Partners for Kids (PFK), a physician hospital organization that manages care for at-risk children in Ohio, originally began as a contracting strategy. It’s since evolved into something that may be looked at as a model for future pediatric care endeavors.
“It started 15 years ago as an organization that took risk for a small number of children, so as not to get stuck with unpaid bills from fly-by-night managed care companies,” says Kelly Kelleher, M.D., vice president of community health services at Nationwide Children's Hospital in Columbus, the founding hospital and finance partner of Partners for Kids. “What happened was the cost of the kids were being managed well and the outcomes were pretty good. So the managed care companies gave PFK more kids. Eventually, the state made Medicaid managed care for poor children mandatory. All of a sudden, the numbers went from 50,000 to 200,000.”
Today, Partners for Kids has contracts with five managed care companies, responsible for approximately 320,000 pediatric at-risk patients in Ohio. It’s formed an accountable care organization (ACO), and if recent research spearheaded by Dr. Kelleher is accurate, a successful one at that. His research team concluded that when compared to Medicaid fee-for-service programs and Medicaid managed care plans, the pediatric ACO did a better job in lowering costs and improving quality care metrics.
Kelleher, who advises PFK on various metrics and measures because of overlap with his job at Nationwide Children’s, is also a faculty member at The Ohio State University College of Medicine. This opened the door for him to research whether or not pediatric patients in at-risk populations can be effectively managed in an ACO environment. He notes that consolidation of health systems has changed the game for pediatricians, making it harder for their populations to be managed.
“The rules for ACOs that the feds are promulgating for Medicare are built around feeding patients into a system through primary care. In many ways, that’s great. In adult systems, the primary care physician should be at the center. But as these places get bought up, they cease to see poor children and Medicaid patients and cease to be a part of our networks, because they’re privately owned and pediatricians have little say in what they can participate in,” Kelleher says. “If we don’t answer the question of whether a pediatric regional ACO can provide more effective care…the adult systems will consolidate and ignore the children.”
He also notes that most ACOs do not have measures specifically for tracking pediatric patients. Pediatricians don’t even have specific electronic health record (EHR) systems tailored for their patients, as a recent report issued by the Agency for Healthcare Research and Quality (AHRQ) noted, making it all the more difficult for them to manage care at a population health level.
Lowering Costs, Improving Care
Despite these challenges, his pediatric ACO shows tremendous promise. Kelleher and his research team compared the cost of care for PFK’s patients from 2008 to 2013 to overall reported costs of Medicaid within Ohio. They also looked at PFK’s performance in quality metrics based on AHRQ’s Pediatric Quality Indicators.
On the cost front, the differences were significant. The costs per member per month for PFK grew at a rate of $2.40 per year, managed care plans grew at a rate of $6.47 per year and Medicaid fee-for-service costs grew at a rate of $16.15 per year. In quality, PFK saw fewer NICU days, fewer visits to the ED for asthma, and a significant increase in the number of well-child visits over the course of the five-year study. There was stability in two other measures and a small decline in two more.
Kelleher isn’t surprised over the differences in cost of care. “Partners for Kids has invested in more and more prevention programs because of the success of some of the early ones. We can monitor reductions in utilization and cost, increases in uses of generics, decreases in hospitalizations for kids with feeding tubes, and other focused pilot projects have yielded positive results,” he says.
The PFK strategy has been a two-fold approach: collecting, integrating, and utilizing patient data across the continuum and thinking about prevention from a population standpoint, specifically in regions. For the first part, the ACO uses technology that Kelleher says is “remarkably simple at this point” thanks to many of the providers, including care coordination notes, being on Epic and claims data being presented in simple graphs. The main challenge is integrating in the textual claims data, which involves the process of “de-duplication” to ensure an individual child doesn’t pop up in multiple sets.
As Kelleher noted, pediatrics has the potential to be forgotten in the new era of population health management, where dual-eligibles and costly adults have gotten the lion’s share of attention. This has not deterred the work at PFK and others, such as the Children’s Health Alliance in Portland, Ore., that are achieving significant milestones in providing quality care for pediatric populations.
It has not deterred the sentiments of Kelleher, who says pediatric ACOs have the potential to “bend the cost curve at the point where it’s most bendable.” He adds that, “Studies have shown that children who start with a bad course are much more likely to have chronic illnesses later in life, chronic mental disorders, develop hypertension and heart disease, and cancer than children who don’t start off on the right foot.”
Moreover, pediatrics lends itself to population health because it’s regionally based, adds Kelleher. It also would reduce the margins for well care since pediatricians can’t afford diabetes coaches and behavioral specialists.
In his work with PFK, Kelleher says the ACO will look to improve upon how it measures outcomes for this population. “We wish the field would come together to figure out an acceptable measure for a pediatric ACO. We have good financial and satisfaction measures. For health outcomes for kids, we want to figure out what should we be measuring and how long we should be measuring it. Is one year sufficient, do we need five years?”