Numerous healthcare industry observers have noted that the opportunities being afforded patient care organization leaders under state Medicaid programs are becoming more and more numerous and robust. Indeed, one area that’s seen a great deal of activity in recent years is around Delivery System Reform Incentive Payment (DSRIP) programs and other Medicaid Delivery Transformation Section 1115 Waiver programs, which are working to expand Medicaid eligibility and clinical services for low-income residents in participating states. In New York state, up to $6.42 billion is allocated to the state’s DSRIP Performing Provider Systems, with payments based on achieving performance improvement results in system transformation, clinical management and population health.
On Tuesday at HIMSS18, being held at the Sands Convention Center in Las Vegas, Joseph G. Conte, executive director of the Staten Island Performing Provider System (SI PPS), one of the 25 PPS in New York, shared with his audience how he and his colleagues have been achieving New York state’s DSRIP project goals, and what they’ve learning from their experiences so far; and he was joined by his organization’s technology partner, who shared insights on the technology infrastructure involved.
“Let me talk to you a little bit about the Medicaid design project, about DSRIP,” Conte told his audience. “New York state is the eleventh state in the country involved in Medicaid redesign, following the lead of California, Massachusetts, and several other large states. $7.4 billion was the agreed-upon amount, working with CMS [the federal Centers for Medicare and Medicaid Services], to create a transformative strategy to redesign Medicaid in New York state. So for 6 million people at $60 billion per annum,” he said, describing the number of New York state residents who are Medicaid recipients, and the annual cost of their healthcare, “CMS and Medicaid have a lot at stake. CMS paid for half of the $7.4 billion.”
Conte told his audience, “This is really a disruptive time for providers, healthcare systems, and patients as well—everything we are used to, is changing over time. The DSRIP program is funded over five years, and it really is a value-based program, to the extent that money must be earned in every period, to receive distributed funds. We have about 75 partners—major hospitals, nursing homes, FQHCs, etc.” And one key focus, he noted, has been the social determinants of health. Broadly speaking, he added, 180,000 of the 500,000 residents of Staten Island are potentially affected by this work.
As for data and information technology, Conte told his audience, “Early on, we recognized that identifying a partner to create a business intelligence platform was very important, and that working with old claims data would not provide the BI we needed. And we found a great partner in SpectraMedix”—the East Windsor, N.J.-based SpectraMedix, a data analytics company that specializes in helping providers transition to value-based payment models. “And the Department of Health recognized early on that their partnership was important in helping us to transform care delivery. Bringing in EMS data, getting information from school health, from all the clinical partners in the community, is incredibly important. And when you’re in an at-risk environment, that is absolutely essential to be successful.”
“CMS and the New York Department of Health agreed on about 60 metrics” that would be used in this program to evaluate outcomes, Conte reported; and, in relation to those metrics, he said, “We’ve been extremely successful in the first several years. Among the gains: “We’ve been able to double the amount of reversible preventable ER use, including around substance abuse.”
Speaking of his company’s technology partnership with SI PPS, Raj Lakhanpal, M.D., CEO of SpectraMedix, told the audience, “In DSRIP, you have lead providers and lead partners, and other partners.” In the Staten Island initiative, Lakhanpal said, “They all started giving data, and that gave us a very robust platform.” What’s more, he said, “The dirty secret in data analytics is that 90 percent of the work involves cleaning, normalizing, and compartmentalizing data. That was very helpful. And then the analytics that were produced, and then the project management office could look at the data and say, these are my projects, these are the metrics we need to address, and to incent.”
In all that, Conte told the audience, “Hot-spotting and geo-mapping have been important. We’ve found out where there have been disconnects between services and the need for services. For example, we discovered that, while the north part of the island had nearly all of the substance abuse clinics, the patients in need of that treatment mostly lived on the south side of the island.”
One of the things that Conte and his colleagues did was to leverage EMS (emergency services/ambulance) data, to address the “frequent flyers”—individuals who literally were accessing medical care in emergency departments as many as 200 times a year. Clearly, Conte said, these people were over-using EDs relative to their levels of care need. Identifying those individuals, and connecting them with care management, was a major step. The same principle applied to mapping the high level of incidence of urgent interventions among school-aged children with asthma, relative to the availability of school-nurse care. With geo-mapping, he said, “You can identify areas that are lacking in key services. And in creating maps of the population, we can filter in on specific conditions, and if we were able to hover over a specific area within a specific map, we have three years of claims data, and can figure out utilization, including hospitalization, medications, etc. And we can filter by demographics, by types of chronic illness etc.”
“Taking the data, normalizing it, and putting into the EDW”—the enterprise-wide data warehouse—in order to do “predictive modeling and simulation around high-risk patients,” has been one key basis for advancement in the initiative, Conte said. One of the next steps will be to obtain data from the Department of Health that is timelier. “Right now, we have data that’s from May 2017,” he said, “which is not so effective in terms of [achieving] of real-time or near-real-time analytics.”
And, ultimately, Conte said, “When you think about improving care for the population, and realize that the patient encounter accounts for maybe 30 percent of those results, you realize you have to work with partners outside the purely clinical realm, to achieve success”—meaning addressing issues such as nutrition/food deserts, housing, and a wide variety of other issues.
One excellent example of that, Conte said, has been around diabetes management. “When we started looking at individuals, we were able to look at the types of services they needed, and understand when talking to individuals out there, they’re important, and providing peer education by fellow diabetics, and providing them access to free fruits and vegetables through the Single Harvest program. These are all things that, again, by looking at cold, flat, one-dimensional data, you’re never able to really raw out of that. But looking at the social determinants, at such demographic elements as country of origin, at where providers are—when you can overlay all those factors, you can have a big impact.”
Asthma management in children and teenagers was another example. “When we started looking at avoidable ED use and hospitalizations, it became very clear that adolescent use of EDs was off the charts,” Conte said. “Children didn’t have medication administration forms on file, so if they had an event, they ended up in an ambulance. And sadly, we’ve had two asthma-related deaths in our schools in the past years. So those triggers prompted us” to take action. “Now, we’re working with emergency departments, school nurses, and emergency services. We want to send people into people’s homes and figure out what the triggers are. Did they get the right advice, the right medication? And if there’s mold, peeling paint, pesticides, cats—all those things that create triggers for asthma—we have an asthma home visit program we run, and that’s made an enormous impact. The plans absolutely love it, because they’re not paying for claims. But getting them to understand that value-based plan redesign is very important. In the value-based purchasing world, this really makes a lot of sense to the health plans in the value-based world. And these individuals go out and educate them, they’re community health workers.”
Meanwhile, intense collaborative efforts with police and other entities, have led to a dramatic lowering in mortality rates among individuals who overdose on drugs. “We’ve been successful in going from the highest mortality rate per capita of the five boroughs in 2016, to the lowest, in 2017,” Conte said. “Working with the police department, district attorney’s office, and recovery centers—that partnership has worked. We do detailing of physicians’ offices, to get them used to the idea that these patients are your patients, you can be trained… and we’ve gone from the lowest amount of medication-assisted interventions, to the highest in 2017.”
In the end, Conte said, several elements have turned out to be critical success factors. “Data is the oxygen that enables delivery system reform. Aggregation and governance of data is key; data form outside sources is essential; data needs to be turned into business intelligence,” Conte said. And such tactics as hot-spotting, geo-mapping, and population health data virtualization are critical tools, he added. “Technology enables program developments and performance improvement.”