When NYU Wagner researchers wanted to find out whether the noise of planes taking off from La Guardia Airport affected the well-being of the people living in the neighborhoods below, they knew just where to look for clues: the huge dataset that is the New York State Medicaid claims file. Through a unique contract with the state Health Department, Wagner scholars are afforded complete, though carefully controlled, access to Medicaid-related information for over 7.3 million New Yorkers, which can be mined for insights into how policies, public services, and social conditions affect health—and how and why residents seek or avoid care.
In this case, the data revealed that those living under La Guardia's takeoff paths suffer increased diagnoses of cardiovascular disease, mental health/substance abuse emergencies, and even insomnia, according to the recently published analysis co-authored by health economist and Wagner dean Sherry Glied. She and the co-authors of the study are part of a small but growing number of scholars making the most of the school’s one-of-a-kind Medicaid Data Initiative directed by Wagner professor of health policy and public service John Billings with Charles Neighbors, associate professor in the Department of Population Health at NYU Grossman School of Medicine. Billings set the stage for this project after years of working with Medicaid data to identify the most frequent users of hospital Emergency Departments and how to reduce costly recidivism. (His foundational study came out in 2009.)
Like divers fathoming ocean depths, a highly skilled team of HEAL analysts helps Billings and Neighbors plumb this reservoir for relevant data in response to requests from the broader research community, policy makers, and healthcare administrators. In connection with the initiative, Wagner is a research hub for the Robert Wood Johnson Foundation's "Culture of Health" project, which considers how education, housing, mass transit, public services, and other social determinants can affect communities’ and individuals’ health.
NYU News spoke with Glied about the Medicaid Data Initiative and what she characterized as its “virtually limitless” potential to distill insights into the challenges facing New Yorkers—such as the potentially deleterious roar of ascending aircraft—and the policies needed to make lives a little, or even a lot, easier.
How did Wagner gain full access to the huge Medicaid claims file?
There’s a history to it. Professor Billings had been working with Medicaid data for a very long time when he obtained—with the support of then-New York State health commissioner Nirav Shaw—a grant from the state Health Foundation to encourage broader research access to the state’s Medicaid claims-and-encounters data. A data-use agreement with the state Health Department gave Billings direct access to the Medicaid Data Warehouse, and an array of possible research projects were identified. Then in 2016 Billings joined with James Knickman, a former professor of population health at NYU Langone, to launch the Health Evaluation and Analytics Lab, or HEAL, to advance the endeavor.
How are the state’s strict confidentiality requirements upheld?
HEAL designed its procedures so that very few people actually touch these data—a very small team of extremely skilled analysts. The computers are not only digitally protected, unconnected to the web, but they are even locked behind cages. The scholars who have collaborated with us on the research do not get direct access to the data, but rather are provided with the relevant summary tables. The very limited number of analysts with hands-on access must sign releases, sign their life away. However, in all honesty, if anyone else tried to figure out how to utilize this database, it would take them forever. In comparison, the team of analysts can turn things around quickly.
Can you discuss some of the research findings so far?
One of the first studies considered the 2014 rollout of Universal Pre-Kindergarten throughout New York City. We looked at the healthcare utilization of children born before and after the program's start date, and found an interesting result that no one had seen before: the kids who got into Pre-K were much more likely to be diagnosed and treated for vision problems. These kids were getting eyeglasses much sooner, with a lasting effect on their educational outcomes.
Working with Professor of Urban Policy and Planning Ingrid Gould Ellen and the NYU Furman Center's highly precise geographical data for New York City, we’ve also seen that while lower-income children move all the time, gentrification was not a predictor of this. Surprisingly, this study found a limited effect from kids' moving so frequently, or on the health of children born in neighborhoods that subsequently went through gentrification.
Did you look for health impacts from eviction?
We did. Many people lose Medicaid in the course of being evicted. When they showed up again in the data returned to the Medicaid rolls, we saw various negative health effects. However, in another study, we found that among those who lived in buildings where landlords made renovations—better lighting or windows, say—their disease burden overall was lighter.
Can the state Medicaid database tell us anything about the health of inmates?
We actually conducted an in-depth project with the Mayor's Office of Criminal Justice that matched discharged Rikers Island inmates to the Medicaid data to see what kind of health conditions the ex-inmates have, and whether they sought health services for them. We found out that people who are frequently in jail are frequently in the hospital when they are not in jail. But one of the things that surprised us is that they often have really high rates of alcohol use. Alcohol use, of course, is not illegal, while drug use is.
What are some of the studies coming up?
We are looking at whether extending Medicaid for longer periods between the mandatory renewal dates would help reduce jail recidivism. This study is based in Westchester.
We’re also evaluating the impact of the COVID epidemic on the health of Native Americans residing in northern New York State, a neglected and disadvantaged population. While they are eligible for public health services, what happens when these services run out of money before the end of the year, which they often do? What happens to those who also fall off Medicaid, and fall through the cracks?
Another paper, which is awaiting publication, turns on a federal policy that makes babies born at less than 1,250 grams, or about 2.75 pounds, eligible automatically for Supplementary Security Income, or SSI, if their families are poor. The data show these children fare so much better than babies born at higher weights, who are not eligible for SSI. In addition, the savings to the Medicaid program resulting from their reduced health care utilization far outstrips the SSI costs. So, babies who received the benefit were far less likely to get urinary infections, for example. Why? Their parents could afford more diapers, in part.
It shows the potential for impact from even small adjustments to rules and policies.
An example of that is what the data revealed about paid sick leave, namely, the benefit leads to regular doctor's visits and better health outcomes for the recipients, keeping people out of the emergency room. The U.S. is one of just two Western countries that does not mandate paid sick leave. We hope to compare our results from the introduction of New York City's mandate to those of mandates more recently introduced in other cities in New York State. Overall, though, the Medicaid Data Project promises an endless number of potential projects. Many, we expect, will have important implications for policy makers.