A Scholar Sees Resilience in the Fragmented U.S. Health Policy Landscape

Kacie Dragan, MPH’16, thought she was building a career in early childhood education when she landed a job at a daycare in her native New Mexico during the summer of 2011, between her freshman and sophomore years at Barnard College. The center served “high-need, low-income” families. The kids lived with foster parents, had parents in prison, or coped with unstable housing, food insecurity, or chronic health conditions. As she got to know them, Dragan saw firsthand how their well-being—and ultimately, their school readiness—was directly affected by Medicaid, SNAP, housing support, and other social safety net programs. “I realized at the end of the day that health is everything,” says Dragan, who joined Columbia Mailman’s Department of Health Policy and Management as an assistant professor in 2025. “Education is important, but if the kids I was serving didn’t live in a world where health was easy to attain, nothing else would matter.”

The next semester, Dragan (sounds like dragon) enrolled in Professor of Sociomedical Sciences James Colgrove’s History of Public Health, the only Columbia Mailman class open to undergraduates at the time. She finished her bachelor’s in sociology, earned her MPH at Columbia Mailman, and worked a few years as a researcher evaluating how universal Pre-K and gentrification affect health before earning her PhD at Harvard. Today, Dragan deploys quantitative methods from multiple fields to evaluate the health impacts of policies that support education, housing, income, and food access. She also tackles methodological questions—like outcome measurement in claims data, variation in hospital coding practices, and ways to link administrative data. “Medicaid has been the way that we’ve tried to address a lot of social determinants of health in the U.S.,” says Dragan. “But there are a lot of questions about whether using health insurance is the right way to deliver social safety net services.”

You’ve set out to bridge the chasm between epidemiology and economics. Why?

Dragan: These are two worlds that haven’t always gotten along or collaborated, even though both bring valuable perspectives and evidence. Put simply, economists worry that epidemiology doesn't appropriately consider the trade-offs or unintended consequences of policies, outside of health outcomes—for example, how traffic safety policies that promote population health, like protected bike lanes or congestion pricing, might harm local businesses. Epidemiologists worry that economists make unrealistic assumptions about human behavior and priorities, such as calculating the dollar value of a human life. A secondary issue—one I try to address in my day-to-day work—is that the fields use very different terminology, even when using the same research methods and statistical approaches. It can make collaboration and knowledge sharing difficult, so I aim to be a bridge between the two.

How do policymakers access your research findings?

Dragan: For all the disadvantages of having a fragmented social service system, each state and agency can experiment on their own. Researchers like me can then come together with state leadership and each other to exchange information. Currently, I’m working with collaborators at New York University to document the variation in primary care in Medicaid across New York State, evaluating which managed care programs are offering higher quality care to enrollees and how we might be able to intervene to equalize quality of care. The work is funded by New York Health Foundation, which has links to the New York State legislature. The idea is that, in addition to peer-reviewed publications, our findings will be compressed into a one-pager and given to folks in the legislature.

What have you learned about the interplay between housing and health for New York State residents insured by Medicaid?

Dragan: We hear a lot about people who are unhoused going to the emergency department for their care. I have a paper forthcoming that looks at the reverse direction of housing and health, analyzing how a health shock affects housing stability. We found that among city dwellers insured by Medicaid, the probability of becoming unstably housed or homeless increases 20 to 40 percent after a major health event. This highlights the importance of protecting people on the precipice of housing instability from adverse health events.

How does housing affect health among homeowners?

Dragan: Other research I’ve collaborated on suggests that having a mortgage late in life may limit homeowners’ capacity to afford medication copays. We found that after paying off a mortgage, older adults increase their out-of-pocket spending on prescriptions by 50 percent, unless they are covered by generous prescription insurance or subsidies, like Medicare. Our work can help inform discussions about changes to insurance policy that could affect health behaviors and social outcomes like filling prescriptions and housing stability.

You’re investigating the health effects of Food is Medicine, where insurers or hospitals provide healthy groceries. How will the Big Beautiful Bill’s changes to food stamp eligibility, or other local or federal policy changes, affect the relevance of your work?

Dragan: The social safety net in the U.S. is designed to be fragmented. Because of federalism and ongoing debates about who’s “deserving” of certain services, many social safety net programs are run on a state-by-state basis or are only available to narrow groups. As the social safety net evolves—as eligibility criteria change or funding mechanisms change—I’m asking how other sectors can adapt and fill in the gaps. Even as Department of Agriculture funding for SNAP changes, Food is Medicine may give the healthcare system the capacity to fill gaps in food access. Cross-sector collaboration is often seen as making things more difficult, but it can also make the social safety net more agile.

What values inform your work with graduate students?

Dragan: I had a career in ballet as a teen, and I was training to be a professional ballerina and instructor. I tell my students that my prior experience teaching ballet to small kids informs how I teach them: I try to keep things fun, assign interactive projects, and focus on opportunities for growth. My years as a dancer also inform my approach to challenging tasks. I was accustomed to getting constructive criticism and learned to be OK with being a “work in progress.” That mindset is essential for thriving in research and academia, especially with peer review. In the course I teach on quantitative data analysis and coding, there sometimes are tears of frustration, just like my former ballet students; I make sure they know I believe they can do it and that progress—not perfection —is the goal.