When I was 23 years old, I took a day trip with some friends and ended up in the emergency room at a nearby hospital. At the time, I had recently moved from China to the United States to study economics. Fortunately, all I had was a bad case of food poisoning, but two months later I received a bill from the hospital for about $5,000. As a first-year Ph.D. student with a monthly stipend of about $1,300, the price was an unwelcome surprise.
This experience helped me appreciate the fascinating peculiarities of the U.S. health care market, which challenged many of the economic principles I was being taught in school. Sadly, my experience is all too common in the United States, where purchasing health services is almost always like shopping blindfolded—it is hard to know what you are getting and how much it costs until after the procedure is over. Surprise bills like mine are so ubiquitous, they’ve become the subject of congressional legislation and ongoing reporting by health care journalists. Unlike so many problems in health care, which can be complex and difficult to change, the lack of transparency around the price of services is something I’ve always believed we can fix.
That’s why so much of my work today at Mathematica focuses on making hospital cost and price data easier to understand and use. I’m part of a team that has worked with the National Academy for State Health Policy (NASHP) and Rice University’s Baker Institute to develop and maintain the Hospital Cost Tool. This interactive dashboard provides insights into how much hospitals spend on patient care and how those costs relate to what hospitals charge (also known as list prices) versus the actual prices paid by health plans. It has become a critical part of NASHP’s Center for Health System Costs, which provides technical support and resources to states seeking legislative changes to curb the growth of health care spending.
Work like this, which draws on empirical economic evidence to inform decisions that affect everyone, has inspired me throughout my career. Sometimes the data can reveal unexpected truths that change the way we understand policies or programs. In graduate school, I worked as a research assistant on a project focused on how health outcomes for patients admitted for major heart surgery varied by how close the patients lived to the hospital they used. It was my first chance to work with a large data set, and it revealed that for patients with high health risks, it’s not always to the patient’s benefit to pursue care at the fanciest, most highly rated hospital—if it’s far from home. The farther away you are from your home, the less care you will get from your informal caregiver network, such as family and friends, and the harder it will be for the hospital to coordinate preoperative and post-discharge care with your informal network. The experience showed me how I could use real-world data to unearth a story that was not immediately evident, one with practical implications.
Although my interest in using empirical data to improve health care decisions has been a focal point throughout my career, the tools and methods I use to analyze those data are always changing. Advances in digital technology create abundant information that refreshes at a speed we could not have imagined a decade ago. These same advances also create challenges for individuals and organizations trying to derive insights from this deluge of complex data.
When Mathematica started using the payer price transparency data that has flooded the internet since 2022, it was dealing with data dumps of 300 terabytes at a time. It quickly became apparent that the price data that hold the promise of promoting health care market competition and reducing price disparities would bear little utility if consumers and providers could not access the data in a meaningful way.
So we stepped up to bridge the gap in data access. I worked with a diverse team of researchers, cloud engineers, data scientists, and web developers to find innovative ways to derive insights from the vast, unorganized price data, using the most up-to-date data technology and analytic tools. Today, Mathematica’s Health Care Price Transparency solution integrates critical data points on provider reimbursement rates, quality, and cost of care, which empowers health systems and providers to develop value-based managed care contracting strategies and service-line adjustments.
As advancing technology accelerates the availability of digital information, it is more important than ever for empirical economists like me to be open to new ways of thinking, collecting data, and conducting analyses. My Mathematica colleagues and I are committed to adapting with the technology to tap into new, vast data sets and elicit more accurate, timely insights that inform better decisions. By doing so, we hope to empower organizations and individual consumers to drive a more efficient, equitable health care system.
