At New Jersey’s Summit Medical Group, Data-Driven Breakthroughs in Population Health

A number of speakers had been scheduled to present on subjects around value-based contracting and population health at HIMSS20, ahead of its cancellation on February 5.

Among the scheduled speakers were Jamie L. Reedy, M.D. and Ashish D. Parikh, M.D. Dr. Reedy is chief population health officer and a family practitioner, at the Berkeley Heights, N.J.-based Summit Medical Group; Dr. Parikh, an internal medicine physician, is senior vice president of medical affairs and quality at Summit Medical Group, whose 800-plus clinicians serve patients in 80 locations across seven counties in northern New Jersey.

Drs. Reedy and Parikh spoke recently with Healthcare Innovation Editor-in-Chief Mark Hagland regarding the presentation they had been scheduled to give, “Using Analytics to drive Patient-Centered Post-Acute Care,” with a focus on how Summit’s clinicians have achieved success by implementing technology to support care coordination and monitor performance across care settings in order to reduce their year-over-year length of stay metrics for skilled nursing facilities. The interview took place before the announcement of the conference’s cancellation. Below are excerpts from the interview.

What are some of the core points of your presentation?

Jamie L. Reedy, M.D.: Reedy: We have been leveraging data for many years. The premise of this presentation is that Summit was really a leader in NJ and the NE in terms of getting into commercial risk contracts in the 2011-2012 time period, before we even got into risk-based contracts with Medicare; we negotiated a number of risk-based contracts then with private payers. We were very innovative in this region in that way. The strategy was led by Jeff LeBenger [Jeffrey LeBenger, M.D.] as CEO, but Ashish and I helped lead the initiative.

What have been some of the key learnings around leveraging data in this area?

There have been a number of key learnings. Timely data exchange between health plans and provider groups is really important; we can’t manage risk if we don’t have the same type of data that our partner has. Arcadia [the Burlington, Mass.-based Arcadia software company] has been an excellent partner, and has helped us to work well; what’s needed are a full scope of data, timeliness, integration and matching of data, to create actionable reports.

Ashish D. Parikh, M.D.: And understanding how commercial risk and Medicare and Medicaid risk differ, and getting the data around the specific population, are important as well. And doing so means not only aggregating data, but also being able to separate it out and being able to translate it back to the clinical side. The clinicians don’t really care about different plans, they want usable data.

The populations covered by public and private health insurers are different, and the contracts are different as well, correct?

Yes, the populations are different, and also, you may have flexibility with commercial payers that you wouldn’t have with Medicare and Medicaid. We call it ‘you-do-we-do.’

Reedy: Yes, and we apply that model to data-sharing, care management, pharmacy management; we’re constantly working with our partner plans to determine which tasks and areas to take on. And as Ashish says, the commercial plans have more flexibility.

What kinds of things have you learned, because of that contracting flexibility?

Reedy: It’s driven by the fact that the populations are different. When you think about the Medicare population, you’re focusing on keeping patients out of the hospital. With commercial payers, you’re working on preventing out-of-network leakage, avoiding avoidable utilization like imaging and lab utilization, as well as that of biologics, infusions, and injectables. So we focus on different areas.

You identify high-utilizers among patient populations?

Certainly, we’re applying care management to patients with chronic illnesses. But with out-of-network leakage, we use our claims data from Arcadia to analyze where patients are going outside of our network as well as outside the health plans’ networks. And where do we need to do a bigger job of partnering? Physical therapy is a good example. We can’t build enough of physical therapy bench, internally. So we need to partner with high-value physical therapy resources. And expensive infusions—where the cost can be three or four times as much in high-cost centers; getting them to a lower-cost, but high-quality partner patient care organizations. And yes, per SNFs [skilled nursing facilities], our March 11 presentation will focus strongly on that area. And Arcadia has worked with us on determining which SNFs are more or less likely to send patients back to the ED.

What have been the biggest challenges around the leveraging of data?

Anytime you’re integrating disparate data, you’ll face process challenges. You’d like ideally to sign a contract and have the data flow instantly, but that doesn’t happen. It takes time to build the connections, and then learn how to use the data meaningfully. But it’s incredibly exciting and meaningful now that we can capitalize on that combined data set, and then take the reporting back to the physicians, to allow them to apply it to their care management of patients. That’s when the magic happens.

Parikh: For example, if we see a spike in ED visits on a Saturday morning, we can see that we need to provide better access to primary care or urgent care on weekends. Or seeing that behavioral health needs a boost, so using the data to justify investing more into behavioral health. And design can impact how well you can control or manage a population, so a self-insured employer can have very rich out-of-network benefits, which doesn’t help us manage those networks well.

What have your biggest learnings been in the Medicare ACO?

We’ve been in the Next Generation ACO program since 2016, we’re now in Year 5. The most successful example there is really related to our skilled nursing facility work. When we first started in Next Gen, we were at 2,000 SNF days per 1,000; we’ve come down now to around 1,300 to 1,400. That really dramatic decrease has been driven by analyzing the data around discharge-to-home rates and discharge-to-SNF rates. It’s allowed our clinical care team to make changes; we couldn’t have done that without data. That’s been the biggest driver to our ability to advance in the NextGen ACO.

Can you drill down on your learnings around discharge rates in those areas?

If you look at the hospital environment where we employ our own hospitalists, several years ago, our discharge-to-home rate was way too low. So our service line chief, Dr. Ahmed, changed how we talk to our patients. They stopped ordering physical therapy consults on every patient admitted, because every time you ask a physical therapist, they’ll say, sure, you need to send them to a SNF. So if you started the physical therapy during the inpatient stay and prepare the family for discharge to home. If you’re going to reduce discharge-to-SNF, you have to not admit to SNFs or you have to reduce lengths of stay in SNFs. And a lot of it is trial and error, it’s figuring out where the roadblocks are, what’s tripping us up. So a lot of it was trial and error, we learned.

Parikh: One of the simple actions that Dr. Amina Ahmed, our service line chief of hospital medicine and post-acute care, implemented was that the hospitalist would walk the patient every day, and they could figure out who needed discharge to a SNF or to home. And that one step changed so much.

What key advice would you offer to those who might pursue this path?

Reedy: In the presentation, we’re going to talk about building a game plan, the step-by-step approach. If you’re going to negotiate a successful contract and create a really robust collaboration, you have to build a proposed timeline. We also have checklists that we’ve developed around what kinds of data you need in order to successfully negotiate a contract with a health plan. And someone once said that results move at the speed of relationships. And Dr. LeBenger and Dr. Ahmed and Dr. Parikh and myself have spent a lot of time really developing growth strategies together. And that’s been a critical success factor.

Parikh: And the data that you get before you even go into these contracts is the key to figuring out where your opportunities are, and it helps you to negotiate your contracts, and where you’re at risk.

Do you have any thoughts around the data and IT infrastructure needed to accomplish this?

Reedy: It’s really critical that you select a vendor partner that has the ability to integrate data from all sources—flat files, HIE [health information exchange], clinical and claims data, Blue Button: we absolutely need visibility into the patients. So they need to be able to bring that data together and make it meaningful inside the platform.