Why We Know so Little About ACOs: The Managed Care Culture at Work

flying cadeuciiThis is the third essay in a three-part series in which I explore the answer to that question. In the first installment  I blamed this problem on the flimsy definition of “ACO.” ACO proponents “defined” the ACO in terms of their hopes for it, not in terms of the mechanisms ACOs would use to accomplish those hopes.

In the second installment  I reviewed a paper published by the Center for Health Care Strategies (CHCS) to document my statement that we have no useful information on ACOs and to illustrate the quandary the hope-based “definition” of “ACO” creates for researchers. I criticized the CHCS paper as well for its cavalier attitude about evidence. The paper relied heavily on press releases and anonymous sources, and ignored the costs providers and insurers incur to set up and maintain ACOs.

In this last installment, I explore the role that culture – the culture of the managed care movement and the larger health policy community – played in elevating the ACO to the status of national health policy and, at the same time, thwarting the production of useful research on what it is ACOs do for patients.

The ACO isn’t the only example of hope-based health policy

If the ACO were the only example of an undocumented and poorly defined health care “reform” that was flogged from obscurity to fame by a few well-placed health policy entrepreneurs, we might dismiss the problems created by the flabby definition of the ACO as an aberration. But the ACO is not the only example of such “reforms.”

To the contrary, the ACO illustrates the norm, not the exception. It is an excellent illustration of how health policy has been made in America since the modern health care reform debate began circa 1970. Over the last half-century, every managed care “reform” that was eventually unleashed on all or large portions of the American populace followed the trajectory of the ACO:

  • It was proposed by a few activists with excellent connections to public- and private-sector policy-makers,
  • it was poorly defined,
  • it was often given a misleading label,
  • it was not shown by evidence to be safe and effective before it was implemented, and
  • it was allowed to stay implemented long after anecdotal and empirical evidence indicated it was not working well or at all. [1]

For an answer to what’s causing this problem, we need to examine the habits of thought within the health policy community that perpetuate this problem. These habits are part of the culture that arose within the health policy community over the last half-century.

In this essay I focus on three dysfunctional habits, mores, or standards that were first adopted by the managed care movement and then spread throughout the health policy community. I then offer an explanation for why these habits became so deeply ingrained and widespread. My explanation is that these habits contributed to the political success of the managed care movement. Just as organisms maintain and strengthen attributes that contribute to their evolutionary success, so the managed care movement nourished the dysfunctional habits that contributed to its political success.

The habits of thought I am criticizing are:

Abstraction. An affinity for abstract labels and phrases (for example, “arrangement,” “accountable care,” “coordinated care”);

Manipulative language. Frequent use of words designed to manipulate (to seduce readers into overlooking debatable assumptions) rather than inform (“health maintenance organization,” for example, assumed HMO doctors had been shown to “maintain health” better than non-HMO doctors because of some unidentified mechanism buried inside the HMO);

Disinterest in evidence. Heavy reliance on raw opinion, little reliance on rigorously conducted experimentation and research (in part because of the poor definition of the “reform” in question), and failure to measure the costs of the proposed “reform” (such as cost of setting up and running ACOs, or the administrative costs HMOs and PPOs impose on doctors and hospitals).

The birth of faith-based health policy

These three habits of thought emerged during the formative years of the managed care movement. Dr. Paul Ellwood, the man who “dreamed up” the HMO concept (as Philip Caper put it), and officials in the Nixon administration deliberately decided to give the HMO a manipulative name, they deliberately decided to “define” the HMO according to their hopes for it, and they enthusiastically made grand claims for the HMO based on anecdotes or no evidence at all. They did not propose subjecting the HMO to a pilot test before exposing large portions of the population to HMOs. Leading Democrats in Congress quickly adopted these same tactics.

The decision to adopt these tactics was initially made by Ellwood, Assistant HEW Secretary Lewis Butler, and three other representatives of the Nixon administration at a meeting at the Dupont Plaza Hotel in Washington, DC on February 5, 1970. At that meeting, Ellwood argued, on the basis of no empirical evidence, that the fee-for-service system was the root cause of health care inflation, and that “health care corporations” that shifted insurance risk onto doctors were the solution. Ellwood initially had Kaiser Permanente in mind as the model for these corporations, but Butler argued against describing this new-fangled entity in any detail. “Why should we specify how to put it together?” Butler argued. “Let the doctors – let everybody do it, figure out how to put it together. Let’s specify what we want it to do. And we don’t give a damn how they put it together…. Let’s describe the thing by what we want it to do, not how it’s formed.” (My emphasis) [2] As Butler explained later to Joseph Falkson, if this new-fangled corporation were “defined” by “what we want it to do,” Ellwood’s “idea could be sold … as a market reform strategy rather than yet another federal program requiring a large bureaucracy … to manage it.” [3]

Ellwood immediately agreed to Butler’s rationale. The five men decided that the only information the Nixon administration would reveal about this new-fangled corporation, other than what they hoped it would do, was that it would be subjected to the reverse of the fee-for-service incentive – so-called “pre-payment,” which was just another word (an unnecessarily abstract word) for “premium payment.” This reversal of financial incentives would lead to good things, and that’s all anyone needed to know.

The five men then debated what to call this corporation. They carefully selected a label designed to persuade rather than inform. They selected “health” rather than “medical” to suggest, on the basis of no evidence, that when doctors bear insurance risk they get better at ordering mammograms and flu shots. They chose “maintenance” because it connotes maintaining health rather than treating illness. And they chose “organization” because it carried less baggage than “corporation,” “group practice,” and “partnership,” three other phrases they considered. As Falkson put it, the new label “was a politically nebulous and, therefore, desirable phrase.” [4]

Before the meeting broke up, Ellwood promised to draft a paper summarizing the rationale for this new HMO idea. The paper, which was drafted in March 1970 and eventually published in Medical Care, contained not a single footnote. [5] It made no mention of the additional costs Ellwood’s proposal would create, including higher administrative costs for providers and the cost of measuring “quality.”

Thus, at the very creation of the managed care movement, its leaders adopted a proposal with no empirical evidence to support it, and they deliberately rationalized describing their proposal in amorphous, aspirational and manipulative terms on the ground that those terms would make it easier to sell their proposal to legislators and the public.

If at this point Democrats in Congress had objected to such tactics, those tactics might have died in their cradle. But leading Democrats, including William Roy in the House, adopted those tactics as their own. Those tactics were successful. Congress enacted legislation in 1972 allowing HMOs to participate in Medicare, and in 1973 Congress passed the HMO Act.

Other proponents of HMOs and of subsequent iterations of what came to be called “managed care” quickly adopted those tactics. They repeated Ellwood’s diagnosis (that fee-for-service is the root cause of health care inflation) and his solution (insurance risk must be shifted to doctors) even though they had no empirical evidence for the diagnosis or the solution. Over the coming decades they would, as Ellwood et al. had, assert their opinions regardless of the paucity of evidence for them, they would use vague language rather than concrete terms, and they would use words designed to persuade rather than inform.

These tactics help explain why managed care has been so successful politically – numerous bills based on managed care theology have passed Congress and state legislatures, managed care proponents occupy leadership positions throughout the media, business, politics, academia and the foundation world, and managed care dominates the professional and lay media discussion of health policy (see, for example, the first page of this paper by David Mechanic  ).

But these tactics also explain why managed care has failed to produce a solution to the US health care crisis, and may in fact have aggravated that crisis.

Other examples of the managed care culture at work

By approximately 2000, it was clear that the HMO and its hybrid spawn had failed. As Arthur Caplan put it in 2001, “Events of the past year demonstrate beyond a doubt that managed care has failed – and failed dismally. The greatest single ethical crisis facing American health care as we move into the new year is what to do about it.” [6]

In the early 2000s, an ostensibly new form of managed care emerged. Its principle elements were “pay-for-performance,” ACOs, and “medical homes,” and the universal adoption of EMRs to facilitate all of the above. But this version of managed care, call it Managed Care 2.0, is also failing to cut costs, is having (at best) mixed effects on quality, and is aggravating other problems, including physician burnout and the consolidation of our system into mega-systems.

Managed Care 2.0 is failing for the same reasons Managed Care 1.0 failed: The habits of thought that proved so successful in creating and selling the HMO also influenced the creation and selling of Managed Care 2.0. The rationale for unleashing P4P, ACOs, “homes” and EMRs on the populace was expressed in abstract and often manipulative language, there was no empirical evidence to support the claims made for those proposals, and advocates pretended the intervention costs of those proposals were zero or so trivial they needn’t be mentioned. [7]

We need research on the culture of the health policy community

In this three-part series of comments, I have sought to explain why we have no information on what services ACOs provide to patients that non-ACO providers do not provide. The immediate cause of this problem is the amorphous, wish-based, un-testable definition of “ACO.” But the more fundamental cause is a set of norms or mores that has evolved within the health policy community that not only tolerates but encourages sloppy thinking. These norms were created at the dawn of the managed care movement, and spread throughout the health policy community, a community which emerged at about the same time the managed care movement was forming. Those norms are not even recognized by the vast majority of health policy analysts, much less analyzed.

The health policy literature is rife with negative opinions about physician “culture.” Those opinions are usually offered to explain why some managed care proposal is necessary, or why some managed care proposal didn’t work as advertised. According to those who opine about physician culture, it is a culture of “resistance” to the bright ideas that flow from the health policy community. I urge the health policy community, especially those members who unabashedly promote ACOs and other iterations of managed care, to consider the possibility that their culture has played a prominent role in the chronic failure of managed care propositions to work as advertised.

[1] In addition to the ACO, the major “reforms” that fit this description include, in approximate order of appearance: The HMO, the PPO, utilization review, drug formularies, report cards, disease management,  pay-for-performance, the “medical home,” and forced use of electronic medical records.

[2] Joseph Falkson, HMOs and the Politics of Health System Reform, American Hospital Association and A. Prentice-Hall Company, 1979, 31.

[3] Ibid, 32.

[4] To give you some idea of how vaguely other HMO advocates defined this entity, consider this statement by Elliot Richardson, Nixon’s HEW Secretary: “I adhered from the outset to the definition of an HMO, namely that it is any arrangement which has the function of allocating prepaid dollars over … a comprehensive range of needed health services. The only thing that an HMO does for you is to create a point at which somebody has to look at how dollars are distributed.” (Falkson, op cit., 53) “A point at which someone has to look at dollars” takes the blue ribbon for vagueness.

[5] Paul M. Ellwood, Jr. et al., “Health Maintenance Strategy,” Medical Care, 1971;9: 291-298.

[6] “In 2001, managed care our No. 1 health crisis,” MSNBC, December 21, 2001.See also

Linda Marsa, “Former JAMA editor laments the state of medical care,” Los Angeles Times, March 26, 2001,


[7] I will not attempt to document that statement here. I have commented elsewhere on the role of the habits of thought I am criticizing in creating and sustaining the hype about P4P http://pnhp.org/blog/2014/08/12/to-err-is-human-the-p4p-fad-illustrates-the-problem/, the ACO (see Parts I and II of this series), the “medical home,” http://pnhp.org/blog/2015/02/24/cmss-medical-home-experiment-is-a-mess/, and EMRs. http://www.healthcareitnews.com/blog/big-data-latest-fad-health-policy For a longer and more documented presentation of my criticism of the culture of the managed care movement, see “Part II: The Managed Care Movement” in a paper I wrote with Ted Marmor http://digitalcommons.law.yale.edu/cgi/viewcontent.cgi?article=1231&context=yjhple.

Kip Sullivan is a member of the board of the Minnesota chapter of Physicians for a National Health Program.

Categories: Uncategorized

Tagged as:

11 replies »

  1. This article and the previous ones by Sullivan offer some useful insights clouded by ideology and what I can only call a bizarre attitude towards political labeling. It’s not just in health care reform where various ideas are given different names to make them more attractive — or has he forgotten the post-9/11 “Patriot Act”? And if his concern is political labeling, rather than merely attacking managed care, how about the anti-managed care legislation signed in 1996 under the Clinton administration, the “Newborns’ and Mothers’ Health Protection Act”? Care to vote against that? (It regulated the stay of mothers in the hospital post-birth.)

    Even more interesting for a liberal commentator, Sullivan doesn’t mention the outrageous, but very successful, relabeling of high-deductible health plans as “consumer-driven health plans.” Like consumers asked for higher deductibles! And, by the way, no surprise: they don’t control costs very well, only shift them.

    So: we know that health care didn’t invent misleading legislative names and that all of this existed well before Sullivan starts his story about Ellwood. But let’s go back to that history, which I researched in my book, Demanding Medical Excellence: Doctors and Accountability in the Information Age. (The 1997 book has a chapter on managed care that talks about horror stories and — much more horrible to the single-payer folks — also talks about accomplishments.) Ellwood tried to get the Democrats interested in prepaid group practices; yes, essentially based on Kaiser. They were convinced that they had the votes to get single payer passed. Ellwood went to the Republicans and they adopted “health maintenance organizations” as the relabeling of those socialist “prepaid group practices” and managed to outflank the single payer folks. Nixon outsmarted them.

    The real history lesson: arrogance in 1973 by the Democrats doomed health reform then. Arrogance in 1993 by the Democrats (Hilary refused to compromise) doomed health reform then. Arrogance by Democrats in Boston, home of the single-payer folks, almost doomed health reform in 2010 when they lost the safest of safe seats to a Republican. Sometimes, a little compromise pays off. (Obama tried to, by the way, but the GOP wanted a death fight.)

    Now, to ACOs. Suffolk U’s Marc Rodwin has written extensively on the pros and cons of different payment mechanisms for physicians across the world. Whether it’s salaries, fee-for-service or some sort of incentive payments, there are pros and cons. ACOs are a managed care hybrid. Conservatives don’t like them because they’re not Medicare Advantage, a free-market invention that benefits from government overpayment, kind of like all those free-maret farmers and oil drillers. Common-sense economists don’t like them because they are built on a FFS chassis and patients don’t even know they’re in an ACO. They are a political compromise, meant to give providers, not insurers, the management power (preventing a political backlash) and preserve maximum choice for beneficiaries (prevention a political backlash).

    They are a transition product, but as the rules change, they may become more effective. I’m not in love with the way CMS has implemented the program, but, unlike Sullivan, I know that political reality is, indeed, a form of reality, and that labels and perception are important. Yes, high prices are a big part of the problem, but we also need organizational structures to start to squeeze down prices and improve quality of care. If that’s not going to be done by Speaker Ryan and Majority Leader McConnell, then structures like ACOs will be needed. Even while the single-payer folks, like their conservative counterparts, nurture dreams of one grand ideological experiment that solves health care reform once and for all.

    Well, their version of reform: access. My issue is quality and safety, but the left- and right-wing access ideologues don’t really care much about that. A different topic.

  2. Fee for service works in dentistry, plastic surgery and lasik….and accountancy, law, architecture, civil engineering (I include optional bundling and optional subscription/fixed price models as simple variants on fee for service)…and now concierge/Direct Primary Care…..I still haven’t heard an explanation of why not medical services. What we have now isn’t really pay for service because of third party payment systems which eliminate patient financial benefit from searching for best price/value offers.

    Civissisus makes a great point that most innovation does come from carefully controlled classical experimental design. But I disagree that it doesn’t come from “empirical evidence”….the empirical evidence is profit and earnings growth in the for profit sector, and revenue growth and budget surpluses in the non-profit sector. But this only works in a positive way when providers and customers are free to choose. EHR’s were for all intents and purposes mandated by HiTech and therefore rolled out system wide on…as Kip says….a hope based initiative that has proved disastrous (some decade in the future we may get a positive result). Now Vermont is looking to force virtually all physicians and patients into an ACO….another “hope based” initiative. The only choice option will be for docs to leave the state….exacerbating a current problem in Vermont.

  3. David, you’ve responded far more diplomatically, and productively, than I might have.

    My wiseass variant would have been something along the lines of “so it’s STILL the prices, stupid, and FFS is still no better than any other available option”.

    I particularly like your observation about the nature of policy innovation in general, which seldom proceeds from the kind of “empirical evidence” that Kip places such great faith in, in any realm of endeavor, much less in medicine.

  4. The last article in this series made more sense. This one is weakened by the same lack of specificity that the author effectively criticizes the health policy community for. For starters, the author lumps together under “managed care” everything from health plans with neutral incentives for physicians (e.g., Kaiser Permanente arrangements) to HMOs and PPOs with fee-for-service physician reimbursement combined with strict gatekeeping, then goes on to suggest that because the latter failed (which was due to consumer and physician backlashes), that all managed care has failed.

    I’d argue instead that fee-for-service institutions and its political supporters have managed to overwhelm most policy attempts to overcome it, from “disease related groups,” which has been twisted over time toward more and more procedure based payment, to the financial squeeze that choked out Southern California’s family-physician groups in the 1990s and their innovative management of care. Fee for service dominates physician and hospital reimbursement still, despite many public and health plan policy attempts to replace it.

    Unless I’m mistaken, Kip, you support Medicare for all Americans. While I can see advantages over today’s arrangements simply because coverage would be universal, there is little proof, which you want us to rely on, showing that Medicare offers greater value (i.e. quality outcomes produced efficiently) than other insurance arrangements.

    We face difficult policy issues because of the high price of current care approaches, the rapid evolution of medical technology, and our entrenched institutions that resist change. While most can agree with your theme that plan-do-check innovation cycles in health policy can be more effectively implemented and tracked, that criticism applies to all policy options, not just ACOs and managed care.

  5. Splendid! Thank you.

    Health care policy has been a tiresome and perpetual
    trainwreck. And, mirabilis dictu, catastrophic policy wreckers are never accountable….although they have repeatedly led the entire nation astray and cost trillions of dollars.

    Docs have been forever dissuaded from coming together in political groups using the old anti-trust arguments Zzzzzz.
    We could have, and should have, come together as groups of patient agents demanding things like billing clarity and transparent prices, adequate mental health coverage, stability in insurance coverage, sufficient numbers of cardiac cath units and trauma centers, questionabe ethics of the hospitalist movement, etc….heavens, there were plenty of issues for docs to get together and gather solidarity and exert political power and not be accused of price fixing. And we should continue efforts to gain power no matter the depressive scene.

    Now, with so many physicians indentured to hospitals, I wonder if we can ever be united forceful leaders in our
    profession’s future. Anyway, missives like this help us greatly and we have to embrace the future. It is nice to see some testosterone once in awhile.

  6. Excellent! I would add that we have to also consider that advocates who launch these ideas are after 1. power and prestige and 2. control over huge funding streams.
    Your series is much more comprehensive than mine, which is expressed in “Reforming the Reformers” in THCB and the Raleigh News and Observer….the THCB version linked below:

  7. Two words describe the entire managed care movement. “Sloppy thinking”

    You did a great job in enlightening anyone that takes the time to read this piece.

  8. Love this series. It was fascinating and is like music to my ears. I now have credible reasons for my distain for the new ACO, “Alternative Payment Model” Care.

  9. This piece, like the previous two in the series, is excellent. My only minor criticism is that if understood properly, “managed care” predates Paul Ellwood by about 60 years, and the history of our healthcare system is precisely that of a continued re-invention of this failing concept. I give the historical overview in broad brush strokes here http://bit.ly/1KHCd1R