Alumni Make Data Shine with Public Health Dashboards
As the sheer volume of health data has exploded over the last 20 years, new tools have become available to help organize and interpret that data. In the first year of the COVID-19 pandemic, public health dashboards proliferated—including several developed by Columbia Mailman faculty to track the outbreak and anticipate its spread. In recent years, alumni have also built dashboards with a wide array of actionable metrics for policymakers, journalists, and advocates.
Building a dashboard is increasingly within reach, even for people without specialized technical skills. Eric Goldwein, MPH ‘19, used AI to help him code the PBJ Dashboard, which tracks staffing trends across 15,000 U.S. nursing homes. “The barrier to entry on creating these data visualizations is lower than ever,” he says.
Nursing Home Insights
At Columbia Mailman, Goldwein took courses in aging and data visualization while earning an MPH in Sociomedical Sciences. After graduating, he secured a fellowship at Long Term Care Community Coalition, a nonprofit advocacy organization for nursing home residents. It was the early days of the COVID-19 pandemic and nursing homes were making headlines, as facilities struggled to contain the spread of the virus. “What happened to nursing home residents during the pandemic was largely shaped by underlying conditions like low staffing,” Goldwein explains. “These problems remain six years later, even if they’re no longer frontpage news.”
As he became an expert in nursing home data, Goldwein, inspired by COVID-19 dashboards, began thinking about building his own. The government website Care Compare helps people shop for a nursing home; it shows several metrics, including a snapshot view of current staffing levels. Launched this year, the PBJ Dashboard stitches together nine years of nursing home staffing data from the federal government—payroll-based journal (PBJ) submissions—to provide a wider window for assessing staffing trends over time.
While Care Compare remains the best tool for consumers, potential users of PBJ Dashboard include state-level ombudsmen who resolve issues between nursing homes and their residents, as well as lawyers and journalists who can see if a nursing home was understaffed during a specific period or whether it has a history of understaffing. Advocates, too, find it useful. “I was talking to someone from Rhode Island who's an advocate who was excited by the ability to look at the metrics and tie them to what she knows through her own grassroots work,” says Goldwein, who now leads 320 Consulting.
Data for City and Congressional Decisionmakers
Big cities like New York and Chicago routinely produce and use metrics and data visualizations to help them set public health policy, but small and mid-sized cities have typically lacked the ability to wrangle their data this way. Launched in 2018, the City Health Dashboard is bridging this gap by providing metrics on over 45 measures of health and drivers of health for over 1,200 cities across the country. Ben Spoer, PhD, MPH ‘14, serves as program director for the City Health Dashboard and the Congressional District Health Dashboard, both of which are funded by the Robert Wood Johnson Foundation and housed at the NYU Grossman School of Medicine.
As a master’s student in Sociomedical Sciences at Columbia Mailman, Spoer developed an interest in urban health, partly driven by his involvement with Hike the Heights, an annual community walk developed by former faculty members Mindy Fullilove and Lourdes J. Rodríguez. At NYU, where he earned his PhD, he worked with geospatial data, including data on the New York City built environment. Today, he and his team are helping cities across the country optimize their public health investments. In Maplewood, New Jersey, officials used the City Health Dashboard to help them relocate their health fair to a neighborhood where people had a greater need for the fair’s screening services. “Cities that find the most benefit from our website are cities that are conscious of the role data can play in their decision making but may not have sufficient budget to hire a data expert to wrangle that information,” Spoer explains.
He and his team of 18 dashboard specialists are continually working to improve and add new features. Early this year, they added a Medicaid enrollment metric to the Congressional District Health Dashboard. This metric caught the attention of a CNN reporter who used the dashboard to show how Medicaid cuts in the One Big Beautiful Bill would affect Republican districts where many people rely on it. “The data told the story,” Spoer said.
The Threat of Disappearing Data
Amidst all the policy turmoil this year, prospect of losing federal data has largely flown under the radar; however, for public health leaders, the potential downside is enormous. Already, CDC data on health equity, gender identity, and COVID-19 has gone dark, and now there is concern that staffing cuts could interrupt data collection across many domains. Moreover, individuals may be less willing to respond to surveys in the current political environment.
The City Health Dashboard and the Congressional District Health Dashboard both rely on federal data sources, including those from the CDC and the U.S. Census. It’s what allows the dashboards to make comparisons across the country. Switching to state and city data sources will be a heavy lift, but Spoer says his team is working on ways to adjust. “Though there’s no way to substitute for rigorous federal data sources, we're developing new tools for statistical analysis every day, and a lot of them hold some potential to help us adapt.”
Anyone shopping for a nursing home might notice that staffing data are now slightly out of date. Could this delay indicate a policy change? For now, Goldwein attributes the situation to “normal, slow government” and a planned transition to a new backend system. Even so, he cautions, “anyone who works with government data should be backing it up on their servers.”