Editor’s note: See additional posts on the Medicare Shared Savings Program Final Rule  and related delivery system and payment reform initiatives by Debra Ness and William Kramer, Douglas Hastings, Mark McClellan and Elliott Fisher, and Don Berwick and Richard Gilfillan.

On October 20, the Centers for Medicare and Medicaid Services issued its final rules for the Medicare Accountable Care Organization (ACO) program.  Simultaneously, the Office of the Inspector General (OIG) of the Department of Health and Human Services, the CMS Innovation Center, and the antitrust agencies – the Federal Trade Commission (FTC) and Department of Justice (DOJ) –  published new policies related to ACOs.

The final CMS rules include a 622 page preamble that is a model of transparency.  Section by section, the preamble describes the originally proposed rules and their rationale, the public comments received in response, and the CMS response to the comments.  In each case, CMS lays out possible alternative rules, their advantages and disadvantages, and the reason for its final decision.

One way of thinking about the final ACO regulations is as organizing a race.  This race is not a 100 yard dash, or a 5k or a 10k but, rather, a marathon.  The organizer, CMS, originally laid out a course that potential entrants found too difficult – too many hills and twisting turns. So now the organizer, CMS, has laid out a course that appears more navigable.  In addition a “head start” is being given to the slower, inexperienced runners through an Advanced Payment program. There is no question that the final rules, along with the new statements from the FTC/DOJ and the OIG, will make it easier and more attractive for physicians and hospitals to form ACOs.  Will these changes be sufficient to entice greater numbers to the starting line?

The Final Rule Is Much Less Burdensome, But One Fundamental Problem Remains

Initial reactions from hospital and physician associations have been cautiously positive, but everyone is still in the process of digesting them and trying to figure out the implications for their own organization.  The rules will make it far less burdensome to create and to operate an ACO.   For example, prospective ACOs will not be required to assume downside risk during the first three years.  They will not be required to calculate their market share for each category of service, and to obtain FTC approval before contracting with CMS.  The number of quality measures has been reduced from 65 to 33.  Assignment of patients to ACOs will be prospective, with a retrospective adjustment, making it much more possible for ACOs to know who their population of patients are.

One basic problem remains – a problem that is built into the core of the Shared Savings Model: at best, the model permits an ACO to receive 60 percent of the savings that it created, with CMS taking the other 40 percent.  To create a dollar in savings, the hospital or medical group must give up a dollar of Medicare revenue.  This dollar of gross revenue would make a contribution to both the fixed costs of keeping the hospital or medical group operating and to the marginal cost of providing the service that, if provided, would gain the dollar of Medicare revenue for the organization.  Each organization will have to decide whether the sixty cents in shared savings that it can, at most, receive is worth more than the dollar in gross revenue that it is giving up.

But the calculation cannot end there – the organization must also consider the cost of creating and operating an ACO.  CMS estimates that the average ACO will need $580,000 in start-up costs and $1,270,000 in annual operating expenses.  Even after spending this money, there is the possibility that the organization will incur the costs of creating and operating the ACO but will not receive any shared savings bonus from CMS, because it fails to generate savings and/or fails to score highly enough on quality metrics.  Additionally, an ACO that chooses the second track (the track that permits it to receive 60 percent, rather than 50 percent, of savings generated), risks having to pay CMS a share of any costs that exceed the predicted costs for the ACO’s population of patients.

In other words, the shared savings “bonus” is not really a bonus.  Under the program as designed even an efficient, high quality ACO will gain less money from sharing in savings than it would have earned if it had simply continued with business as usual.  And there is no real bonus for quality – at best, an ACO can receive the maximum amount of shared savings possible – i.e., at most 60 percent of the savings it created for CMS – but no additional funds for quality.  This is a fundamental flaw in the design of the program – a flaw created by Congress, and which only Congress, not CMS, could remedy.

It would (in theory) be possible for Congress to create legislation that would, going forward, make the payment rates to all physicians and hospitals lower than they would otherwise be.  Part of the savings from the lower payment rates would be retained by CMS, and part would be put into a pool that would reward high quality ACOs.  This would save money for Medicare and would make it possible for an ACO that both generated savings and had exceptionally high quality to earn more than it would have earned if it had not joined the ACO program.  Such a program would not involve the government in picking winners and losers.  Physicians and hospitals in an efficient, high quality ACO could earn more than they would have earned if they were not in an ACO, even after accounting for the costs of creating and operating the ACO.  But physicians and hospitals in ACOs that did not perform well would be better off financially if they had not been members of an ACO, because they would not receive a bonus, yet would incur the costs of creating and operating the ACO.

Why Might Providers Form ACOs?

There are many other factors that hospitals and physicians will consider in deciding whether to create an ACO.  Most savings are likely to come from reducing the volume and cost of services provided by hospitals; since physicians would share in these savings, this gives them an incentive to join ACOs (in other words, a dollar in revenue lost by a hospital generates savings in which physicians can share).  Similarly, more savings are likely to come from reductions in the revenue of specialist physicians than from primary care physicians, so the latter have more incentive to join ACOs.  Nevertheless, specialist physicians who want to retain their primary care referral base are likely to participate in ACOs.  An ACO can also help physicians purchase and effectively use electronic medical records, and can put care management processes in place that help physicians score better on measures of quality that will soon be publicly reported by Medicare and by health insurance plans.

ACOs can provide many benefits for hospitals besides sharing in savings.  First, many hospitals are operating at full capacity; if an ACO enables them to reduce unnecessary admissions and readmissions of Medicare patients, they can share in the savings from the ACO program AND replace these admissions with more profitable commercially insured patients.  If hospitals contract as ACOs with private health insurance plans and reduce unnecessary admissions of commercially insured patients, they can open beds for patients for whom they did not previously have capacity, thus expanding their market share.  Everyone benefits (except hospitals that lose market share).

Second, CMS and private health insurers are increasingly providing financial and public reporting incentives for hospitals to avoid unnecessary readmissions, reduce inpatient infections, and score well on quality measures.  Public and private payors are also moving toward bundled payments for some procedures – e.g. for orthopedic and cardiac surgery.  Hospitals need cooperation from physicians to respond to these payor initiatives, and participating in an ACO may be a good way to gain this cooperation.

Third, ACOs provide a means for hospitals to create tighter ties with physicians and to funnel resources to them (e.g. for purchasing electronic medical records); these are attractive objectives for many hospitals.  The ACO policies just promulgated by the OIG will facilitate hospitals’ efforts to work more closely with physicians and to share resources with physicians.

Who Will Create ACOs? 

Hospitals have the capital to create ACOs, and can do so by linking with physicians; they can employ the physicians and/or work with physicians organized into one or more large medical groups, an independent practice association (IPA), or a physician-hospital organization (PHO).  Large medical groups and IPAs could take the lead in ACO formation and establish loose affiliations with hospitals.  In these arrangements, more of the savings and bonuses might go to the physician entity, with savings for hospitals occurring on a more selective targeted basis such as bundled payment savings for specific procedures or conditions.

What about physicians in the small and medium-sized practices that are still very common in the U.S.?  For the past decade, many physicians in these practices, and many physicians who have just completed training, have been choosing to join a large medical group (if the opportunity exists) or to become employed by a hospital.  There are many reasons for this trend; if physicians begin to believe that ACOs will become common, the trend is likely to accelerate.

Physicians who do not wish to become employed by a hospital or to join a large medical group do have an alternative.  The ACO concept provides stronger motivation for physicians to “clinically integrate” into an IPA (or a PHO, which is essentially an IPA linked to a hospital).  Clinically integrated physicians can gain the benefits of participating in the CMS ACO program AND the benefits of negotiating payment rates and other contract terms with health insurance plans.  In addition, clinically integrated organizations can help physicians with care coordination and other organized processes that make it possible to provide higher quality care and to gain higher scores on quality performance measures.

The Advance Payment Program

The Advance Payment Program announced by the CMS Innovation Center simultaneously with CMS’s publication of the final ACO rules will provide assistance to physicians (and to small hospitals) that lack capital to create an ACO.  The program will provide $170 million to up to 50 ACOs to help with start-up costs.  In selecting the 50 ACOS, preference will be given to ACOs that are small, rural, and care for a higher percentage of Medicaid patients.  $170 million is $3.4 million per ACO – a significant sum.  The funds will be gradually recouped by CMS through savings generated by the ACO.  If an ACO fails to generate sufficient savings, CMS will not require the ACO to pay back the remaining balance.

These terms might make it very attractive for a medical group, IPA, a Federally Qualified Health Center, or critical access or rural hospital to accept the Advance Payments and to choose Track One of the ACO program, which gives the ACO a somewhat smaller percentage of savings, but does not give it any downside risk.  These advanced payments can be used to create clinically integrated programs to improve care.  Even if the program does not succeed as an ACO, it will benefit from having these programs – and from the experience of creating them – without bearing any financial risk.  This should give a significant boost to those organizations serving largely safety net populations and to Medicaid ACO innovations occurring in Camden, New Jersey, Colorado, and California.

A Diffiult Task for The Antitrust Agencies

Antitrust law makes it illegal for independent physicians (i.e. those in separate practices, independent from each other and not employed by a hospital) to negotiate jointly with health insurance plans unless they are financially integrated (to oversimplify, this means taking capitation together) or clinically integrated.  Put briefly, the FTC considers physicians in an IPA or PHO to be clinically integrated if they invest substantial financial and human capital in creating and maintaining organized processes to improve the quality of medical care and to control the overall cost of care through increased efficiency and reduction of the amount of unnecessary care provided.  The FTC permits clinically integrated IPAs and PHOs to negotiate with health plans as long as their market share (and thus their negotiating leverage) is not excessive.

The initial CMS proposed ACO rules would have made it mandatory for organizations that include independent physicians and that have a high market share in even one specialty area to gain FTC or DOJ approval before they could be designated as an ACO.  The final rule drops this requirement; the new FTC/DOJ statement simply asserts that the agencies will continue to enforce the antitrust laws.  If large numbers of ACOs are formed, the burden on the agencies to identify ACOs with enough market power to substantially raise payment rates – and thus the cost of health care – will be substantial.

There are other problems with the final rules.  Notably, when CMS performs risk adjustment annually to predict the expected cost of care for an ACO’s population, it will use only demographic variables, and not also the patients’ diagnoses, to update the risk adjustment for beneficiaries who continue with the ACO.  CMS made this decision because of concern that some ACOs might inflate their beneficiaries’ risk status through “upcoding.”  Unfortunately, it seems likely that the decision will discourage participation by organizations that, because of their location and/or their reputation, are likely to attract high risk beneficiaries.

IF And IF And IF . . . U.S. Healthcare Could Be Fundamentally Changed

Ten years from now, will October 20, 2011 be remembered as the day that a fundamental, irreversible transformation of the U.S. healthcare system began?  CMS projects that 50-270 ACOs will sign contracts with Medicare during the first year or two of the program.  It appears that an additional 30 ACOs will be formed in response to the CMS Innovation Center’s Pioneer Program.  Meanwhile, some large health insurance companies are signing ACO-like contracts with major provider organizations – the Blue Cross Blue Shield of Massachusetts Alternative Quality Contract is the best known to date – and in some cases are providing financial assistance to the organizations (on October 18, for example, Blue Shield of California announced $20 million in grants to 18 provider organizations to assist them in forming ACOs).

It is plausible that in 2015 there will be perhaps one hundred Medicare ACOs and many more working under ACO contracts with health insurance companies.  Indeed, the CMS and health insurance plan programs will be mutually reinforcing, leading to more ACOs being created than would be the case if only CMS or only private health insurers had ACO programs.  IF these ACOs save money for payors and score well on quality metrics, and IF patients and physicians like their experience in ACOs, there is likely to be substantial pressure on hospitals and physicians to form ACOs, and on health insurance plans to create more opportunities for them to do so.  In that case, the organization of healthcare in the U.S. could change quite rapidly.

This change will happen only if CMS and health insurance plans make it possible for ACOs to function effectively (for example, by providing timely and accurate data to them) and if physicians and hospitals are able to make very substantial improvements in the ways that they provide care for patients.  This will not be easy.  But if done right, ACO programs could make it possible, for the first time, for physicians and hospitals to be paid for systematically improving the quality of care they provide for their patients, rather than simply for providing a high volume of services.

It is important to recognize that this is going to be a marathon spread out over many years.  Just like in a marathon, the “runners” will have to adapt to the pace of the race, the runners around them, timing, and weather conditions.  And, what happens next year if the left-handed jump shot artist from the South Side of Chicago is no longer President of the Organizing Committee for the marathon?  Will the new President continue to supply the Gatorade to the runners?

In 1919, with Germany’s economic and political system in crisis, the revered sociologist Max Weber spoke to the Free Students Union of Munich University.  He told them that “Politics is a strong and slow boring of hard boards.  It takes both passion and perspective.”  So it will be with ACOs – will there be sufficient numbers of hospital, physician, and political leaders with the passion and perspective to stay the course?