During a daylong virtual meeting of the Population Health Delivery Council, hosted by The American Journal of Managed Care®, a panel of experts discussed population health, unmet needs in digital health, technology solutions to support data transparency, and opportunities for collaboration between industry and health care stakeholders. It was moderated by Dennis Scanlon, PhD, director of the Center for Health Policy Research at Pennsylvania State University and editor-in-chief of The American Journal of Accountable Care®.
Paul Tang, MD, MS, adjunct professor, Stanford Clinical Excellence Research Center at Stanford University, kicked off the discussion in the morning session by posing the question of how we should think of populations—or whether we should think of them at all. Most evidence-based medicine for populations is based on randomized controlled trials, he explained, which removes confounders and leaves a homogenous population. In a large practice, that will represent only a very small percentage of the actual population they treat.
“If we base our evidence and guidance on 1 in 20 people, what should we do for the other 19?” he asked.
Ochsner Health organizes around populations of customers with similar needs, explained Kenny Cole, MD, MHCDS, Ochsner’s vice president of clinical improvement. Because individuals who are healthy have very different needs than individuals with chronic conditions, Ochsner reorganized its care delivery around those with chronic conditions based on those who have multiple health determinants versus those with fewer. In this way, it has redeployed its resources.
Want to publish your own articles on DistilINFO Publications?
Send us an email, we will get in touch with you.
Dave Chokshi, MD, MSc, vice president and chief population officer of NYC Health + Hospitals and clinical associate professor at NYU School of Medicine, considers population health as a proactive approach to addressing avoidable human suffering.
“I think it’s really important, particularly when we talk about patients with chronic conditions, to focus on the idea that our ultimate goal is to figure out how we can address and potentially prevent some of that suffering,” he said.
An important element of population health can be identifying what can be done to intervene earlier and in a larger way to make a difference at the individual level. This means addressing the factors underlying chronic illness for a population, which requires understanding why some people develop, say, diabetes or asthma in the first place and try to prevent it.
“I think that requires a little bit more of a life-course perspective,” Chokshi said.
In addition to prevention, MGB Partners, composed of Brigham and Women’s Hospital and Massachusetts General Hospital, decided to see how it could tackle complex patients with multiple comorbidities, explained Despina Garalis, PharmD, RPh, director of population health management programs, Partners Community Physicians Organization, founded by the hospitals. They set up care teams including nurses, social workers, and community health workers, and they used this team approach to help patients with direct outreach.
For outreach regarding prevention, Garalis’ organization utilizes its information-rich electronic health record (EHR). They’re still tweaking their process to better understand how to tackle social determinants of health.
“If you don’t handle and tackle some of the social determinants, you really can’t help with prevention, and then it just leads into [patients becoming] medically complex,” Garalis said. But there are challenges: “Are you limited with the community resources? How does a primary health care system deal with social determinants when you don’t have the community to kind of help with that?”
In using data to better understand their population, Cleveland Clinic found that untreated behavioral health needs are among the biggest drivers of care, as are unmet social determinants of health, such as health literacy and financial support, said Jessica Hohman, MD, MSc, associate medical director, Cleveland Clinic Accountable Care Organization (ACO); investigator, Center for Value-Based Care Research; and research associate staff, internal medicine, Cleveland Clinic.
The lack of patient engagement, another driver, is related to poor outcomes for patients as well as high spending for Cleveland Clinic, she added. Cleveland Clinic is not only trying to understand which patients are not engaged, but why they aren’t engaged.
“How do we engage patients?” Hohman asked. “How do we meet them where they are? How do they want to be engaged in their care to really move their care forward?”
The fee-for-service system remains among the biggest challenges to adapting to value-based care and population health, but there are also operational difficulties. Getting meaningful data in the hands of the provider in real time is crucial so they can use the data in the context of decision-making, Chokshi said.
“I think by and large…most providers want to do the right thing,” Chokshi said. “They often know what the right thing is to do, [but] we just haven’t set up systems to allow them to achieve that.”
Juliann Molecavage, DHA, associate vice president of quality and primary care at Geisinger, added that the old way of operating—in which payers would provide monthly reports on the population of patients—overwhelmed providers, because they didn’t know what to do with all that information. Now, providers can look at the diseases faced by the majority of their patients and determine the evidence-based preventive measures they can take that can make a real difference. Those real-time data are right in their EHR; claims data prompt providers at the point of care with actions to take.
Identifying and Engaging Patients
Each panelist provided some real-world examples of how their organization has used different approaches to identify patients who need help and provide them with appropriate care.
For instance, ChristianaCare launched a pilot to care for patients with uncontrolled diabetes that included embedding behavioral health consultants in each primary care practice, explained Amanda McMullen, PhD, RN, CPHQ, vice president of population health and value-based care for ChristianaCare and eBrightHealth ACO. The primary care providers identified which patients needed help and referred them to the behavioral health consultants, who talked them through their medication regimen. The consultants also taught the providers the strategies they were using. As a result, the targeted patients saw a 35% improvement in their diabetes control.
In focusing on heart failure, Cleveland Clinic found that some of the tools for capturing and targeting patients with this condition were sometimes inaccurate. The first task, then, was to identify these patients accurately and create a standardized way to capture them in the EHR, explained Hohman. They also focused on transitions in care, not just when patients were leaving the hospital, but also when they were going in and out of skilled nursing facilities and long-term care facilities.
Cleveland Clinic also found that their mortality rates for heart failure were much higher than their hospice utilization rates. In addition, the length of stay of patients who did get referred to hospice was generally very short. “They were being referred right as they were actively dying, and we have worked hard to change that,” Hohman said.
Patient engagement was another important piece of the puzzle for organizations doing population health work. Anthony Keck, MPH, chief population health officer of Ballad Health, noted that at the South Carolina Department of Health and Human Services, where he was previously director, a program existed to manage high-utilizing uninsured populations, and part of that was an engagement screening to measure patients’ feelings of responsibility for their own health. The department found that the members of high-utilizer populations felt similarly engaged in and responsible for taking care of their own health as people in any other populations. The issue was that these high utilizers had barriers.
For instance, “they felt just as responsible to take their medications on time, but they couldn’t get to the pharmacy to get them refilled,” Keck said, “or they didn’t have the money to pay the co-pays.”
Keck added that much time spent discussing patient engagement amounts to a “dog whistle for folks to blame the patient,” yet in his experience, in working with several programs, he found that background factors help create barriers that prevent some high utilizers from fully engaging in their health. They don’t “make a choice to simply give up on their own health,” he said.
Cole added that patient engagement also comes from trust in the people providing care. In an organization in which he previously worked, the word “noncompliant” was banned and “nonadherent” was viewed as being just as bad. The result of this mindset was usually that the care team members’ ability to earn trust successfully affected their ability to foster patient engagement.
Empathy and the ability to identify with patients is a critical skill, Cole said, perhaps even more so today, when George Floyd’s death has resulted in nationwide protests over racial discrimination and, along with the impact of the coronavirus disease 2019 (COVID-19), has put an increased focus on racial equity in health care. Health care organizations have to think about the employees hired by practices, with whom the patient is asked to engage. In some cases, community health workers are being used “as a patch,” he said, because they reflect the population being treated more than the doctors and nurses do.
Treating patients based on their individual symptomatology can be at odds with what is needed in population health, which requires scaling something with a degree of standardization, said Devdutta Sangvai, MD, MBA, executive director for Duke Connected Care; associate chief medical officer for Duke University Health System; and assistant professor of family medicine, pediatrics, and psychiatry for Duke University Medical Center. However, in working one-on-one with patients, team members can uncover unanticipated issues that may be of prime importance to that sole patient.
“Until you find a way to move forward on that [individual life issue, patients] aren’t going to pay attention to their heart failure or their diabetes,” Sangvai said.
Connecting Value to Payment
The program then shifted to the area of linking value to reimbursement; Scanlon asked the panelists what they believe that actually means. Cole noted that value should be the outcomes that matter to patients divided by the cost of delivering those outcomes, but Scanlon played devil’s advocate and pointed out that the outcomes that matter to patients are likely to be varied and diverse. How would a system standardize that?
Cole responded that health outcomes can be categorized in larger buckets. For instance, capability is related to functional outcomes; comfort is defined as the alleviation of physical and emotional suffering; and calm is when patients achieve capability and comfort in an appropriate environment free of mayhem and confusion.
The challenge is making sure that providers ask patients what they really want and listen to the answers, according to Molecavage. Instead, providers usually go in with an idea in mind of what they want for the patient. Garalis agreed, saying that patient goals are not always the primary focus because there are so many external pressures.
Payers have set hundreds of goals for different disease states, added Keck, which is an artifact of the fee-for-service system. Moving away from that system would give providers more time to engage with patients on what is actually needed to improve their health.
Taking a step back, population health has been developed and shaped by external pressures, Garalis said. If, instead, the system had begun with a patient-first foundation, it would look quite different, she said: “If I could start it all over again, I would say, look at the patient and see what that patient needs holistically, then build off that.”
Tang agreed, adding that he would “just like permission to do the right thing.” If all the perverse financial disincentives were removed and if we measured what matters most to individuals, then the United States would get the most beneficial and least costly system by “tapping into the intrinsic reward system.”
Chokshi said that although he agrees with the fundamental importance of payment reform to advance population health, he is disappointed with how little the needle has moved in the last decade of payment reform. Looking ahead to the next decade, he stated, organizations need to be able to identify and stratify their population; there needs to be greater utilization of high-quality community-based care, pushing the focus away from hospitals and further into neighborhoods and homes; and health care needs to do a better job of meeting patients where they are.
Sangvai’s wish list for a health system that uses population health management requires a degree of tolerance, because “such a change won’t happen in 1 or 5 years. It’s going to be a long-term approach.
“I think to do population health management the right way, you’re fundamentally changing the health care system,” Sangvai said. “That takes time, patience, and accommodation.”
Focus on Patient-Centered Asthma Care
In the afternoon session, the conversation around population health took a deep dive into asthma, a chronic condition that affects nearly 24 million Americans. Scanlon first asked the panelists to answer a poll question on whether their institution has a formal disease management program in place for asthma. He was surprised that only 40% replied that they did; 60% said no.
Garalis noted that while her organization does not have a formal program, it does monitor adherence to medications and reaches out to nonadherent patients. Sangvai added that his institution’s focus on asthma was not quite a formal program, per se, but asthma interventions have been created that aim to decrease patients’ medication burden.
Hohman outlined the components of Cleveland Clinic’s program: It employs a multidisciplinary care coordination team to reach out to patients, ensure they have adequate supplies of medication, and regularly update their asthma action plans. Although care utilization has decreased, Hohman noted the challenges of attributing any changes to the intervention, as patients were not randomized into the program.
Similarly, Chokshi’s team identified patients with poorly controlled asthma based on utilization criteria and sent community health workers to conduct home visits and connect families with needed services such as mold removal or pest control—even providing new pillowcases and cleaning supplies to help reduce potential asthma triggers in the home.
“Even though we took a very clinical and utilization-oriented approach to identifying patients, the patients who we enrolled in the program were predominantly from low-income neighborhoods, families who are living in poverty, and often racial and ethnic minorities,” Chokshi noted.
Scanlon next asked the panelists to discuss how they approach patient engagement, which often involves the whole family in cases of pediatric asthma. Cole stressed the importance of gauging a patient’s health literacy and self-efficacy beliefs to “understand from a 360-degree view” any barriers that might stand in the way of their using an inhaler effectively. According to Molecavage, Geisinger provides an asthma action plan that allows patients to identify their “phase” of symptoms as green, yellow, or red; the plan outlines steps they can take to prevent themselves from reaching the red phase.
When Scanlon asked about critical unmet needs in the asthma space, Tang identified medication management as an important issue, especially for older patients. He suggested that daily pill packs may simplify patients’ regimens, but pill packs need to be made more affordable to ensure equitable access, perhaps through partnerships with industry. Keck took a broader view, focusing on the accountable care community’s efforts to address upstream drivers of disease. “There is a correlation between adverse childhood experiences and asthma, especially with violence in the home and poverty in general, so…we’re trying to get much deeper into the social determinants to set them up for later on down the road,” he said.
A polling question asking if patients’ social determinants of health are predictors of their health outcomes received nearly unanimous answers of “yes,” with Tang asking if he could add an exclamation point to that vote. In response to Scanlon asking the panelists what their organizations are doing to address these social determinants, Garalis explained that Partners identifies patients’ needs during screenings at annual visits and connects patients to service organizations on that basis, but capacity and funding are challenges, she said. Keck added that despite millions of dollars being spent to address these issues, “the connection is so inefficient. It’s really a logistics/distribution type of problem, getting the right patient to the right service at the right time.”
McMullen and Molecavage acknowledged the difficulty of making these connections, which motivated Christiana Health to invest in the Unite Us platform and Geisinger to choose Aunt Bertha as tools to link patients to community entities based on their social determinants and needs. In particular, Molecavage highlighted the positive effects of food pantries on patients’ cost of care, utilization, and disease management, “and then their families are getting the benefit of it, too, because the education is happening and the nutritious food is being provided to the entire family.”
Stemming from the conversation on social determinants, Scanlon asked the panelists to discuss their organizations’ perspectives on equity and disparities. In the Deep South, Cole said, the mindset of “actuarial fairness” is difficult to move against, but Oschner is doing so by embracing a more progressive mindset, spurred in part by a New England Journal of Medicine article addressing health care inequities in Louisiana laid bare by the COVID-19 pandemic.1 Keck added his praise for Oschner’s transformation, noting that the health system did not accept Medicaid when he worked there decades ago. “It was not an easy path,” Keck said, but “it can be done if organizations put their mind to it. It is about organizational leadership and vision and keeping people accountable.”
Chokshi added that COVID-19 has revealed for many organizations that health equity, instead of being a secondary concern, is the “main event.” Referencing a framework from the National Academies of Sciences, Engineering, and Medicine report on health care sector activities to better integrate social care, Chokshi highlighted the 2 most salient of the report’s “5 A’s” as adjustment and advocacy.2 Adjustment is exemplified by Ochsner’s transformation, which was achieved by turning the spotlight inward on its own role in inequities; and advocacy, Chokshi said, is “about lifting our voices together to challenge things like income inequality or mass incarceration or these really entrenched social issues that ultimately we see spilling over into our emergency departments and hospitals.”
Scanlon then asked the panelists to discuss ways health systems can partner with industry. From Garalis’ experience working in retail pharmacy, she suggested trying to find creative ways to incentivize medication access by making drugs more affordable. Cole also suggested working with pharmaceutical manufacturers that are interested in providing tools and resources, but he noted that in order to succeed, every player must be aligned around a value-based payment model.
With the conversation turning to value-based care, Scanlon asked the panelists if they have seen asthma outcomes improve now that the Medicare Access and CHIP Reauthorization Act is in place. Two-thirds responded “no,” and the remainder were unsure.
Chokshi, who answered “no,” explained that the health system as a whole has made progress on the “diagnosis,” as it has reached a consensus on moving from volume to value, but has not yet landed on a “prescription” to fix the problem. So far, alternative payment models have been “hitting more singles than home runs,” Chokshi said, in terms of the marginal shared savings realized, but he is hopeful that in the coming years more “doubles or triples” will occur, by taking into account a broader set of factors and crafting more calibrated interventions within population-based models.
Scanlon then asked the panelists to identify the greatest opportunities for population-based asthma management. Tang mentioned the need to educate the public, including children, about how climate change and the environment affect health, because asthma is too large a problem to tackle through medication alone. Keck brought up opportunities to redirect money wasted on low-value care to more productive efforts, such as early childhood education and lower drug costs for patients.
Educating children about wellness and the environment was repeated by McMullen in answer to Scanlon’s final question about the social determinants that could make the biggest difference in asthma outcomes. Sangvai and Garalis both identified high-quality housing as a crucial unmet need, and Molecavage and Hohman touched on access issues in their responses, with Hohman hoping systems could bring a multidisciplinary approach to asthma to places convenient to patients, such as day care providers, schools, and community centers. Chokshi called for an intensive “kitchen sink” approach that would require cross-sector cooperation to match local organizations with the scale of investment required to operationalize these programs.
Wrapping up the session, Scanlon said that he had learned a lot