns reflected in the Dartmouth Atlas and the pricing patterns in markets with high concentrations of care by leveraged providers that will drive a region’s overall medical costs.

  3. Um, hm, and by warding off diabetes and obesity-related problems, we preserve people for cancer, dementia and whatnot, at the same time extending the drain on social security resources.
    Do you really believe Conagra controls my diet?
    I’m not terribly familiar with the “death panel” controversy, but I don’t think the reformistas made it sufficiently clear that, in no event, would people be restrained from using their own resources any way they please, wrong-headed though they may seem. And I’m talking about requiring people to contemplate end-of-life issues at a time when they are still healthy, not when they’re on their deathbed. With Medicare and Medicaid, you have an existing repository for their decisions. You could, of course, extend the requirement to insurance.

  4. “I don’t see why, upon first qualifying for Medicare or Medicaid, people are not required to study on end-of-life issues.”
    Just try to get that past Republicans/Evangelicals who won’t scream “death panels”. The dead bill tried to address this issue. And why would you limit this to Medicare/Medicaid?
    “I don’t see how “wellness” and “lifestyle” regimens can have much impact here, since everyone will eventually croak anyway.”
    It’s not the croaking that’s killing us, it’s the cronic disease from our “lifestyle” choices. No plan, public or private, can contain costs when there’s an exposion in obesity and diabetes and all the medical problems they cause. But apparently this country has decided that the sales/profits of corporate food production/distribution take precedence over healthcare and it’s costs.

  5. Yes, Peter, insurers even double check their arithmetic, and conferences between treating physicians and the insurer’s medical experts, on catastrophic cases, are not uncommon. They just don’t have the authority, though, to control what Medicare can’t.
    I don’t see why, upon first qualifying for Medicare or Medicaid, people are not required to study on end-of-life issues. Most people have a horror of ending up as a human rutabaga.
    I don’t see how “wellness” and “lifestyle” regimens can have much impact here, since everyone will eventually croak anyway.
    The more I think about, most NYC doctors would probably think they had found paradise in McAllen.

  6. “I don’t see how you pierce the defense of “medical judgment.”
    Get rid of fee-for-service. Impose hosptial budgets where peer pressure and oversight prevents one or two docs from using their over zealous billing practices to steal income from other docs. Peer review and enforcement of best practice and comparative effectiveness. Bonuses for comimg in under budget. Is there ever a medical meeting between hosptial management and medical staff about budgets, except bill til you drop? Does insurance do any oversight for kind/amount/frequency of billing other than outright fraud to the insurance company?

  7. Not to utter blasphemy in the inner sanctum, but what is on the table to control those providers who, with monetary gain aforethought, engage in the practice of “overutilization”? I don’t see how you pierce the defense of “medical judgment.”
    Actually, McAllen is hardly “forsaken.” It’s the destination of choice for “Winter Texans.” Now,El Paso is “forsaken” squared.

  8. My Healthcare Economics professor argued this for the entire semester, and made very compelling points to support his claims. It is possible that we could take advantage of these price differentials, but I think any savings would fall way short of “reform”.

  9. Margait & Tim
    Margait –The Dartmouth reserachers will tell you that there is more than one root cause. An excess supply of specialists and beds leads to over-treatment. (Build the beds and someone will fill them.) When doctors own the hospitals, more patients are hospitalized. A “money culture” can develop in a town like McCallen– read the whole story in the June 1 New Yorker. It’s a matter of what becomes the norm in a particular community– if everone is doing it, and it’s lucrative, others begin doing it. (You’ll find that the McAllen docs admit this in the story.)
    If you read more of the Dartmouth reserach (www.dartmouthatlas.org) you’ll find that WEnnberg and others have written extensively about “why” it is happening as they talk about “supply-driven care” as well as other reasons for overtreatment, including treatment that is driven by teh physician’s preference (preference-sensitive care).
    For instance, in some medical communitites, you’ll find that women with breast cancer are much more likely to have a full mastectomy, while in other communities they are more likely to have lumpectomies. Reserach shows that the cancer is not more serious in places where more mastectormies are done–and women in these places do not care less about saving their breasts. It’s just that it has become the custom among physicians in that areas to do mastectomies. That’s what “most surgeons” do most of the time. That’s the norm.
    What is interesting is that however the reserachers slice and dice the numbers–looking at individual hospitals, looking at larger regoins, comparing people who suffer form particular diseases, onlyl looking at academic medical centers in different parts of the country. adjusting for the patients’ race, income, underlying health of the population, etc. the results are always the same.
    Wide variations in intensity of care that cannot be explained by medical need or patients’ preferences.
    The DArtmouth reserach has been controversial since it began in the early 1970s. In the 1980s a great many people were trying to prove it wasn’t true. For obvious reasons doctors and hospitals were not eager to acknowledge that in some places they were overtreating people.
    They tried to poke holes in the reserach in every way possible, looking for rationalizaitons, other explanations. Dartmouth reserachers reponded by adjusting for different factors, doing studies where they compared patients who all sufffered from the same disease, comparing academic medical centers around the country, tying to think of every possible explanation that might account for such wide variations in how intensive care is in different places.
    At this point, it’s very unlikely that someone is going to find a major hole in the reserach that no one else thought of. It’s like thinking that someone may discover oil by drilling in Manhattan–that for all of these years, no one else noticed it.
    By the mid to late 1990s, virtually everyone agreed: the Dartmouth reserach is correct. These wide variations had nothing to do the patients medical needs or patient druthers. And where care was more expensive and more intensive, outcomes were not better–sometimes they were worse.
    See Shannon Brownlee’s escellent book: “Overtreated.”
    When I wrote about this both in my book and in articles in 2006 and 2007 you would be hard-pressed to fnd a medical expert who didn’t agree: the Dartmouth reserach had been questioned from every angle. It stands up.
    The only question was : What do we do about it?
    Then Obama was elected, and it became apparent that many in D.C. wanted to push ahead with health care reform. Both Obama and White House Budget Director Peter Orszag were very famliar with the Dartmouth reserach as were some members of Congress and the Medicare Payment Advisory Commissoin, which often quotes the Dartmouth reserach in its reports.
    At that point, some specialty organizations and others who profit from our current health care system became nervous. In particular they were nervous because the Dartmouth reserach suggests that we don’t need more specialists. The hospitals that profit from training specialists want to train more.
    So a campaign of misinformation about the Dartmouth research began. People said there really is very little waste in our health care system. The fact that people undergo more surgeries than in any other developed country is a good thing–though admittedly, we have no medical evidnece that Americans are healthier as a result . .
    The Dartmouth researchers themselves are still refining how they do the research, making varioius adjustments– this is what the new Health Affairs article is about.
    But as they explain why the variations in some areas may not be quite as extreme as some studies suggest, the variations are still very wide–and very expensive. The main thrust of the research holds up.

  10. I have an HSA, which is ok for the young and healthy, although I’m feeling good enough to get a high deductible health insurance. I’m thinking that maybe these forms of health insurance, with little paid in premiums, could possibly bypass this ‘wage index’ problem in big cities? Well, i’m hopeful…

  11. Those numbers are rather significant in their variation. It would be good to see how they correlate with measurements of health of people living in the respective areas. Is there any possibility that some people are benefiting by paying more?

  12. Maggie,
    The Dartmouth folks are very good at documenting WHAT is happening, but not WHY it is happening in some places and not others.
    Why do we see more “entrepreneurial” physicians in those particular areas? Why are there too many Cardiologists in a forsaken place like McAllen?
    I can see doctors flocking to Miami, New York, San Francisco and even Eugene, but not to Harlingen and Monroe. So why are they?
    We seem to be lacking the root cause.

  13. Maybe I’m not understanding this.
    Even though:
    “Darmtouth has been doing this reserach for more than 20 years; they have thought all of the possible explanation for these variations, and have discoverd
    only a few explanations…(sic)” – Maggie Mahar
    (In other words, “the science is settled”.)
    …they do another study, in response to critics, looking at other factors they had not looked at before, and adjust several regions by double digits (in percentage terms), some as much as 30 percent!?
    According to Mr. Goozner:
    “…some big cities with high poverty concentrations that also serve as training grounds for future physicians may have been unfairly lumped in with areas that overuse health care services.”
    In other words, they were significantly wrong and their critics were right? Is that what they just said?

  14. Merrill–
    Very informative piece.
    I would point out only that the fact that Manhattan hospitals are paid more for training residents, that is not the whole story.
    Manhattan hospitals are paid more because they are very powerful and NY politicians have struck deals with Congress and Medicare to make sure that they are paid way, way more– much as Florida hospitals have.
    In addition, when you look at actual utilization of medical resources, Manhattan patients are generally over-treated.
    Toward the end of the current Health Affairs article, the authors point out: “Measuring hospital days per decedent in the last two years of life for chronically ill patients implies that the Manhattan Hospital Referral Region is 78 percent above the national average—far exceeding the rates in Los Angeles (43 percent above average) and Miami (48 percent).
    (The Hospital Referral Region–level measures of end-of-life utilization are based on the decedent’s ZIP code of residence and thus include everyone in the Medicare sample with a chronic illness as determine by Iezzoni risk adjustment, including those without a hospital admission.)
    Manhattan patients spend more days in the hospital, where more specialists poke and prod them; they undergo more tests and more treatments. Moreover, Jon Skinner, one of the authors of the article, points out that when the reseracherse how many specialists see the patients, they don’t count residents.
    In Manhattan’s teaching hospitals, patients are treated by an unusual large number of specialist–plus countless residents. (The residents aren’t counted because they can’t bill.)
    I’d rather be in a quieter hospital where there is less going on– a better chance of getting out alive.

  15. What I don’t get about these conservations regarding the Dartmouth Atlas of healthcare data is that often too you compare ‘apples to oranges’ in terms of regions even though they adjust for some patient demographics (age, sex, race).
    Both of the TX regions (McAllen and Harlingen) are some of the poorest regions in the U.S. with incredibly high rates of unemployment (especially chronic unemployment), food stamp utilization, etc.
    Frankly, I am more interested to see how these regions compare against other hospital regions along the U.S. border in several states including CA, AZ, NM, and other parts of TX. I would also be interested to see how they compare against other rural parts of America that have similiar characteristics including parts of the Appalachia including KY, TN, WV, OH, PA.
    Same on the urban areas. Compare Detroit to other midwestern and northeastern regions that have undergone massive deindustrialization since the mid-1970s and on.

  16. Margait–
    Some of the Dartmouth studies adjust for overall “underlying health of the population”.
    For instance they factor in the fact that health should cost 30% more in certain areas because the overall environment is so much less healthy– and usually, the populatin is poorer as well.
    Evern after factoring in undelrying health of the population, they find dispairites confirming that care is more expensive in certain regions.
    Darmtouth has been doing this reserach for more than 20 years; they have thought all of the possible explanation for these variations, and have discoverd
    only a few explanations:
    –a medical culture which encourages “doing more” (and is oven a money-driven culture where doctors are extrremely entrepreuurial as in the Texas town Gawande wrote about in the New Yorker. A large number of physician-owned hospitals also leads to more hospitalizations, more surgeries, adn higher costs. Louisiana and Texas lead the nation in the number of physician-owned hospitals and they have among the highest costs. Also, when physicians lease or own diagnostic testing equipment and use it in their offices, they do twice as many tests.
    –A large number of hospital beds adn specialists leads to higher costs–and over-treatment. It’s not that specialists consciously “do more” when there are more of them, but they have more time. There are no fixed rules as to how often a cardiologist should see a heart patient. If he works in a town where there are twice as many cardiologists, he will have more space in his appointments book, and so may see her every three months rather than every six months. And more appointments lead to more tests. . . Similarly if he knows that there are plenty of beds available, he is more likely to hospitalize a patient . . .
    As Karen Davis, President of the Commonwealth Fund points out, two factors drive high health care spending in the U.S.
    1) we do more– more tests, more surgeries, more patients in ICUS, more intensive end-of-life care, more implants (hip, knee, heart devices). Our hospital stays are shorter, but more happepns to use while we’re there. We see more specialists and fewer primary care docs.
    2) we pay more for virtually everything–except nurses and primary care docs.

  17. What I would really like to see is if there is any correlation of these numbers with intangible factors like perceived well being, clean water, crisp air, beautiful scenery, stress levels, healthy behaviors (diet, exercise), leisure time…. Maybe take the Gallup-Healthways findings and adjust for where the local population of seniors spent most of their lives.
    I just can’t escape the indication that the lowest spending areas seem to be pretty pastoral and laid back by comparison to the high ones.