A Clinically Integrated Network Forges Success in Southeastern New Jersey

March 26, 2018
In New Jersey, Shore Quality Partners is succeeding as a clinically integrated network, through clear strategy and excellent execution—and successful leveraging of data and IT to support clinical transformation

As the U.S. healthcare system shifts from volume to value, provider leaders are finding that there are many different ways to advance, and that in particular, there are many different ways to participate in ACO (accountable care organization) arrangements; but that one irreducible critical success factor is physician engagement.

That certainly has been a core awareness at Shore Quality Partners CIN, a clinically integrated network based in Somers Point, New Jersey, just south of Atlantic City. Formed in March 2014, Shore Quality Partners is owned by Shore Medical Center, a freestanding community hospital, and encompasses about 240 primary care physicians and about 190 specialists, 65 of whom are employed by the hospital, with the rest in independent practices.

Shore Quality Partners was formed as a network of providers aligned to support quality improvement and cost efficiency. The CIN is governed by a board of directors and several physician committees, who are responsible for developing and implementing clinical integration initiatives and dealing with all matters affecting the organization. The CIN’s board consists of a total of 12 directors, of whom nine are physician directors (a mix of primary care and specialist physicians), and three hospital directors.

As articulated on its website, “Shore Quality Partners brings independent, contracted and employed physicians – whether primary care, specialist or hospital-based-physicians, together with Shore Medical Center to form a network of providers aligned to support quality improvement and cost efficiency. The Clinically Integrated Network (CIN) is governed by a Board of Directors and several physician committees, who are responsible for developing and implementing clinical integration initiatives and dealing with all matters affecting the organization. The CIN performs various functions and activities: developing and managing arrangements with payers for payment of bonuses, care management fees and other incentive compensation; developing and implementing care guidelines, policies, procedures and protocols; gathering and analyzing data; designing and implementing care management initiatives; aligning with additional providers; managing IT infrastructure implementation and operations; and measuring performance and distributing incentive funds based on performance.”

With regard to the CIN’s founding, the organization’s website also notes that, “In collaboration with its Physician Leadership Council, SMC [Shore Medical Center] identified the need to respond to the changing market while allowing community physicians to retain their practice independence and autonomy. Shore Quality Partners emerged as a dynamic model designed to engage and align interests among community providers and SMC. This effort brings together physicians and SMC to enhance collaboration and achieve clinical integration with the support of Shore Medical Center’s resources and infrastructure. By proactively seeking opportunities with interested payers, Shore Quality Partners positions its participants for success under new payment and care models.”

In terms of the data analytics and population health needs involved, Shore Quality Partners has been partnering with the Piscataway Township, N.J.-based HealthEC, in order to achieve success in leveraging data analytics to support its intensive clinical transformation work.

Since the outset, Cliff Frank has been Shore Quality Partners’ interim executive director. Frank spoke recently with Healthcare Informatics Editor-in-Chief Mark Hagland about his organization’s push into value-based healthcare. Below are excerpts from that interview.

Tell me about the origin of the impetus for formation of the clinically integrated network?

At the end of 2012 and in early 2013, we realized that we didn’t have enough money to employ our way to a big medical staff; it just costs too much to employ physicians and cover their overhead. So we knew we needed some other vehicle. And a lot of medical staff didn’t want to be salaried, and in any case, we didn’t have the money. Same canoe.

Are you an MSSP?

We are part of an MSSP [Medicare Shared Savings Program], Track One. We have a direct shared savings contract, commercial, with Horizon Blue Cross, and a separate one with MJ Health for Medicaid; and we have a pay-for-quality contract with a Medicare Advantage program, Clover, which up in northern New Jersey and active here. They’re owned by the same organization up in northern New Jersey.

What is the name of the MSSP?

Allegiance ACO, and they’re based in Trenton. They have about 5,000 patients over there, and we have about 75,000 patients over here.

And what is the size of your commercial ACO population altogether?

About another 30,000 lives.

When did you join Allegiance ACO? And when did your commercial contracts begin?

Commercial ACO started about April 2014. The Medicaid ACO with M Health started in January 2016. And the Medicare ACO started in January 2017.

So you’re in the second year with the MSSP?

Yes, that’s correct.

What are you hoping to achieve overall?

Well, the first thing is, we’d like to earn a bonus for our doctors; and that’s easier said than done, believe me. But we’re certainly about improving the quality of care for the patients, and we’ve got a good track record of having done that on any number of metrics, and earning a bonus. And how we get there is important. One of the things that ACOs across the country have, is a lot of care that leaks out of the network and goes elsewhere. We have a lot of that—60-70 percent leakage. And as a hospital, we only have 30 percent market share in our country, so you’d expect leakage. And we don’t do open-heart and certain other procedures. And we’ll never eliminate leakage, never; but if we could go from 70 percent to 50 percent, that would be wonderful.

How have you been able to get physician buy-in, and what have the results been?

Our experience has been that, once they trust us, it’s not so hard. Getting them to trust us was really hard. And it wasn’t so much an issue of trust per se, as it was, getting us to fit into their workflow. Yes, you’re right, they did not want us messing with their workflow. For example, for nine months, I had a social worker working with us, and the phone wasn’t ringing. And I said, what’s the matter? Give us your most difficult cases. And nothing happened. Well, it turns out, making a phone referral is not how they do business. As soon as we invented a fax-based referral form, boom, 50 referrals a month. In other words, so much of this is not about any intellectual or emotional resistance, but about finding the workflow-based path that fits their patience; and as soon as we do that, boom! They’re with us. We help simplify their quality metrics—what’s in their nominator? What’s in their denominator? Helping them identify patients who should be de-attributed. Otherwise, it’s overwhelming. And Medicare, Medicaid and commercial, we will only talk about one of the three types in a meeting, which means that we’ll be there three times. Otherwise, they get totally overwhelmed. They get overwhelmed anyway. At least this way, they can keep it a little bit straight.

And what have some of the results been?

One of the things we did early on was to identify our diabetics, as other organizations have done. We actually went out and found a certified diabetic counselor, and she is floating around to all the practices. And we’ve taken our diabetic spend down to half, from $3,000 to $1,500—and it’s not just because of the diabetics educator—but we’ve noticed that when we hand a list of diabetic patients to our providers, and they say, hey, you haven’t seen this person in months—they do better, they refer patients to the diabetes educator, etc., and that has a financial benefit as well.

So a lot of this is process?

It’s all process. And sometimes, you can link an outcome or process, sometimes, it’s an article of faith. But either way, if you have patients who are high-risk who aren’t seeing doctors, intuitively, you know it’s a bad situation about to get worse. So, fix it, see the patient. OK!

Tell me a bit about physician participation in governance? You have physicians on your board?

We do. We have a 12-person board, with nine doctors and three hospital administrators. Everyone has one vote. It takes two votes to do anything. So immediately, the doctors realized they had veto power, and they loved it. And the hospital was perfectly willing to try it, because, without it, there was no physician engagement, and no evidence of trust. And we’ve never had a divided vote, never; it’s my job to make sure we don’t. But just knowing that they had the power, was very helpful. Second, we try not to “meeting them” to death. We did at the beginning, we were a little bit naïve. Originally, we had four committees—now, we have just one, which covers care management, quality, and data, and that operations committee meets once a month, and the board meets every other month. The other thing is, with 33 primary care physicians, we can ramp up to something quickly. That’s where small works. If we were in central or northern New Jersey, with 300 primaries, this would not work. And the fact that I and my staff are out in the doctors’ offices a lot, means that we can address misunderstandings before they spread.

In terms of misunderstandings, those can be like the classic phone game or grapevine process that gets out of control, correct?

Yes, that’s right, and by being close, we find things out, and we’ve established a track record, so people will at least bring and issue to us and wait for our response, before they wind up, based on mistaken information.

What has been the hardest thing in all this?

The hardest thing has been getting consistent behavior and messages from the payers that are consistent from moment to moment. They zig, they zag, they drop the ball, they send us crap for data, and then change the formula. So the hardest thing is maintaining trust among all the parties. Because when the payers do that stuff, and they may have very good reasons for doing that—they change computer systems, quality metrics, or whatever—but change is not something that providers are happy with, when they don’t understand the whys, the hows and the wherefores.

And the physicians are not going to sit down long enough to hear a paragraph-long explanation. So that erodes trust between us and the payers, and can then erode trust between the doctors and the administrative arm. We’re suddenly coming in, looking for different information in their EMR; or they could submit a manual code and now they have to submit G-codes; every month, and it makes them crazy.

What’s been most successful, what’s worked best, in all this?

Giving doctors data that is actionable. I’ve made plenty of mistakes handing doctors a list of patients and a stack of claims and saying, here’s what’s going on with your patients. And they say, what the hell is this? So I’ve gotten more focused, and that comes from having a data system that works. And having something like HealthEC that has the data and inferential statistics around what’s going on with the patients.

I’ll give you a couple of examples: one is amazingly simple, but so important. Out of HealthEC, for every patient, we get a risk score, both a retrospective and a prospective risk score. What we want to know is, what’s the risk of something bad happening to that patient? We run the numbers, and we’ll look at the patients with the high risk scores. And we’ll ask who they are. A bunch, 1,500, will be seen for cancer, chronic kidney disease, COPD, whatever—OK, those are being managed by specialists. Then we get to 350 who are patients who are high-risk who have not been seen by their PCPs in more than six months. And we tell the PCPs, it’s probably a good idea to see these patients. And that is an incredibly effective intervention to make.

Another example: we give the primaries those lists, and they see the patients. Another example of how we use the data—one of the things about data is that it can be used as a weapon, and it’s really important that we not do that, because that’s how you destroy trust. So in order for data to be useful, it has to have a point; it has to have the right numerator and the right denominator. So one of the things that’s really, really hard is that when the payers send me data, they organize that data based on whom they paid, like Shore Physicians Group—well, that’s an amalgamation of 20 PCPs and 45 specialists. So when I run data on Shore Physicians Group, it’s all a big ball. What I get from HealthEC is totally different; they break out the data down to the MPI, the individual provider ID, level. So they can take a tax ID and break it up into the 16 different MPIs providing care, and I can say, this MPI is doing great, and that MPI is spending like crazy. Now, you show that data to the doctor, they can use it.

What is most important in terms of managing the data and IT in order to be successful?

There are two pieces, the numerator and the denominator, and they both have to be counted correctly. For example, a patient goes to see a radiologist, and they go to the hospital for the diagnostic imaging study. The hospital generates one bill, the radiologist generates another bill, because of the professional and technical components. You need to make sure when you count that visit, that you count it as one, not two, or all your radiology statistics will be wrong. You have to have a vendor who understands that. You would be amazed by how many vendors don’t understand that a global fee and a technical/professional, are not the same. So that’s why I mean about the numerators—how much stuff got done? Measuring how much stuff got done, and who did it, and is that a leak, or something in network, and how much did it cost? All of that I worry about, yes.

But the other part is the denominator. Over how many people? Just because there were 40 hospital admissions, that doesn’t mean anything. It’s over how many people? 40 in a month might equate to 90 per thousand, and we’ve budged 20. Uh-oh. So the problem is, you have to make sure you’re counting the number of members correctly each month; so if a member terminates and loses their coverage in May, and now it’s December, is that member in your membership? The partial year should count. You ought to have four member months in your denominator while you had exposure to that member. So all of that matters, when you compare your experience to your target, or compare it to last year. Things change. Last year, hips and knees were done inpatient in hospitals, and now they’re being done outpatient in the community? Holy cow, that’s a whole different situation. So, understanding the nooks and crannies of the data, is important, and it’s really hard, if you’re just trying to learn that from scratch, and worse, if your system is just learning it from scratch, because your systems will mislead you.

What would you say to CIOs and CMIOs, as they prepare to develop clinically integrated networks, and participate in ACOs and population health management?

I’ll answer it this way: managers and executives go from the general to the specific, because they get to pick the level of detail they want to get involved with. That’s not how physicians think; they go from the specific to the general. If you’re going from the general to the specific, you’re asking physicians to trust you from the beginning, which they won’t. Starting from the specific, painful as it is, is worth the investment.

What will happen in the next couple of years, for your organization, and for the industry?

I’ll start with us. We will gradually continue to get better. It won’t be in huge leaps, but rather,  inch by inch. Achieving shared savings will get harder and harder. Even if you only periodically re-base, you’ll eventually get to the point where there’s no more savings to get. But it gets harder and harder after the low-hanging fruit is gone. In a broader sense, there is a collision coming between those groups that do well in shared savings and managing care, and payers who do not want to let go of the risk because they don’t want to let go of the reward. And there are many groups out there in Next Gen ACO and other arrangements that would love to step up and do a Medicare Advantage deal, but can’t get payers to step up and work with them. So many providers will convert to risk plans, because payers won’t let go for the risk. Fifty percent, when someone else is setting the benchmarks, is not enough of an incentive.