The Great ACO Experiment

Compiled by Ken Congdon, Editor In Chief, Health IT Outcomes

Are ACOs the answer? Five Medicare Shared Savings Program ACOs share their perspectives on the positives, negatives, progress, and long-term sustainability of the program.

There are a lot of experts out there, including famed Harvard Business School professor, Clayton Christensen, banking on the inevitable failure of the ACO (Accountable Care Organization) model. These pundits say the model won’t be able to change physician behavior or save money on a grand scale in the long run. But what do existing ACOs think? These organizations are on the frontlines of this initiative. They are living the change and arguably better positioned than anyone to make a determination as to whether or not the model has promise.

While there are well over 5,000 hospitals registered in the United States, there are only between 200 and 300 ACOs currently formed. These ACOs come in a variety of flavors, widely characterized by their structure, the payer, and the level of risk the ACO accepts. However, the biggest ACO movement to date was catalyzed by CMS — The Medicare Shared Savings Program (MSSP). I had the opportunity to speak with the leaders of five MSSP ACOs (all from the July 2012 cohort) to get their insights on the program. Do they think it will fail? Maybe — but that doesn’t necessarily mean the initiative won’t achieve its ultimate objective.

Q: What progress have you made to date with the MSSP?

Dr. Slonim: Basically, we’ve just been getting ourselves organized. We spent a good part of the first year putting a team together. I had to hire people to fill various roles within the ACO (e.g. finance, human resources, clinical care coordinators, etc.). Then, we focused on collecting the quality data required by CMS. There were 23 quality measures that we were responsible for submitting patient data through CMS’ GPRO (Group Practice Reporting Option) tool. Now, we’re waiting for CMS to provide us with the benchmarks from all the collected ACO data. We can then run this information through our analytics engine to understand the baseline performance of our patient population and start making process improvement decisions based on this information.

Dr. Rastogi: We have been working diligently to build the IT infrastructure required to harvest the data requested by CMS, convert that data into the required format, and transmit this data to them. These steps required not only enhancing our EMR interoperability, but also adopting data analytics technology to identify high-risk patients and other trends in our patient population.

Dr. Newman: Last month, we submitted (on time) a report on all the quality measures required of us by CMS. This effort will just provide us with our baseline data. The earliest we could potentially see any potential shared savings bonus from this activity would be the second half of 2014. We’ve also been busy developing embedded case management and palliative care programs that we did not previously have and hiring the necessary employees. Finally, we’ve invested quite heavily in developing our analytics capabilities.

Abdella: We submitted our GPRO data to CMS on the 23 quality measures they requested. Basically, CMS provides a random sample of beneficiaries that they believe are assigned to you, and the ACO goes on an Easter egg hunt to find the bits of information they want to see on those patients. We also developed a clinical integration committee that includes several specialists in the area of diabetes and heart disease. This is our designated focus area by CMS because these are the areas where a lot of our claims come from. This clinical integration committee is responsible for engaging with these patients and ensuring the preventative measures and treatment plans are being followed that will reduce the likelihood of hospital admissions.

Q: What has your biggest challenge been in participating in the MSSP?

Dr. Rastogi: It’s been the incredible amount of time and manpower necessary to not only put the technology infrastructure in place, but also to administer compliance with CMS’ demands and deadlines.

Abdella: One of the challenges is the fact that the patient has all the choice right now. Patients can choose to be treated by any provider they desire. Engaging patients in potentially different referral patterns is a part of the equation that I don’t think anybody is talking about right now.

Interoperability has also been a challenge. For example, I’m dealing with multiple EMR systems currently, and four of them don’t export data. We had to operate parallel systems and run separate reports to get all the data for CMS located in one place. IT systems need to be more open. I’m hearing that progress will likely be made in this area in Meaningful Use Stage 3, but the MSSP pilot will likely be complete between now and then.

Q: What’s next on your ACO road map?

Dr. Slonim: The very next step is to gain an understanding of our patient population once we get the collective data back from CMS. We need to understand which patients are the most expensive to care for and which patients have had quality of care issues. Specifically, we’ll be looking for ways to reduce ER visits and repeated hospitalizations.

Dr. Lundquist: First, we need to get our analytics platform up and running. We are about seven months into an 18-month deployment of a fully-functioning system that will allow us to analyze the patient cost, quality, and satisfaction data we collect. This is essential because we have to be able to prove to CMS that we’ve made improvements from both a cost and quality perspective.

The second step is building standard care models across the provider communities within our ACO. Our ACO consists of many providers — some owned by a health system, while others are not. We need to ensure we establish and enforce a standard way for collecting and measuring data across the ACO.

Third, we need to deploy care coordinators to really start changing the way we deliver care. These care coordinators will reach out to patient populations with chronic conditions and strive to ensure preventative services are maximized to keep these individuals out of the hospital.

Abdella: A main focus for us is determining how to handle a patient’s transition from the hospital or skilled nursing facility back to primary care. How are we going to wrap that person in their patient-centered medical home? We think the biggest savings and quality improvements are going to come from identifying patients with complex illnesses and putting together detailed care plans for them.

Q: Have you seen any positive results from your involvement in the MSSP to date?

Dr. Rastogi: According to the data CMS has provided us, we’ve already improved on our costs. We simply did some rudimentary changes (mostly to workflow and documentation) that have made an immediate impact. This is encouraging. Also, an unanticipated benefit has been in patient satisfaction. One of the jobs of our newly hired care coordinators is to contact patients within 24 to 48 hours of their discharge from the hospital to see how they are doing, confirming they are adhering to discharge instructions, and to schedule their follow-up doctor’s appointment. The positive response we’ve received from patients has been unbelievable.

Dr. Newman: Our preliminary data from CMS showed us that we had the most potential for shortterm savings in reducing unnecessary hospitalizations and ER visits. Thus far, we’ve been successful in doing that by intercepting ambulatory sensitive conditions and getting these patients treated on an outpatient basis. We have also seen some improvement in some core quality measures such as diabetes care, and some preventative measures such as cancer screens, colonoscopies, and mammographies.

Abdella: We found the data we collected for CMS to be very informative. For example, we identified a lot of missing documentation, which clued us into opportunities for immediate process and workflow improvements.

Q: Are there any changes you would suggest to the MSSP?

Dr. Rastogi: I think it would have been easier from an organizational standpoint if CMS would have notified us as to when we were going to be an effective ACO earlier. They notified us of our July designation at the beginning of the month, so the clock started ticking almost immediately. We had little time to prepare or ramp up, which made the process more challenging.

Dr. Lundquist: I do not believe that health providers, especially primary care physicians, need to be put at any risk under a payment model that supports an ACO. If an ACO is successful, preventable ER visits and hospitalizations are eliminated because of better care delivery. A primary care physician shouldn’t have to take a pay cut in years one, two, three, or four, in order to get there.

Dr. Newman: There are some uncertainties that exist with CMS’ current attribution model. Basically, CMS provides you with a list at the beginning of the program based on the patients in your population from the previous year. You really don’t know which patients you are responsible for until the end of the performance period. It would be helpful if these populations were more clearly defined up front. Furthermore, since the data is historical as opposed to real-time, you can identify where problems existed. It does not help you manage patient conditions contemporaneously.

Q: Why do you think there is so much negativity surrounding the ACO concept?

Dr. Slonim: I think a lot of people expect ACOs to be a plug-and-play solution, but they’re not. This is a learning process that will probably take five years or so to come to fruition. Right now, many of us are learning as we go. There’s a segment of the healthcare population that is really nervous that there isn’t a consistent plan of action. The good thing is CMS hasn’t over-engineered the MSSP. All they’ve done is given us metrics. They haven’t told us how to get to the end goal. With more than 200 ACOs containing some of the brightest minds in healthcare working on this, I’m confident at least a handful of us will come up with something that works and be able to share more broadly with others.

Dr. Rastogi: I think a lot of folks relate the ACO concept to the capitation that occurred decades ago, which basically focused on rationing or denying care. ACOs are getting a bad rap by association here. However, I have gotten to know the goals of what we are doing and how it’s different than what’s been tried before. ACOs do not ration care in any way at this point. They are really just focused on trying to take the waste, duplication, and inefficiencies out of the system.

Dr. Lundquist: The healthcare industry in the U.S. has been built on the fee-for-service model for decades, and providers have become very successful at earning money under this model. These providers learned how to play the game well, which is why you see escalating costs. ACOs will upset the way providers get paid, and health systems are nervous about changing the transparency of their fee schedules.

Q: Do you think the MSSP ACO model is viable?

Dr. Lunquist: We believe wholeheartedly that ACOs are the future of healthcare. I think it’s hard to argue with the tenets of the patient-centered medical home, and that is the underpinning of a good ACO. So, ensuring primary caregivers and specialists buy into the concept of bundled payments and total cost of care is essential.

The end game is that the ACO model will work to manage populations differently and better. It will challenge us to look for the best patient encounter that might be in a physician’s office, but it could just as easily be a telemedicine encounter, an e-visit, or an exam at a retail clinic.

Abdella: I don’t think the current MSSP model is sustainable. In our rural community, in particular, there is only so much cost we can take out of the system before we start to hurt ourselves. We don’t have enough infrastructure or resources to continue to cut costs without compromising service. There’s only so much blood you’re going to get from that stone. You can’t continue to share savings, but I do think an upside gain share with providers has potential merit.

Dr. Rastogi: I think the ACO concept has merit, but I don’t know if it will survive in the exact form that it is today. However, whether or not the ACO as it currently is defined comes to fruition is inconsequential. Ultimately, what is important is the fact that we need to bring accountability to the provider side of healthcare where physicians can access patient data, effect change (to both care quality and cost), and share in the rewards of improving these outcomes.

Dr. Slonim: I’m not sure if the MSSP, or ACOs as a concept for that matter, will be around for more than three or four years. Ultimately, I believe the program will lead us to the next stage of improving the cost and quality of care conundrum. It’s likely that most MSSP ACOs will “fail.” I think they’re even expected to fall short in quality and financial targets. What’s important is what we learn from those that succeed, and try to figure out ways to generalize the findings to others. What’s very clear is we need to make sure we are documenting and improving quality as an industry while documenting and reducing expense. Whether that ultimately takes the shape and form of ACOs as they currently exist remains to be seen.