4

Effective Population Health Policy and Science: Finding Common Ground

The final panel of the symposium focused on finding common ground to move population health forward. Joneigh Khaldun, director and health officer for the City of Detroit Health Department and emergency room physician, provided a local-level perspective. Rahul Gupta, state health officer and commissioner of the West Virginia Department of Health & Human Resources Bureau for Public Health and internist, shared his state-level perspective. A viewpoint from the federal level was given by Ellen Marie Whelan, chief population health officer at the Center for Medicaid and Children’s Health Insurance Program (CHIP) Services at the Centers for Medicare & Medicaid Services (CMS) and pediatric nurse practitioner. The panel was moderated by Joshua Sharfstein, vice dean for public health practice and training at the Johns Hopkins Bloomberg School of Public Health. (Highlights of this session are presented in Box 4-1.)

FINDING COMMON GROUND

To start the panel discussion, Sharfstein asked the participants to comment on how they establish common ground for progress at their respective institutions. Khaldun said that public health in Detroit has seen a lot of transition recently (with at least five health commissioners over the past decade). As the current Detroit health commissioner, she has been focused on the public health ecosystem, working to identify the players (e.g., hospitals, emergency medical services, police, schools) and their strengths and needs. The focus has been on bringing people together, aligning around

common goals, and identifying and filling gaps. She noted that Michigan has a diverse population and, although the city of Detroit does not fully reflect the rest of the state demographically, initiatives in Detroit significantly affect overall outcomes for the region and the state.

Gupta said there is an art to public health that complements the science. Having been a local health officer for 6 years and then a state health officer for about 4 years, he emphasized the importance of building trust and credibility within the communities the department serves. The foundations of trust and credibility are accountability, transparency, and a data-driven approach to action. He likened the latter to showing the community their reflection in the mirror. To begin to understand the opioid epidemic in West Virginia (which still ranks as first in the United States for opioid overdose deaths) the health department initiated a “social autopsy.”¹ Together with several other agencies in the states, they collected data to develop a more complete picture of the individuals who died, and to then identify themes across the population. Information was gathered not only on the medical cause of the drug-related death, but also on the person’s employment status, educational attainment, health insur-

___________________

¹ One of the earliest uses of the term was in the 2002 work of sociologist Eric Klinenberg, titled Heat Wave: A Social Autopsy of Disaster in Chicago.

ance status, marital status, incarceration status, and a range of other social factors that make up the whole individual. An opioid response plan was developed, and the process was open to the public for comment, Gupta continued. Nearly 500 comments were submitted online (the de-identified comments are posted for the public to see), and a public meeting was held. Within months, he said, a policy was drafted, and laws were passed. “The bottom line,” he repeated, “is transparency, accountability, and a data-driven approach that now several other states and local jurisdictions are also following.”

In her role as chief population health officer at a federal agency, Whelan said she looks at the range of population health management. With regard to finding common ground, she said there are several points for which there is general agreement. First, most agree that the fee-for-service approach to health care results in the receipt of services that are paid for, but it does not create a healthier population. Second, there is broad agreement that there are limited resources, and the amount of resources is not going to increase. Maximizing the available resources is essential. Third, there is agreement that spending is not necessarily associated with better outcomes.

Having worked in both the current and prior administrations, she said that approaches to moving population health forward span the political spectrum, and there are many approaches that are not partisan. The Center for Medicare & Medicaid Innovation (CMMI) was created by the Patient Protection and Affordable Care Act to identify different models of care delivery and payment that would be more effective than the fee-for-service model. One model, for example, is the accountable care organization. Another model, the bipartisan Medicare Access and CHIP Reauthorization Act of 2015, moves Medicare away from the fee-for-service model to a quality-based payment model. Whelan noted that the secretary of the Department of Health and Human Services has identified value-based health care as one of his top four priorities.

There is an array of approaches aimed at paying for outcomes, and the intent of payers and health systems is to maximize health through the health care delivery system. As an example, Whelan said that health insurers are spending hundreds of millions of dollars on housing for covered individuals because they have come to understand the effect housing has on achieving improved health outcomes.²

___________________

² Whelan referred participants to a Forbes article. See https://www.forbes.com/sites/brucejapsen/2018/08/14/to-keep-you-healthy-health-insurers-may-soon-pay-your-rent/?sh=36c3142867ce (accessed December 23, 2020).

STAKEHOLDER ENGAGEMENT

Sharfstein asked panelists how they engage their audiences at the city, state, and federal levels as partners in population health efforts. As an example, Khaldun described the unintended teen pregnancy challenge that was launched in the city of Detroit (where a teen is 2.5 times more likely to become pregnant than a teen across the rest of Michigan). Over a 2-year time span, all of the long-acting reversible contraceptive providers in Detroit were contacted and asked about their practices, including which products they prescribe as well as which they do not have access to, their office hours, the types of insurance they take, and challenges in serving the teen population in need of contraceptives. A campaign was launched to educate the community and to connect them to a network of providers. A hotline was established, as well as a health department clinic that is open after hours and on weekends. All patients are seen by providers, regardless of insurance status, she said. Because Detroit lacks a robust transportation system, the health department provides transportation through Lyft. This fits in with the overall approach of looking at the entire ecosystem and filling the gaps where needed, she said.

Gupta said that the state of West Virginia’s approach is results oriented. Process is important, he said, but one can get bogged down in the process and lose sight of the goals. As an example of engaging the population, he discussed the Management of Maternal Smoking (MOMS) program. Smoking is a significant health concern in Appalachia, particularly in West Virginia. The rate of smoking during pregnancy in West Virginia was about 29 percent compared to about 8 percent nationally, he said. The rate of smoking among pregnant women on Medicaid was about 42 percent. Rates were even higher in counties in southern West Virginia, where nearly half (49 percent) of pregnant women were smoking. The MOMS program engaged federal, state, and local stakeholders in planning. The program includes education and training for providers, universal screening of women, connecting women to the Tobacco Quit Line, a text-to-mom program that sends reminders, and an evaluation of the payment policies of the state’s Medicaid partners. He said that 2 years after the launch of the program, data from the Centers for Disease Control and Prevention’s (CDC’s) Pregnancy Risk Assessment Monitoring System show a nearly 25 percent reduction in smoking among pregnant women (from about 29 percent to about 22 percent). Engaging stakeholders, patients, and providers as partners is important, he said, as is developing clear goals for the outcomes desired so that programs can have a long-lasting effect.

Sharfstein asked Whelan what incentives might encourage health care, such as Medicaid programs, to be more engaged in the structural issues underlying health (e.g., employment, environmental conditions,

access to food). Whelan referred to two CMMI models that are working to bridge the community and clinical care aspects of health. The accountable health communities model is funding bridge organizations that refer people to community services. Data are being collected to determine how referral on social issues influences health care costs. The Integrated Care for Kids model was just announced and will also collect data on how supporting upstream identification and treatment efforts for children affects costs and quality of care. She noted that, by law, all CMMI models must be rigorously evaluated. States are also taking creative approaches and are looking to measure outcomes of value, she said, including metrics such as school readiness or reading by third grade. In that regard, Sharfstein referred participants to a roundtable workshop on education metrics and population health.³

Looking beyond the health care sector, Sharfstein asked panelists how they communicate and engage with the business sector about solving community health problems. Gupta emphasized having an agenda that is data driven and working to understand the needs of the community being served and the issues that drive stakeholders. Public health agencies “can no longer afford to exist in a vacuum,” he said. As an example, he described the development of clean indoor air regulations in West Virginia. In listening to the stakeholders, the Kanawha-Charleston health department gained an understanding of the root causes of resistance by businesses owners to these types of policies. Providing the right assistance to business owners led them to embrace the clean indoor air policies. Khaldun agreed with the need to genuinely engage the community in the development of population health programs and policies. The community “may not have the evidence, they may not be as articulate, they may not have a degree,” she said, but they understand the challenges the community is facing and should thus be equal partners in creating solutions. To garner community support it is important to be able to describe real, tangible benefits that programs are designed to achieve, she said. Communication across stakeholders needs improvement, and public health needs to learn to speak the language of the stakeholders when engaging, she continued.

DATA NEEDS

Sharfstein asked panelists to comment on what might help them move their work forward. Khaldun emphasized both the need to listen to

___________________

³ Information about the workshop is available at http://nationalacademies.org/hmd/Activities/PublicHealth/PopulationHealthImprovementRT/2018-JUN-14.aspx (accessed December 23, 2020).

the community and the need to recognize that there is politics in everything. Having the evidence is necessary but not sufficient for success. It is important to consider the stakeholders. Public health educators need to help their students understand politics, policy, and how to engage with policy makers, she said. Policy makers are less interested in large meta-analyses and more interested in human stories, real outcomes, and how a policy will benefit them or their constituents. Gupta said that academic partners in population health can help to address the research gaps. For example, there is a great need for research on the opioid crisis. The needs of the community should inform the direction of the scientific inquiry, which, in turn, drives policy.

Whelan urged those in population health to find opportunities to partner with the health care delivery system. She noted that CMS has been working to develop in-house partnerships by reaching out and working with colleagues at the Health Resources and Services Administration and CDC.

DISCUSSION

Responding to the needs of communities and patients, Terry Allan of the Cuyahoga County Health Department raised the issue of dealing with politics in the midst of a crisis (e.g., the Flint, Michigan, water crisis or the Elk River, West Virginia, chemical spill). Khaldun said that a key lesson from these situations is the need to really listen to people’s concerns and focus on the facts and the science while still being respectful of the community. Gupta noted that the chemical spill in 2014 resulted in 300,000 people in 9 West Virginia counties having contaminated drinking water. The first lesson in these situations, he said, is “Never let crises go to waste,” but seize the opportunity to make changes; in this case, enacting stronger clean water policies to protect the public’s health. The second lesson is that meaningful work can sometimes put one at risk of losing his or her job, but it is important to not let that stand in the way of doing what is right for the public.

Lih Young noted the challenges of creating real change that improves people’s well-being. She asked about mechanisms of accountability and ways to ensure that programs are effective, have value, and are actually meeting the needs of the intended recipients. Khaldun said that silos and competition impede the ability to deliver change in public health. In a given city there will often be different hospital systems competing for the same patients or for the same grant funding. In addition, when programs evaluate their effectiveness, they are often evaluating a specific population, and it is not known whether those with the highest need are being served. She said that aligning programs, identifying the gaps, and sharing

data across health systems and with public health will help to generate better accountability and understand effectiveness. Whelan noted that there is now a dataset for Medicaid that provides an opportunity to compare data across states and link health data with the services an individual receives, such as the Special Supplemental Nutrition Program for Women, Infants, and Children or Early Head Start. Previously, this was not possible as every state was collecting data differently. However, all states are now collecting data in a standardized format. One of the underlying issues, Gupta added, is that there is a thriving “medical industrial complex” that can be at odds with the creation of health and well-being. Social determinants also come into play, and he emphasized the need to focus upstream on prevention of adverse health outcomes.

Engaging Health Care Providers in Population Health

A participant said that, as payment systems are changing, providers will be paid to keep people healthy. Unfortunately, health care providers are often unaware of the work of public health professionals that can help them achieve this goal (e.g., convening community stakeholders). He asked how the public health profession could reach out to the health care professions more effectively to work together on solutions that foster health. Instead, he said, providers reach out to public health management consultants for cost-control measures. There are missed opportunities, Gupta said, and in some cases, there are disincentives. As an example of a missed opportunity, he said approximately 9,000 inmates in West Virginia have substance use disorder. Public options for intervention are not available in prisons and across the country—only about 150 of approximately 5,000 prisons offer some type of treatment for substance use disorder. This is a large population that is not receiving treatment or vocational education to prepare them for reentry into communities.

Khaldun said population health needs to be better able to persuade decision makers of its value (including speaking the other stakeholder’s language) and to simply be bold and insert itself into the process. As an example, she said that during the hepatitis A outbreak in southeast Michigan, she engaged all of the Detroit emergency departments to urge them to start screening and vaccinating the most vulnerable people (e.g., the homeless or those with substance use disorders, who often end up in the emergency department). Public health provided support for providers, making it as simple as possible for them to acquire vaccines, necessary refrigerators, and so on. All of the emergency departments in Detroit are screening and vaccinating for hepatitis A now, she said.

Sharfstein observed that there are two schools of thought about how population health can assert and insert itself. One approach is to speak a

common language and find common ground to work together. The other approach uses power to accomplish goals, including regulatory authority, orders of the health commissioner, and political power. These two approaches are not mutually exclusive, he said. Public health has power, and in the examples discussed, people have not been afraid to use it.

Rural Health

Joni Nelson of the James B. Edwards College of Dental Medicine at the Medical University of South Carolina pointed to the challenges of addressing rural health disparities and added that certain policies may hinder rural health outcomes. Gupta said the fact that many rural communities still do not have broadband Internet connectivity is a significant challenge that needs to be addressed. Some physicians in West Virginia still receive health alert network advisories via fax machine. Access to high-speed Internet is needed for rural areas across the nation to take advantage of telemedicine. Building Internet connectivity is as important as building roads. To address health inequities, rural and frontier areas need to be brought into the twenty-first century, he said, adding that there are rural areas in West Virginia that still do not have clean drinking water. It is necessary to identify the root causes behind why small water systems cannot be adequately maintained.

Khaldun reiterated that Detroit is geographically large and lacks a robust transportation system. As such, one way they are approaching public health in Detroit is to bring services into the community. It should not require a day off from work and travel on three buses to come to the health department, she said. Mobile strategies based in neighborhoods meet people where they are, and this approach aligns with rural work as well.

Whelan added that CMS now has a rural health strategy. When addressing any issue, it is also necessary to consider what differences there might be for frontier areas. CMS has a learning collaborative of rural states that facilitates the sharing best practices. CMS, as the convener, can help rural health agencies determine how to maximize their efforts. Payment models for rural hospitals can also be unique. CMS is working with rural hospitals that are struggling with volume and expertise to help them transition toward more value-based purchasing.

Investing in Children

Sanne Magnan of HealthPartners Institute asked about building common ground for policies and investment in very young children. Investing in children is an area where a lot of common ground exists, and few

would disagree that it is necessary, Whelan said. One of the barriers to progress is that children are not expensive from a health care perspective. When seeking to invest for the most cost benefit, investing in children does not generate much near-term savings. The challenge is to change the discussion to focus on the long-term implications (perhaps decades away) of focusing on early childhood areas. She suggested taking a multigenerational approach, investing in both the child and the family at the same time. Another approach is to work with employers who are purchasers of health care and show them how investment in the health of a child has implications for the parents who are their employees, as well as for growing an able workforce in the long term. The argument must be made that intervention early on can have longer-term returns.

Khaldun agreed that common ground exists for investing in children. An issue, however, is that society overall lacks demonstrated compassion for poor and minority children. For example, does society believe a black child in Detroit is as important as a white child living elsewhere in the state? Clearly, they are equally important, and it is important to “bring that humanity to all of society,” she said.

In West Virginia, the decline in teen pregnancy is attributable not only to access to services, but it is also a multigenerational issue, Gupta said. Women who used contraception are now mothers themselves and pass that knowledge to their teens. He also suggested the need to move children’s health initiatives further upstream to before a woman is pregnant, addressing issues such as adverse childhood experiences from a multigenerational standpoint. A participant agreed that taking a life-course approach shows how much needs to be done before a child is born. In many states, a woman becomes eligible for Medicaid when she becomes pregnant and becomes ineligible 6 weeks after giving birth. This policy “treats women as a vessel for delivering babies with no concern for their health in the long run,” he said. To ensure healthy babies there also needs to be attention to the prepartum and intrapartum periods. He noted that South Carolina now pays for group prenatal care.

Sharfstein noted that there are important opportunities for public health to draw attention to these issues. The evidence of profound injustice abounds, particularly as it affects children (e.g., the effect of immigration policies).

Priorities

To close the session, Sharfstein asked panelists to suggest the most promising areas on which to focus over the next 5 to 10 years. Population health is a multigenerational issue, Gupta said, and urgent action is needed because the next generation is at risk. Premature birth, neonatal

abstinence syndrome, and smoking during pregnancy, for example, have lifelong ramifications for an individual from the moment they are born. For the medium term, Gupta said that obesity, tobacco, and substance use and mental health should be priorities. For the long term, policy approaches are needed that move away from medicalization and toward a population health approach.

Whelan said a better understanding of total cost of care is needed. Most of the focus has been on the costs of health care and shared savings based on the cost. Total cost of care is a broader circle, and might include, for example, the cost of using meals-on-wheels to decrease social isolation and food deserts. As another example, how might initiatives that increase school success factor into the total cost of care for children? It is also necessary to understand the total cost of care and improvement over time. What are the longer-term implications of an intervention, not just for the following 30, 60, or 90 days, but after 5 or 10 years?

Khaldun agreed and added the need for enhanced accountability. She suggested that funds need to be withdrawn if broader community health or population health outcomes do not improve. “Incentives have to be aligned with accountability,” she said.