The ACO Model Faces the Eye Test

The pros and cons of accountable care organizations and what their inception means for ophthalmology.

Walter Bethke, Managing Editor

At some point in ophthalmologists’ careers, they may have entertained fleeting thoughts of what life would be like ringing up big reimbursement charges like cardiologists do. However, with the threat of increased oversight and decreased income posed by the advent of the accountable care organization provider model, ophthalmologists may ultimately be grateful for their relatively efficient, modest billing. Big billers such as cardiology may find themselves wearing targets on their backs if and when ACOs take root.

In this article, experts in the field of medical reimbursement and the implementation of new provider models such as ACOs discuss the pros and cons of this new approach to health care and give ophthalmologists a look at where they might fit in.

The ABCs of ACOs

Before delving into the potential advantages and drawbacks of an ACO and the viability of a relationship with one, it helps to know exactly what the Centers for Medicare & Medicaid Services means when it uses the term.

The new Patient Protection and Affordable Care Act empowers Medicare to contract with health-care providers who provide care to at least 5,000 patients to create ACOs: networks of hospitals and/or physicians working together to coordinate care for a certain population of beneficiaries while trying to eliminate unnecessary tests, procedures and hospital stays. In practice, the ACO closely tracks its spending in serving these patients and then, depending on how the ACO is structured, at the end of the fiscal year it will either receive a monetary bonus based on the amount of savings it generated compared to a comparator population of similar patients, or it may get hit with a financial penalty based on its overspending.

However, the ACO isn’t all about saving money. The other major ACO goal is maintaining quality. To ensure patients are receiving quality care—as well as to prevent simply rationing care to pump up the savings numbers in order to reap bigger bonuses—the ACO is required to hit certain quality metrics, which currently number 33. The quality metrics fall under the headings of patient/caregiver experience, care coordination, patient safety, preventive health and at-risk population/frail elderly. (Some of the specific quality measures by which physicians will be judged appear in Table 1, p. 39.) 

Table 1. A Sampling of ACO Quality Measures1 
Quality Domain
Patient/caregiver experience
Timely care, appointments and information
Patient/caregiver experience   How well your doctors communicate
Patient/caregiver experience   Access to specialists
Care coordinator/patient safety   Falls; Screening for fall risk
At-risk population—Diabetes   Diabetes composite: Hemoglobin A1c control (<8 percent)
At-risk population—Ischemic vascular disease   IVD: Complete lipid panel and low-density lipoprotein control (<100 mg/dL)
At-risk population—Coronary artery disease   CAD composite: Drug therapy for lowering LDL cholesterol
The ACO final rule from CMS allows for an ACO to choose its level of risk in the first three years of the project: one-sided or two-sided. In the one-sided model, an ACO formed in 2012 will share in savings, but not losses, for the three years. For future contracts, though, the one-sided ACO can’t choose that option again. The two-sided risk model, which the CMS final rule says is for “experienced ACOs,” offers the opportunity to share in a greater percentage of savings in exchange for sharing the risk for losses starting in the ACO’s very first year.1

Questions About the Model

As ophthalmologists contemplate their potential relationship with ACOs, it helps to hear what some feel are drawbacks to the ACO model. Here are some of the main concerns that have been raised:

 • Ophthalmologists’ reimbursement will decline in an ACO. Shachar Tauber, MD, is an ophthalmologist in Springfield, Mo., and is chairman of an ophthalmology group practice owned by Mercy health system, a large hospital-based provider. Dr. Tauber and his colleagues are currently transitioning to an ACO, which will be finalized in January 2013. He says the ACO model appears to be a solid idea, but there’s the potential for a shake-up in reimbursement levels. “Will reimbursements go down?” he asks. “Compared to the time before the Affordable Care Act and ACOs go into effect, yes. Someone looking at the tea leaves will see that, in two years, he’s not going to get paid the same amount of money for the same amount of work that he’s doing now. Is he going to be paid more fairly in an ACO than he would as a private practitioner? Will he have fewer costs and overhead expenses and will his take-home pay be greater? Or will he exchange all that for a better quality of life, i.e., will there be less running around inquiring about access for patients and less time spent on marketing? I don’t know... . In the discussions I have with other ophthalmologists—some of whom are great friends of mine—many of them have left academic medicine and have gone into private practice and are making multiples of what I’m earning. I don’t know what the answer is for them. If I were a private practitioner, my biggest fear would be about how I could dare invest in capital if I can’t predict what my income or access to patients would be for the next year. That’s the area where the ACO model and participating in an integrated system works really well ... . The integrated ophthalmologist is now a viable career path.”

Stephen Kamenetzky, MD, professor of clinical ophthalmology and visual sciences at Washington University School of Medicine, is the American Academy of Ophthalmology’s adviser to the AMA Relative Value Scale Update Committee and he’s a consultant to the Academy’s Health Policy Committee. (He notes, though, that his views on ACOs are completely his own, and not the Academy’s.) Dr. Kamenetzky thinks the ACO’s method of determining how much you saved or overspent by comparing your practice to another one with similar patient disease levels and other actuarial data is difficult. “The problem is that, with time, practice patterns change and it becomes harder and harder to save money versus the [government or other organization chosen] control group, because your control group often behaves similarly to your group, in that it’s saving money too,” he says. “This makes it harder and harder to keep realizing savings. I think people change their behavior in response to what’s being measured. The problem is what’s being measured isn’t always the right thing to measure; choosing a comparison group that is really the same as yours is tough.”

Dr. Kamenetzky is also concerned that the entire ACO approach may, unfortunately, eventually evolve into capitation. “I think capitation is what this approach really is,” he continues. “This is just another variation of fixing the size of the pot first. This means x amount of dollars per patient per month or per year, and if you can’t do it for that amount, you’ll have to come up with some more money because you can’t stop taking care of them.”

Elliott Fisher, MD, director of the Center for Population Health at the Dartmouth Institute for Health Policy and Clinical Practice, is one of the architects of the ACO concept. He and Mark McClellan, MD, former administrator of CMS, are co-directors of a joint Brookings Institution-Dartmouth program to advance ACOs. He views the concerns about ACOs becoming capitated programs in this way: “They already have a virtually capitated system,” he says. “So, it’s a target budget, and all of the costs for the included patients come out of that target budget. So the incentive is there already for providers to try to reduce costs while hitting quality targets. It’s not hard capitation; the penalties for going over it—the risk bearing—are shared. Certainly, under the Medicare Shared Savings program [ACOs with one-sided risk], there’s no penalty for going over. Under the Pioneer program [which shares both upside and downside risk] there are slightly different ways of sharing the risk and having some risk borne by providers.

Table 2. Highlights of the ACO Agreement1 
Subject    ACO Final Rule 
Three-year ACO “tracks” and their risk levels   Track 1, one-sided risk: Share on first dollar in savings after a minimum savings rate is met, no share in losses; Track 2, two-sided risk: Higher sharing rates for money saved, but also share in any losses
Assignment of patients to ACO   Two-step process: 1) for patients who have received at least one primary-care service from a physician, use the plurality of allowed charges for primary-care services rendered by PCPs; 2) for patients who haven’t received primary-care services from a PCP, use the plurality of allowed charges for primary-care services rendered by any other ACO professional

Proposed quality measures   33 measures; get paid just for reporting on the measures in the first year, then get paid for reporting and performance in the second and third years
Electronic health record use   Not required, but is weighted higher than any other quality measure for quality scoring purposes
“There may be additional payments [to physicians] for other services,” Dr. Fisher adds, pointing out the potential upside. “The ACO model encourages these organizations to both support and pay for services that aren’t otherwise paid for under Medicare or by private payors, such as funding for care management, determining how to put quality protocols into place, or participating in clinical improvement and clinical integration activities. The downside is there may be fewer procedures. ACOs will want to be quite thoughtful about trying to reduce avoidable, non-beneficial care, such as elective stents or elective imaging procedures in cardiology.”

Jennifer Jackman, senior vice president of accountable care at Monarch HealthCare in Orange County, Calif., one of the Brookings-Dartmouth pilot ACO sites, says she is aware of specialists’ concerns, as well. “In the initial go-round, I know that a lot of specialists and hospitals are concerned that ACOs mean fewer referrals,” she says. “Right now, it’s not really going to change anything. In the future, it could potentially mean less business, but it should be business that wasn’t necessary. Less waste, you could say. That frees up your capacity to perform services for people who need them.”

 • Patient freedom of choice hinders ACO savings. Part of the Medicare ACO concept is the stipulation that patients retain their freedom to choose their provider. Critics wonder, then, how an ACO can reliably save money if its patients are free to leave it to visit a non-ACO provider that doesn’t have to conform to any of the ACO saving or quality requirements. 

The need to keep patients in the ACO, and the need for them to take an active role in their health care are so important that the question of what incentives ACOs could legally offer patients came up during the comment period of the Federal Register’s ACO announcement. In the discussion, CMS ruled that gifts that are directly tied to advancing clinical goals, such as free blood-pressure monitors, would be OK, but that gifts such as jewelry and baseball tickets would be out of bounds.1

Ms. Jackman says patients need to be energized with a commitment to their community ACO. “Sure [being able to leave the ACO] can hurt savings, and it’s a concern,” she says. “But people are looking more to see how we can engage patients to want to stay in the community, and to think that maybe they don’t need to go to the super-specialized hospital for their knee replacement, for instance. Maybe they can stay in the community and have it done if we can give them data on it.”

 • ACOs will violate anti-trust laws. Some critics of commercial ACOs have said that one ACO can eventually get big enough in an area that payors have no other options of provider networks, allowing the ACO to actually increase the cost of care. “I’ve heard this complaint coming from health plans,” says Ms. Jackman. “Every competitive area is unique. Certainly, here in California, I know the state attorney general is looking at some recent mergers and consolidations in hospital systems that have created foundations with physician groups, thinking that it may be driving up the cost of care. And, in some cases, it may have been true. But there is so much oversight already that it would seem very difficult to create a situation where you’d truly have a monopoly.” To address anti-trust fears, the Department of Justice and the Federal Trade Commission have begun offering voluntary anti-trust evaluations to prospective ACOs to help address potential problems.

 • Previous ACO projects didn’t generate significant savings. In 2005, CMS started an ACO project called the Physician Group Practice Demonstration that involved 10 multispecialty groups. Reviewers applauded the PGP’s quality scores; in 2009 all of them achieved at least 29 of 32 preset quality goals. In 2010, seven of them reached the benchmark levels on all the quality measures, with the rest doing so on at least 30. They also increased quality scores on diabetes, heart failure and cancer screening measures by at least nine percentage points over the five years.2 Unfortunately, in terms of the other main ACO goal, saving money, the demonstration didn’t fare as well. Only two PGPs surpassed the 2-percent savings threshold in the first year, and only five could do so at three years. And the savings varied greatly: One clinic earned half the total savings, three others earned about 15 percent and a fifth earned just 5 percent.2

In 2009, Blue Cross and Blue Shield of Massachusetts performed another ACO-like experiment. In it, BCBS entered into agreements called Alternative Quality Contracts with 11 provider organizations, in which the caregivers agreed to receive a global payment for the projected cost of care for its patient population for the year. This contract put providers at risk for excessive spending and rewarded them with bonuses if they reached certain quality benchmarks.

In an analysis of the Massachusetts project, health-care researchers found that the quality of care provided by the AQC providers did improve when compared to a control group of organizations. Specifically, chronic care management, adult preventive care and pediatric care all improved.3 When it came to savings, the AQC yielded savings of 1.9 percent in 2009 and 3.3 percent in 2010. However, the analysts point out that the spending numbers don’t include the quality bonuses or end-of-year budget reconciliation. “In year one,” the researchers write, “total Blue Cross Blue Shield payouts to groups in the contract probably exceeded savings under the global budget. In year two, savings achieved by the intervention group were generally larger than surplus payments received. However, total payments to groups from BCBS of Massachusetts, including surplus sharing, quality bonuses and infrastructure support, probably exceeded the savings achieved by most groups that year.”3

“That’s the most solid criticism [of the ACO model,” says Mercy’s Dr. Tauber. “That any model that was tried didn’t work well. However, in those other programs, I don’t think they thought it out as well. I don’t think the EMR technology, when it first came out, was as good as it is now. Other models are working out pretty impressively since the Massachusetts model went through, though. For example, major employers such as Lowe’s are finding several clinics across the country that own the expertise of cardiovascular disease. If a Lowe’s employee in Louisiana needs cardiac surgery, for instance, Lowe’s will have him flown to the Cleveland Clinic. That actually represents a cost savings for them versus the wildly different price structures that we have in the United States. There, the Cleveland Clinic will have a panel that will approach the patient’s preop and postop care as a team focused on ethical care and cost savings; they won’t simply operate because he showed up with a note. It’s a fascinating model.”

Dr. Fisher says there were aspects of the Massachusetts experiment and the CMS demonstration project that either mitigate the disappointing savings results or show how they’re more positive than they seem. “I think it’s important to recognize who the population covered by the Alternative Quality Contract is,” he says. “In Massachusetts, it was an under-65 commercial population that was remarkably healthy. Spending is highly concentrated on sick patients. It’s really hard to save a lot of money in a commercial population because there just aren’t that many sick people.

“In the CMS PGP demo, on the other hand, as our recent paper in JAMA from September 2012 showed, on average the savings were modest,” Dr. Fisher continues. “They were only $114 per person, the most conservative estimate that we could come up with. However, there were dramatic savings among the dual-eligibles— people eligible for both Medicare and Medicaid—which amounted to $532 per person. So, there were substantial savings in the sick.”4 

ACOs and Ophthalmology

Now that the ACO inception period is under way, executives and physicians alike are gaining experience with the new model and coming to grips with the challenges of making it work for primary care and for specialists such as ophthalmologists. 

"There may be additional payments [to physicians] for other services...the downside is there may be fewer procedures. ACOs will want to be quite thoughtful about trying to reduce avoidable, non-beneficial care."
—Elliott Fisher, MD
Ms. Jackman’s Monarch HealthCare ACO is rooted in primary care but involves specialists as well. “Specialists are included in our Medicare ACO, but not ophthalmology yet,” says Ms. Jackman. “We’ve got cardiology, endocrinology, pulmonology and a few gastroenterologists. Those are the specialists in our Medicare ACO through whom patients can actually be aligned to the ACO. Medicare considers those to be primary specialties, and ophthalmology isn’t one of them.” Ms. Jackman alludes to one of the reasons ophthalmologists may have some breathing room during ACO development: To be viable, either as a Medicare ACO or even a commercial one, an ACO needs patients. And since ophthalmology isn’t one of the specialties that contributes patients to an ACO’s ranks, it’s faint on the radar screen at the moment. Also, ophthalmology is one of the more efficient specialties, which also makes it a low-priority target. “Doing an analysis of our claims data, we’re targeting what we think are the high opportunity areas, where we see a lot of waste in the system or a lot of opportunity to improve the coordination of care so patients don’t fall through the cracks. So, ophthalmology isn’t going to rank high on that list right now, because they’re not the ones helping manage the sickest of patients.”

While Monarch looks at ways to get other specialists involved, Dr. Tauber’s 16-doctor eye-care practice is about to begin practicing as part of the ACO, and he provides a taste of what things will be like. “We still use a reward system that most ophthalmologists understand: You eat what you kill,” he says. “That’s how we get paid in clinic, but we’re also employees of Mercy. We don’t yet know what the quality markers will be. Are we participating in treating a disease entity in which we must show a reduction in diabetic eye disease? Or are we to show an improvement in diagnosis? Those points are still being negotiated. I don’t think we as ophthalmologists are going to be touched by it as soon as primary-care physicians are with regard to such markers as overall diabetes, overall hypertension, etc. I think we’re probably going to be presented with, ‘Here is the number of patients we anticipate having at Mercy. If you see these patients and effectively and ethically treat them, and we can show a reduction in or meet some sort of marker that hasn’t been elucidated yet, we will save costs. And that savings will be refunded to Mercy, which will be translated to a bonus to the physician.’ ” 

As an example of how ophthalmologists can make an impact in an ACO-type environment, Dr. Tauber recounts a story that started two years ago. “Mercy came to us and asked how we were going to improve the percentage of patients getting diabetic eye exams, which then sat at 60 percent,” he recalls. “I said that I thought we should get non-mydriatic cameras and have primary-care doctors use them. Doing this, we picked up 800 eyes who had never had eye exams, and 20 percent of them had disease. And the disease wasn’t all diabetic retinopathy, it was also AMD, glaucoma and cardiovascular disease. We actually found some significant issues. Also, some patients ended up being sent to ophthalmologists, so we generated income that wasn’t there before.” 

Health-care experts say that by being efficient, low-priority targets to the ACO movement, ophthalmologists have more options when it comes to dealing with ACOs. “One way ophthalmologists can participate is to have discussions with ACOs that are developing and see how they can both maintain their independence while also providing services to the system and gaining access to the ACO’s patients,” says Dr. Tauber. “Systems are going to need to meet requirements. Though there are systems like Mercy that will have an in-house ophthalmology department, many ACOs will not. You might be able to participate as an independent contractor while meeting the ACO’s requirements. The other option is we put new ophthalmologists fresh out of training into the ACO model for a couple of years. We could forgive their school loans in exchange for a commitment to a number of years, and pay them a respectable income, too. It would almost be like the Peace Corps of medical care. We have to get creative.”

Though Dr. Fisher points out that there’s always the risk of an ACO contracting with one ophthalmology group, or a small number of groups, in a community, and thereby taking away a major source of referrals for the other, non-ACO ophthalmologists, Dr. Kamenetzky thinks there are enough question marks surrounding ACOs to give ophthalmologists time to maneuver. “Insurance companies and large health-care networks have the luxury of making mistakes because they have the capital to correct for them,” he says. “Plus, they’ve made mistakes before so they know every decision they make isn’t going to be right, so they approach it in a way that says, ‘We think we need to do this because this is the direction health care is headed.’ I think individual ophthalmologists can say, ‘Let’s see for sure that this is the direction health care is headed before we make the jump, because we might not get the best deal we can get, or might mistakenly team up with an organization that doesn’t know what it’s doing, though it looks good on paper.’ There is some time here for analysis. I really believe that.”  REVIEW

1. Medicare Program; Medicare Shared Savings Program: Accountable Care Organizations, Final Rule. Federal Register 76:67801 (2 November 2011) p. 67801-67990.
2. Wilensky GR. Lessons from the physician group practice demonstration—A sobering reflection. N Eng J Med 2011;365:1659-1661.
3. Song Z, Safran D, Landon B, et al. The ‘alternative quality contract,’ based on a global budget, lowered medical spending and improved quality. Health Affairs 2012;31:8:1-11.
4. Colla C, Wennberg D, Meara E, et al. Spending differences associated with the medicare physician group practice demonstration. JAMA 2012;308:10:1015-1023.