Health Policy

May 3, 2021

VUMC’s Lee among group to earn Kenneth J. Arrow Award for health economics research

Scott Lee, MD, PhD, MPA, MPhil, assistant professor of Medicine at Vanderbilt University Medical Center, has received the Kenneth J. Arrow Award for health economics research given by the International Health Economics Association.


by Jake Lowary

Scott Lee, MD, PhD, MPA, MPhil, assistant professor of Medicine at Vanderbilt University Medical Center, has received the Kenneth J. Arrow Award for health economics research given by the International Health Economics Association.

Scott Lee, MD, PhD, MPA, MPhil

Lee and colleagues at the London School of Economics and the Massachusetts Institute of Technology received the award for their 2020 paper published in American Economic Review that covered a lengthy study about the recruitment of health practitioners in Zambia and its relationship to significant improvement in measured health outcomes.

The award is the highest given annually for health economics research, and “needless to say, I’m hugely honored,” Lee said.

According to Lee, this is the first study to test the impact of incentives on who enters the health workforce, and how this effect on recruitment influences performance.

The study began in 2010 after a brief meeting with the Zambian Ministry of Health, which was planning a recruitment effort for a new national community health worker program. At the time, in Zambia and many other African countries, many rural communities had only one trained health worker, and these workers were often treated more like volunteers than health professionals. The Zambian government was planning to convert this informal workforce into a new civil service cadre, a change that would offer a career ladder, with opportunities for supervisory positions and more advanced clinical roles.

With this new program, Lee, his fellow researchers and the Zambian government saw a potential “double-edged policy sword.”

“It might attract those who want to make a long-term difference in their communities, but it might also draw in those who are primarily motivated by self-interest, which is the opposite of what you want in your local doctor,” Lee said.

So, they designed an experiment that would promote the program in two ways across the country — by focusing recruitment messaging on career opportunities in half of the 165 communities participating in the first recruitment wave, and in the other half emphasizing community service and social impact.

Ultimately, 314 candidates were selected for the new position, trained, and deployed to the local communities covering a geographic area the size of France. Importantly, the actual job benefits were identical between the two groups, such that any differences in performance could only be due to the types of applicants that the two recruitment approaches attracted.

They found the most capable and service-oriented candidates were recruited when career incentives were included in the messaging, which delivered distinct improvements in several measures of health outcomes. Community health workers who were informed of the new position’s career benefits at the application stage generated a 30% increase in health facility utilization for childbirth, a 24% increase in well-child visits, a 20% increase in polio vaccination and a 25% decrease in child malnutrition, among other improvements in maternal and child health outcomes.

In view of these large health impacts and the overall rigor of the experiment, in 2018, even before the study was published, the World Health Organization cited the Zambia experiment as the main source of evidence to recommend that other countries offer opportunities for professional advancement in their national community health worker programs.

“There are not many policy interventions that lead to 20-30% improvements in health outcomes across the board,” Lee said. “While this was just one study and we should always be strengthening our evidence base of what works and doesn’t work in health policy, my personal view is that we can do a lot more to support and empower health workers working in very difficult settings, and this study shows that doing so more broadly could save millions of lives.”