Nudging Populations Toward Better Health

In a conversation with The Regulatory Review, physician–economist Kevin Volpp discusses how behavioral economics can inform the design of interventions to promote healthier decision-making and improve population health outcomes.

Preventable deaths in the United States have continued to rise over the past decade. Globally, experts estimate that preventable chronic diseases account for more than 60 percent of all deaths. By 2030, the global “cost of chronic diseases” is projected to reach $47 trillion. In the United States alone, 90 percent of the $4.5 trillion spent annually on health care is directed toward addressing chronic diseases and mental health conditions.

Drawing on decades of experience, Volpp explains how behavioral health interventions can address these public health challenges in ways that are scalable and sustainable. He describes how health systems can leverage behavioral “tools”—financial incentives, social norms, and “choice architecture”—to improve rates of medication adherence, vaccination rates, and nutrition. These tools can also reduce harmful behaviors such as smoking and distracted driving.

Although “behavioral economics strategies” have led to measurable improvements in health outcomes, Volpp emphasizes persistent challenges in scaling even the most promising initiatives. Highlighting his team’s recent work, including its efforts to expand access to “food is medicine” internationally, he shares insights on how practitioners can confront challenges and build solutions with long-term viability. Volpp underscores the importance of promoting transparency and forming strong, aligned research partnerships to ensure that behavioral health interventions can succeed in real-world settings.

Volpp is the Mark V. Pauly President’s Distinguished Professor at the Perelman School of Medicine and Health Care Management at The Wharton School of the University of Pennsylvania. At Penn, he serves as the founding director of the Center for Health Incentives and Behavioral Economics and co-creator of the Penn Way to Health platform, which has been used to support behavioral interventions in more than 420 studies across the United States. Volpp was also a key figure in launching the Penn Medicine Nudge Unit, “the world’s first behavioral design team embedded within a health system.”

In his research, Volpp applies insights from behavioral economics to test innovative ways of improving patient health and promoting provider engagement. He has authored over 300 articles, and his work has been recognized widely, receiving the National Institutes of Health’s Matilda White Riley Award for career achievement in social and behavioral sciences, the Society of General Internal Medicine’s John M. Eisenberg National Award for career achievement in research, and the Association for Clinical and Translational Science’s Distinguished Investigator Award for career achievement and contributions to clinical and translational science.

The Regulatory Review is pleased to share the following interview with Kevin Volpp.

The Regulatory Review: What drew you to the study of behavioral economics? What role does behavioral economics play in your research on population and global health?

Volpp: Behavioral economics is embedded in all the research I conduct. My colleagues and I have conducted countless studies that strive to improve population health using behavioral economics. I was drawn in more than 20 years ago as a physician–economist who was taking care of patients. I could see that our health system did not do a good job of prioritizing healthy behavior in our efforts to improve health.

I can share recent examples of projects I’m working on through Penn’s Center for Health Incentives and Behavioral Economics (CHIBE).

First, I’m the scientific lead on an American Heart Association Health Care by Food^TM initiative, which is working to advance “food is medicine” interventions that can treat, manage, and prevent diet-related diseases. This is an interesting opportunity from a behavioral change standpoint. We’re thinking about the kinds of programs that clinicians would want to refer people to, that patients would want to sign up for, and that payors would want to pay for.

Second, a CHIBE team recently worked on a study that demonstrated how behavioral interventions can decrease handheld phone use while driving, which, if scaled, could reduce distracted driving and crash risk in the population.

In another recent study, my team leveraged principles of behavioral economics to encourage patient engagement with population health screening programs. We found a 45 percent to 65 percent relative increase in the rate of patient engagement.

On a global level, CHIBE researchers are targeting health conditions in populations as diverse as Singapore, India, Peru, Uganda, and South Africa, among many other countries.

For example, our researchers are using text messaging that leverages the “fresh start” effect to encourage people living with HIV to return to care in South Africa. We are drawing insights from participant perspectives to learn more about incentives for adherence to tuberculosis preventative therapy and reduced alcohol use in Uganda. In addition, we are adapting and scaling a successful diabetes prevention program in India.

TRR: How have insights from behavioral economics shaped your understanding of how to address public health problems?

Volpp: I think behavioral economics can help shape how we address or think about public health problems. We know that people often have difficulty behaving in their best interests. For example, we know that smoking is bad for your health, but having that knowledge doesn’t make quitting smoking easier.

I agree with what Daniel Kahneman and others have emphasized: The world is already challenging to navigate, and instead of pushing people harder, we should focus on making things easier.

That’s why behavioral economics can be so useful in tackling pernicious health problems. It allows us to preserve people’s choices and autonomy while nudging them toward better choices, such as opting for healthier food or using social norms to discourage clinicians from prescribing antibiotics to patients who probably don’t need them.

It’s also important to see how interventions work in different contexts. Helping patients navigate a chronic health condition in California may look different from doing so in Alabama. An intervention designed to increase uptake of a vaccine in the United States may not work in Kenya, or vice versa. Addressing public health problems requires looking at issues from multiple perspectives. In the end, we are just trying to use behavioral economics to help people improve their health.

TRR: What is a behavioral health intervention? Could you provide an example of an intervention that you have studied or developed to address a public health problem?

Volpp: A behavioral economics intervention uses tools such as financial incentives, commitment devices, social norms, choice architecture, or gamification—among many other techniques!—to help people address challenging health domains, including exercise, medication adherence, smoking, vaccination, and nutrition.

One of our studies tested behaviorally informed financial incentives for smoking cessation among employees at General Electric, and we found a tripling of long-term smoking cessation rates. This research led to a benefit design innovation adopted by General Electric for its 152,000 U.S. employees. We replicated these findings with 2,500 CVS Health employees as well.

TRR: What types of public health problems benefit most from behavioral health interventions?

Volpp: We look at public health problems from both an individual and system perspective to maximize our impact.

On the individual level, the types of problems that behavioral science can help with the most tend to relate to behavioral choices. How can we get people to be more physically active? How can we nudge people to choose healthier food options? How can we make it easier for people to take their medications? The types of individual public health problems that benefit most from behavioral health interventions tend to relate to conditions where behavior plays a large role in determining people’s health.

At the system level, CHIBE-affiliated faculty are using behavioral health interventions to improve health in creative ways. For example, Christina Roberto and her team found that taxes on sugar-sweetened beverages significantly decrease consumption. In addition, Aaron Richterman, Harsha Thirumurthy, and other colleagues have found that cash transfer programs were associated with a 20 percent reduction in mortality for women and an 8 percent reduction in mortality for children younger than 5 years old in low- and middle-income countries.

TRR: How do you measure the success of behavioral interventions in health care settings?

Volpp: I’ll answer this question using two examples of successful interventions.

Recently, we found that semi-automated orders for pharmacist referral were associated with a doubling in appropriate statin prescribing compared with usual care. We measured success of this intervention by seeing a large increase in people who would benefit from a statin getting a statin prescription, and we also saw this project as a win because it reduced clinician’s workloads by batching orders so that the primary care provider could sign off on all the statin orders for their eligible patients at the same time. This successful approach is being rolled out across Penn’s health system.

In another project, Shivan Mehta and his team created a program for colonoscopy preparation, sending behaviorally informed texts to patients to help them prepare for the procedure. There were promising results from a learning pilot—a 28 percent increase in show rates to colonoscopy appointments—but then the team saw null results in a larger sample. An important part of the value proposition was in easing the staff and clinical team’s phone call burden and in focusing their efforts on patients who needed it the most, rather than calling every single patient with instructions. The following year, the intervention was scaled to 20,000 patients, and the team saw a 6 percent increase in show rates.

So, I would say you can measure the success of behavioral interventions in a few ways. Does the intervention improve patient outcomes? Does it do so in a cost-effective way? Does it reduce clinician workload? Does it make it easier for staff to do their jobs?

TRR: What are some of the challenges of implementing behavioral health interventions at scale and in ways that promote health for all?

Volpp: There has been a lot of success using behavioral economics strategies to improve health, but some of the most efficacious interventions suffer from what John List has described as substantial “voltage drops” in effectiveness trials. Moreover, many promising ideas fail to scale. It’s important that behavioral economics-focused approaches work in the real world and not just in research studies.

We have adopted a few strategies for avoiding voltage drops. First, we strive to be intentional in the creation of research partnerships with organizations that share our interests and have the capability to test ideas in ways that mimic scaled implementation. Second, in much of our work we have been building human-centered design into the front end to design programs that are inherently more engaging to potential participants. Third, we use a combination of tests to assess efficacy among trial participants and effectiveness among all those offered the program, focusing on both efficacy and program engagement. Finally, we aim to be as transparent as possible in terms of methodological rigor, methods, and data to ensure that studies can be replicated successfully.

We look to increase health for all through several different centers and initiatives. Our teams are looking at improving representativeness in clinical trials, bolstering economic opportunity to improve health, developing evidence-based best practices and guidelines for conducting research and improving outcomes across a range of urban and rural populations.

TRR: You co-created the Penn Way to Health (W2H) platform, an online platform used to facilitate “behavior change interventions.” How has W2H expanded opportunities for collaboration in the process of designing, implementing, and evaluating behavioral change interventions?

Volpp: W2H provides technology infrastructure for behavioral change interventions. It has supported about 450 programs and is currently supporting 163 active research and clinical programs, touching the lives of nearly 2.2 million participants across all 50 states. Teams from more than 15 universities have used the platform to facilitate their research.

Teams that have used W2H have observed improved patient outcomes, reduced readmissions and mortality, decreased burdens on care teams, and many other positive outcomes.

The platform focuses on automation and flexibility and can develop solutions for specific patient populations anywhere in the United States. Beyond the software, W2H also provides research teams and clinicians with information on design, content, and communication frequency.