Integrating Data Science with Public Policy

Adele Balmer is equally at home bird watching along the Virginia coast, knocking on the doors of politicians in Washington, D.C., and crunching health data at her desktop computer. What connects these activities is her enduring passion for science and for using science to drive positive change.

As a CZI State Science and Technology Policy Fellow, Adele Balmer used her data science skills to investigate healthcare spending in Virginia.

Balmer is a doctoral student studying population ecology and ornithology at Virginia Commonwealth University (VCU), where she honed her data science skills. She is also a 2021 Commonwealth of Virginia Engineering and Science (COVES) Fellow, one of a handful of students selected to participate in a program designed to strengthen connections between the scientific community and state government. The COVES Fellowship Program, now in its second year, is an initiative of the Virginia Academy of Science, Engineering, and Medicine, supported by the Chan Zuckerberg Initiative. Our Movement for Science program backs efforts to put scientific evidence at the heart of policymaking, including at the state level.

For 12 weeks earlier this summer, Balmer worked with Virginia’s Senate Finance and Appropriations Committee to understand the factors driving Medicaid spending. Her results could eventually help to improve the health and wellbeing of Virginians and bring down the state’s healthcare costs.

Over the summer, you completed an intensive science policy fellowship. What catalyzed your interest in science policy initially?

Balmer: When I was a Master’s student at Auburn University, I got involved with the National Association of Graduate-Professional Students, an advocacy organization, and ended up becoming their southeast legislative coordinator. I went to Washington, D.C., a couple of times a year and met with policymakers about student loan interest and other issues that were important to graduate students. At that point, I got on the science-policy kick.

As part of the COVES Fellowship Program, you spent 12 weeks working with the Senate Finance and Appropriations Committee. What drew you to that part of government?

Balmer: I interviewed with different departments and agencies as part of the fellowship’s matchmaking process. Finance and Appropriations had never had a fellow before and they didn’t have a specific project in mind. Instead, they said, “You can come in and figure out what you want to do.” That is what made me want to work with them.

I also know that money talks. I wanted to learn how to communicate using dollars and cents. Instead of just appealing to people’s heartstrings, I wanted to be able to say, “This action will improve healthcare affordability, which will save you and your taxpayers this much money.” At the end of the day, that’s what taxpayers care about, so that’s what their representatives care about.

You eventually decided to work with the Health Subcommittee on a massive database called an All-Payer Claims Database (APCD), which collects healthcare claims from public and private payers like Medicare and Medicaid. Why is the APCD important?

Balmer: The whole idea behind it is to provide a more transparent view of where and how much money is getting spent on healthcare and ultimately to improve affordability, efficiency, and cost transparency. Virginia is one of 31 states with an APCD as the result of legislation that was passed in 2012 and mandated in 2019.

To give you a sense of its size, I got access to just one-tenth of the database and there were 17 million data points!

Can you give me an example of what you found when you dug into the data?

Balmer: I decided to focus on Medicaid spending because that’s where a lot of the Commonwealth’s annual budget was going. I discovered that there are vast differences in per capita spending in different parts of Virginia. That made me wonder, why is one locality spending thousands more per person than another? And what choices could policymakers make to improve the health of Virginians and thereby lower healthcare costs?

To answer these questions, I combined a bunch of datasets together. Then, I looked at more than 30 factors that could be driving up Medicaid costs and created a data model to tease out the most likely ones. The factors were things like teen pregnancy rates, access to a primary care physician, number of mental health care providers, physical inactivity, smoking, diabetes, and alcoholism.

My analysis showed that certain factors, including low access to parks and high teen pregnancy rates, were directly influencing Medicaid expenditures, even when income and poverty rates were taken into account. The good news is that these types of factors can be influenced by decision-makers. For example, support for something like a more comprehensive sex education program that includes different methods for preventing unintended pregnancy may decrease Medicaid spending. My analysis suggests that regardless of your beliefs, that would be a smart move financially.

What do you hope the impact of the project might be?

Balmer: All revenues, state and local tax policies, and spending bills go through the Senate Finance and Appropriations Committee. I’m hoping that when they’re making decisions on where to appropriate funds, they will take this type of analysis into consideration. But my goal during the fellowship was really to get the ball rolling and to be able to say, “Here are things that need to be investigated further.”

As a graduate student at Virginia Commonwealth University, Balmer studies prothonotary warblers, a bright yellow songbird that migrates every winter from the Eastern U.S. to Central and South America. They are used as indicator species for the health of Virginia’s wetlands.

What experiences or skills did you draw on to do this work?

Balmer: As a graduate student, I develop mathematical models. Biologists at VCU have been tracking a population of prothonotary warblers, a neotropical migratory songbird, since 1987 and have amassed a really, really large dataset. My research focuses on determining what has affected their population over the last 34 years and predicting what will happen to the population as the climate changes. But math is math. During the fellowship, I was able to use my statistical knowledge to work on health instead of population ecology.

Do you have a message for other scientists or graduate students who might be thinking about doing public policy work?

Balmer: I think everybody should learn how to talk to political representatives so elected officials can make informed decisions on behalf of their constituents. I work in the field of ecology, which is full of passionate people. But the legislation that comes from our research is just not as strong as it could be. I think part of the problem is that we’re not presenting the research in a way that is relatable enough to make other people care. I think learning how to communicate your message effectively can go a long way.

What has the COVES Fellowship inspired in you? What are you thinking of doing next?

Balmer: I still have to finish my Ph.D., but I think I would eventually like to work for the State Department. They have an Office of Global Change, which works on a broad range of international climate change issues, from clean energy to the impact of climate change on human health.

Learn more about CZI’s Movement for Science program.




Supporting the science and technology that will make it possible to cure, prevent, or manage all diseases by the end of the century.

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

COVID-19 Unemployment Rates in Comparison to The Great Recession

What Is The Difference Between A Data Analyst And A Data Scientist?

Machine Learning for Employee Attrition Modeling

Exploring uncharted datasets

Udacity Data Visualisation project

Ethical Responsibilities in Data Collection

Graphing Feature Importance with Scatter Polar Plots

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Chan Zuckerberg Initiative Science

Chan Zuckerberg Initiative Science

Supporting the science and technology that will make it possible to cure, prevent, or manage all diseases by the end of the century.

More from Medium

The Role of Bottom-up and Top-Down Information Cycles in Visual Processing

Addressing Coded Bias with Responsible AI

Traffic Accidents in London City: Perspective of Two Personas