Nemours/Alfred I. duPont Hospital for Children, Wilmington, Del.
Seemingly everyone in health care is on board with the notion that dollars should reward value rather than volume or intensity of care. There is far less agreement about what value-based care should actually be. Today’s versions have been shaped by a hodgepodge of quality measures, re-engineered delivery systems, and alternative reimbursement schemes. Each embodies value for someone—but not necessarily for all of those who deliver, pay for, and use health care services.
It’s fair to say, however, that value-based care as we know it today is still in beta. A National Academy of Medicine paper earlier this year acknowledged that there is still scant evidence about what works among today’s payment-reform models and which skills physicians and patients need to be effective players within them. As it evolves, value-based care’s flaws will be exposed and, ideally, addressed. “We have to build this plane while we fly it,” remarks Alan Balch, CEO of the Virginia-based Patient Advocate Foundation.
Many of today’s value-based care models trace their roots to the Institute for Healthcare Improvement’s Triple Aim. Arguably, their sustainability may hinge on how true they stay to that trinity of improving population health, the patient experience, and per capita costs.
Some incarnations of value-based care, such as ACOs and episode-based payments, have strong population-based components. David J. Bailey, MD, president and CEO of Nemours, a pediatric health system spanning six East Coast states, says provider groups struggle to understand what population health means.
“You see the term used to talk about a sophisticated form of chronic disease management. That’s not population health, as far as I’m concerned,” says Bailey. To him, improving population health compels health care systems to form partnerships with community-level agencies to address social determinants of health.
Three decades’ worth of research has proven that education, income, housing, employment, racism, and violence are potent influences on health. In some localities where the need is great, Medicaid ACOs and special needs health plans—the Trenton Health Team in New Jersey and Amida Care in New York City come to mind—have developed relationships with community-based groups outside of health care to address these influences, but on a national level movement has been slow. A dozen years ago, Nemours got ahead of the curve, establishing a division dedicated to addressing social determinants of health in collaboration with community organizations.
In 2012, Nemours took what it learned from this effort and implemented a pilot designed to improve asthma outcomes. As described by Bailey in a Harvard Business Review article in June, patient-centered medical homes affiliated with Nemours expanded their care teams to include social workers and psychologists, while Nemours used a $3.7 million CMS innovation grant to hire community health workers who worked with families to identify asthma triggers in their homes. Nemours also used its influence to push for policy changes to end idling of buses around schools and to replace the types of pesticides used in public housing.
Improving population health entails health plans forming partnerships with community organizations, says David J. Bailey, MD, CEO of Nemours.
The results were remarkable: 60% and 44% reductions in asthma-related ER visits and admissions, respectively, and a $2,100 per-child reduction in annual medical costs. Bailey says Nemours has been able to sustain these outcomes since the three-year pilot ended in 2015.
Health outcomes in the United States compare poorly to those in other developed nations, Bailey believes, because so little attention is paid to social determinants of health. “We’re the only nation that spends more on health care than we do on social services,” he says.
Which brings us back to “what is population health?” In Bailey’s view, today’s value-based care arrangements task the country’s health care systems with responsibilities that they may not be well-equipped to handle. Still, he says that health systems have a duty to understand the role they play in a community and become a willing and collaborative community partner.
“It’s easy to get overwhelmed and say, ‘I can’t cure poverty or reduce the dropout rate’—you can go through a thousand can’ts. But we can influence, we can convene, we can inform policy.”
All this activity comes down to what’s best for the patient, or as IHI frames it in the Triple Aim, the patient experience. In the eyes of Balch, at the Patient Advocate Foundation, which helps patients with arbitration, mediation, and negotiation for care- and expense-related issues, the patient focus is a crucial ingredient missing from many of today’s forms of value-based care.
To improve the patient experience, clinically defined outcomes should be supplemented with outcomes of interest to the patient, says Alan Balch, CEO of the Patient Advocate Foundation.
“One of the challenges with [quality] measurement is that we don’t really have good ways to measure outcomes that matter to patients,” he says, noting that patient-reported outcomes and the technology for measuring and reporting them are limited. “We really need a moonshot, if you will, to come up with innovative ways to measure outcomes that matter to patients.”
In a Health Affairs blog in May, Balch and Darius Lakdawalla of Precision Health Economics, a health economics research firm with ties to academia and big pharma, argued that patients are willing to make tradeoffs, depending on what they want out of their care: fewer side effects in exchange for an acceptable degree of efficacy; avoidance of work life disruptions; access to an oral medication in place of an infusion that might leave them wasted for three days. If we are moving toward a system that places more cost on the patients, they wrote, it is reasonable that their preferences be reflected in formularies and coverage policies.
It may be easy for a payer to dismiss this as the stuff of quality of life. But Balch says listening to the patient might reveal variables affecting outcomes that the payer can grasp, such as willingness to adhere to a regimen, recovery time, or the ability to stay out of the hospital.
“Payers and clinicians struggle with this. They say, ‘The outcome we’re going to measure is survivorship. Or patient satisfaction.’ I understand the need to measure outcomes in a clinically defined way that may be appropriate for a population of patients or the average patient, but there needs to be a way to augment that with outcomes that are more specific to the patient in front of you because they may be very different from the average patient.”
Balch thinks of it as having a menu of options that reflect outcomes the patient identifies as important. As described in the foundation’s recent report, “The Roadmap to Consumer Clarity in Health Care Decision Making,” those options could supplement standardized outcomes measures that a payer or health system wants to use.
Integrating patient-reported outcomes into clinical value assessments means patients and clinicians will have to learn to communicate in meaningful ways. As a means to that end, Balch holds up as an example the work of the Life Raft Group, whose super-registry for patients with gastrointestinal stromal tumors (GIST) fosters collective learning. It also has led to identification of research gaps. For instance, researchers used patient-reported and clinical data to gain clarity on the effects of different drugs as third- and fourth-line therapies for GIST that helped inform the sequencing of therapies.
Now that CMS is collecting patient-reported outcomes, assessments of their use are gaining steam. Writing in Value in Health in June, Lee Squitieri, MD, of the David Geffen School of Medicine at the University of California–Los Angeles, outlined a consensus panel’s recommendations for the role of patient-reported outcomes in value-based payment reform: a clear rationale for the measurement and the intended context of its use; use of measures appropriate for the indication; and some level of standardization.
Balch believes that the best patient-reported outcomes are windows into a patient’s actual or potential clinical outcomes, adherence to treatment, and the financial consequences for the patient. “To the extent we’re defining patient value and satisfaction as the ‘hospitality suite’ I think we are doing a disservice to patients and the role they should play in measuring and paying for value.”
Lowering per capita cost is the final dimension of the Triple Aim, but that leaves some important unanswered questions in today’s value-based care models. “When folks talk about reducing costs, a pertinent question is ‘Reduce costs for whom?’” says Bailey at Nemours. “Is it for the patient? For the health system? For the payer?” In Nemours’s asthma-care model, savings benefit the payer—a situation that Bailey says “still has not been corrected.”
Like Nemours, many health systems have one foot planted in a fee-for-service environment and the other in a value-based world. On this, Bailey offers a cautionary tale.
The asthma pilot reduced ER visits and hospital stays—exactly the goal—but it meant a loss of revenue to Nemours. The hitch is that this sort of value-based model of care adds administrative costs that has kept Nemours from expanding it to other chronic diseases as rapidly as Bailey would like.
“An issue that isn’t spoken much about is the increased cost of maintaining multiple care models and reimbursement models,” says Bailey. “The administrative work to produce the data needed to support the outcomes in a simple pay-for-performance plan is remarkable … it will drive costs up for health systems in a mixed reimbursement environment.”
Now, pile on the complexity of value-based care: It may mean pay for performance, bundled payments, shared risk, or full risk. Bailey observes, “When you say ‘value-based,’ you are indicating one of a whole host of different reimbursement mechanisms that require their own administration and skill set.”
One way to confront this reality is through internal efficiencies. Nemours did this through the adoption of Lean principles, which encourage employees to identify and eliminate wasteful processes. Applying Lean allowed Nemours to maintain its asthma-care model—with its higher administrative costs—even after the grant for the pilot ended.
Lean and value-based care are compatible, says Bailey. “You want to strive for outcomes that are important to patients, and whether you are in a value-based or fee-based environment, the object is to drive waste out of the system.”
By way of example, Bailey describes how pre-op nurses and advanced registered nurse practitioners at Nemours came together to map processes for outpatient surgery, then made recommendations for paring away the wasteful ones. As a result, pre-op bloodwork, physicals, and surgery are scheduled on the same day, reducing overhead and making the whole experience much more convenient for families.
The dimensions of the Triple Aim are intended to work in concert, making value-based care an ongoing lesson in systems theory. Overreliance on any one dimension skews the application of value. Yes, fee for service has perverse incentives, but so may value-based care, points out Bailey—pointing to incentives not to provide care, the danger that has shadowed every venture into managing care.
“The reimbursement models are morally equivalent—it’s what you are able to do within a reimbursement system that makes one better than the other. Regardless of reimbursement system, we need to do what is best for the patient—and the rest will take care of itself.”