The decision to participate in Medicare’s Shared Savings Program requires thought on multiple fronts. Not only must hospitals determine whether they are able to create enough value to justify enrollment, but the program offers two payment options — one of which shares both risk and reward, while the other pays smaller bonuses with no risk of loss.
Although the potential payouts are appealing and participation provides a valuable education in enacting an ACO model, the program is far from established. The success rate — particularly for the higher-risk option — is low so far. In the program’s inaugural year in 2012, four of the 114 organizations that participated opted for the riskier of the two payment options. Only two of these reaped the benefits: Rio Grande Valley Accountable Care Organization Health Providers and Heartland Regional Medical Center in St. Joseph, MO, who will be awarded $2.9 million in bonuses.
Three out of four organizations failed to earn Medicare bonuses in the first year, according to Modern Healthcare. Twenty-nine organizations (including Heartland Regional) will share $128 million in bonuses.
Despite the lukewarm rollout, the program is gaining traction. As of Jan. 1, there were 341 ACOs participating in the Medicare Shared Savings Program, including 123 new organizations that filed to participate in the program this year.
The federal Centers for Medicare and Medicaid Services accepts applications for MSSP only once a year. Any ACO wishing to participate in the program in 2015 must file a formal notice of intent to apply by May 30 and submit a complete application by July 31.
With those deadlines just around the corner, Healthcare Dive examined the self-audit that ACOs need to perform to determine if MSSP is right for their organization.
How do I find 5,000 Medicare beneficiaries?
The first thing a qualifying ACO needs to assess is the scope of its panel. The price of admission to MSSP is 5,000 fee-for-service beneficiaries assigned to the ACO. Without that critical mass, an ACO is ineligible. An organization needs to ensure that it has a sufficient number of primary care physicians participating in its network to draw the necessary volume of Medicare Part A and Part B beneficiaries to meet the mark.
PYA principal Martie Ross advises that applicants ensure that their population is substantially larger than the 5,000 benchmark. (PYA recently published its second edition of a white paper “road map” for applying for MSSP.)
“The rule of thumb we use is that you need to be at 7,500 in your panel because of the way the attribution rules work,” Ross said. “They are brutal.”
Am I in a position to create dramatic savings in a short period of time?
MSSP rewards increases in efficiency, not organizations that already are efficient. An organization whose processes already are good and organization already is lean is not likely to benefit from participation in the program.
But hospitals that do have room for improvement must also be in a position to invest in improved processes and infrastructure. According to Modern Healthcare, Heartland Regional invested approximately $2 million in new care management, data analytics staff and new software to support clinicians. The $2.9 bonus payout will cover that, but the hospital had to be in a position to fund those investments in the first year.
Moreover, although no formal decision has been made, there have been rumblings that CMS may eventually do away with the “upside-only” risk alternative. If that comes to fruition, the stakes will much higher if an ACO misjudges its ability to create savings:
“CMS made noises when it published the rules in 2012 that only they would only offer upside-only contracts for the first three years,” Ross said. “2015 will be the third year. Who knows how likely this is, but in 2016, will you only be able to get in the program if you accept risk?”
Do I have the data to paint an accurate picture of my population?
Creating efficiency depends in large part on an ACO’s ability to properly define its population. Total cost of care is a broad metric that provides limited actionable insight. Distribution of diagnoses and services, on the other hand, allows organizations to not only identify high-risk and low-risk patients, but also identify rising-risk patients — those patients who appear reasonably healthy now but are likely to move into the high-risk bucket within 18 months.
That data is admittedly difficult to find. Jeff Hoffman, senior partner at health care consulting firm Kurt Salmon, notes that payers are the only companies that have close to complete information on a given patient. The data available to hospitals and physicians is limited to care performed at their facilities. Worse, according to Martie Ross, the delay in ICD-10 has further set back the documentation that drives good population data.
“ICD-10 is all about how well the physician documents the patient encounter. Better documentation drives the description of the patient population you serve,” Ross said. “If you’re missing diagnosis codes because the physician never documented the information that would drive that diagnosis code, you don’t know your population.”
Ross uses diabetic patients as an example. “I may be able to describe how many patients are diabetic, but I haven’t described all the complications associated with that — that patient looks like they have one chronic condition when in fact they have five. ICD-10 drives you to capture that documentation.”
The strength of the connection between ICD-10 and population description is up for debate. According to Brad Sitler, the principal industry consultant at the SAS Center of Health Analytics and Insights, creating a population database that is predictive of patient outcomes is more dependent on leveraging nontraditional data: Socioeconomic data combined with EMR data and payer data.
“If you have a provider and payer who have a non-strained relationship and they could strike up a deal to share data, you could actually have a holistic view of a patient,” Sitler said. “You could start leveraging that data today to start building risk strata, and start looking for patients that are on the trajectory to becoming high risk. That data is very predictive.”
It's predictive, Sitler notes, of behaviors, not necessarily patient outcomes — although the hypothesis is that it does both.
Using that data for risk analysis, Sitler says, is only the jumping-off point — the information ACOs can use to identify opportunities for creating savings. The real value in leveraging that data is to change how an organization interacts with patients outside of the office. Tailoring interventions to the individuals, Sitler said, drives behavior change, which improves outcomes and creates savings.
What happens next?
MSSP has been described as “a race to the bottom.” Like radioactive half-life, the margin of savings is going to get smaller and smaller, and at some point, all the fat will be trimmed out of an organization. Once you’re done, says Ross, you’re done. The financial incentive from the program is over.
And while it seems likely that will always be some complex services (such as heart surgery and kidney transplants) that will be reimbursed on a fee-for-service basis, Hoffman believes basic care providers who have gone down the MSSP route won’t be able to survive for long with a foot in both fee-for-service and value-based care. Improved care translates to a decrease in billed services, and the provider starts to lose money. So when the savings payouts dwindle, hospitals must have a long-range plan to cover overhead and ongoing investment in needed infrastructure. MSSP, ultimately, is an opportunity to develop the tools, processes and the team to succeed in a care-based environment, but it is a short-term strategy.
The solution, according to Hoffman, is for ACOs to partner with payers on the premium level. Some organizations are selling an insurance product; others are selecting a payer and forming a risk-sharing partnership. And Hoffman says he is helping to partner a large number of health systems and medium-size payers. The payer continues to sell insurance and process claims and administration, but it shifts 50% of the management of the risk to the health systems.
“The value for the health system is: ‘I’ve invested in these tools correctly, and I’ve got physicians and partners in the medical community and we are taking the sickest members of the population and managing them better. By doing it right the first time, we can save money and we will earn dollars from the premium rather than, Every time we bill a CT we make a dollar,’” Hoffman said.