[ Winter 2015 ]
Assistant Professor Jessica Cohen is bringing a behavioral economics perspective to public health interventions in Africa.
Here’s the thing, says Jessica Cohen. You can design a public health program or product that works wonderfully, that will spare suffering and save lives. And you could tell people all about it, when and why they should use it, and all the benefits.
“But, if you don’t think about the barriers to change, how the intervention actually fits into people’s lives and the human response to what you are doing, then all that work you have done, this incredible public health tool you’ve created—it could be all for naught.”
An assistant professor in the Department of Global Health and Population, Cohen is a behavioral economist who is pushing to bring the methods and insights of her field to international health projects, especially those in Africa. Rather than banish to the sidelines the messy aspects of human psychology such as procrastination, behavioral economics directly draws on them to explain economic and other aspects of human behavior.
Most of Cohen’s research has involved malaria, a devastating infection that kills 600,000 people annually, most of them children. The disease has attracted A-list funders (the Gates Foundation, the Clinton Foundation, former President George W. Bush’s 2005 President’s Malaria Initiative). But malaria also plays to Cohen’s strength as a behavioral economist: it’s widespread, the symptoms are nonspecific, and there are dozens of decisions to be made at the individual and government level about how much time and effort to invest in its prevention and treatment: “Malaria is one of those diseases where you can’t crack the code without understanding human behavior,” says Cohen.
Malaria and ear infections: surprising similarities
For example, biologically, malaria and childhood ear infections have nothing in common. But from a behavioral economics point of view, Cohen sees a striking resemblance: “There are so many behavioral layers, so many judgment calls, involved with both conditions, especially for parents. Is it serious enough to need treatment or is it just a cold? That question frequently comes up with ear infections and with malaria—although if you miss a serious malaria case, the consequences can be dire. Medications for malaria are given out when the illness isn’t really malaria, just like antibiotics are prescribed when kids really don’t have bacterial ear infections. In both cases, we need to worry about resistance.”
As the mother of a 2-year-old daughter, Cohen understands the fog of calculation surrounding ear infections: Mia has had two or three of them—but Cohen has taken her to the pediatrician six or seven times. “I guess I am batting about .300,” she says. “But so often there’s no problem and you’ve missed half a day of work for no reason and your child has been exposed to other illnesses at the pediatrician’s office.”
What is behavioral economics?
Behavioral economics weaves together human psychology and economics to explain the decisions we make. In many ways, it’s a response to traditional economics, with its classic supply-and-demand curves and assumptions that people are rational and fully informed. By contrast, behavioral economics sees rationality as one factor—a limited one—and instead looks for patterns and explanations for why people don’t behave in their self-interest.
One key concept in behavioral economics is “anchoring”: we humans tend to make, or “anchor,” choices based on reference points that have little, if anything, to do with the matter at hand. For example, the quantity of food people eat at a meal is heavily influenced, not just by their perception of hunger, but by the size of the plate or bowl that the food is served on.
A shocking experience
Cohen is 36 but seems much younger. She is 5-foot-2-inches tall with dark brown hair and brims with intellectual and kinetic energy. “People sometimes think I am a teenager,” she says with a laugh, adding that past attempts to look older with makeup and business attire were fruitless. “When I give talks, I get a sense that people are amused by the fact that here I am, looking like a teenager, while I am talking about advanced topics in public health and economics.”
That Cohen is at Harvard Chan and toiling on projects in Africa is the result of several serendipitous career swerves. As a student at Wesleyan University, she had no plans to pursue a career in economics until she was inspired by a charismatic economics professor. As a first-year graduate student at MIT, she intended to focus on domestic labor issues—until an opportunity to work on a project in Kenya opened up after she took a course on international development, jointly taught by two academic superstars, Harvard’s Michael Kremer and MIT’s Esther Duflo.
“I’d like to tell you that I went to Kenya to save the world, but it was more of a selfish decision,” confesses Cohen. “Being a PhD student can be a soul-sucking experience. To be honest with you, I was looking for a shock to my system.”
The shock was delivered—by the sight of the terrible health conditions among glue-sniffing street children and the stark austerity of health care facilities and by an altogether different smell: a rich earthiness mingled with the smoke of burning trash. Cohen took several newcomer’s risks: an overnight bus trip to Nairobi that left her, alone, in the bus station at 4:30 a.m. and a road trip to the dangerous northern part of the country. “I just did dumb, dumb stuff,” she says. “I think the part of my brain that assesses risk was just not working.”
The best price is free
Although Cohen was hired to work on an agriculture project, she and two colleagues started an NGO in 2004 called TAMTAM (TAM stands for Together Against Malaria) to distribute bed nets free through clinics providing prenatal services. At the time, other programs charged for bed nets, although at a heavily subsidized price (the full price of $5 to $7 is completely out of reach for many African families).
Cohen was presented a golden research opportunity. Kremer and others had already pioneered “random evaluation” of international development projects, which involved the same kind of rigorous, side-by-side comparisons made by randomized controlled clinical trials in health and medical research. Along with Pascaline Dupas, who is now an associate professor of economics at Stanford, Cohen organized a study that involved 16 health clinics in the Western Province, a rural area of Kenya.
What Cohen, Dupas, and their co-investigators discovered was that charging even a small amount for the bed nets significantly dampened demand for them. Studies of other public health interventions in developing countries, including water purification tablets and deworming medications that get rid of intestinal parasites, have likewise found that free is the most effective price.
Is this really so surprising? At the time, it was. In international development circles, there had been a widespread belief that free distribution was well intentioned but misguided. Having people pay a little, the thinking went, was an efficient way of identifying individuals who would really use and value the intervention. Free bed nets ended up in a heap in a corner or were made into clothes.
Since then, however, the World Bank and other organizations have been persuaded by Cohen’s and other investigators’ findings to back free distribution of public health products with widespread benefits which leads to more total use of the products—even if many people don’t use them at all. Cohen says this shift underscores the need for random evaluation. “Nearly all development policy has been based on whims and anecdotes and fashions, not good evidence,” she says. “We need to do a lot of these experiments to find out how to overcome seemingly strange behavioral choices, such as people paying $12 a year to treat malaria with drugs but not willing to pay 50 cents for a bed net to prevent the infection in the first place.”
When to pay a price
Cohen’s malaria research also includes an important study in Kenya of the pricing of the artemisinin-based combination therapy (ACT) recommended by the World Health Organization. In this case, the results were the opposite of the bed net findings: A lower subsidy and therefore a higher price led to “advantageous selection”: more people took the ACT treatment who actually needed it. Subsidies are essential, Cohen says, but a higher price makes people more selective about using the treatment drugs, so there’s less inappropriate use by people who believe they may have malaria but don’t.
When Cohen investigated subsidizing rapid diagnostic tests for malaria, about 60 percent of those who tested negative went on to buy treatment drugs anyway. “We need more research into how people perceive malaria and how certain they are that they have the disease when they have it,” she says.
Along those lines, Cohen is collaborating with Caroline Buckee, an assistant professor in the Department of Epidemiology and associate director of the School’s Center for Communicable Disease Dynamics, on a project that would connect data about perceptions of malaria risk to the actual risk. Cohen says malaria programs could use that information to calibrate and modify their malaria control efforts, to remain effective even as the perceived risk fades. “Right now we have these huge efforts that bring the rates down—but then the foot comes off the gas, the mosquitoes are still there, and the disease comes back. Everyone gets frustrated.”
For public health, the key question is: When it is beneficial to charge a price for an intervention and when it is not? Research has shown that the best price for many prevention efforts is free, explains Cohen. It may not make economic or public health sense, but we human beings are much less likely to take steps to prevent illness than to treat it. We procrastinate. We forget. “If, on top of that, it costs something, then demand for preventive products like bed nets is too low,” she says.
The behavioral economics of treatment is a different story. People recognize that they need it and are much more willing to pay. But there’s a sweet spot between charging too much, so people can’t afford the treatment, and charging too little, which can lead to indiscriminate and inappropriate use of a treatment.
Cohen is now branching out from malaria to explore family planning and maternal health in east Africa, with department colleague Margaret McConnell, assistant professor of global health economics. One project involves post-partum contraceptive services, which are crucial to the spacing between births that improves the chances of bettering both maternal and child health. As Cohen explains, it’s difficult to get women to even think about family planning after they’ve had a baby, let alone do something about it. “You have a lot of competing priorities. You are tired and overwhelmed. It’s hard to coordinate with your partner what type of contraception you should use. And there are a lot of mistaken beliefs about not needing it now and putting it off —that contraceptives have major side effects, that they interfere with breast-feeding, and that you can’t get pregnant if you are breast-feeding.”
Many government health programs offer vouchers that make family planning services free, but women don’t use them—suggesting that price and availability are not the issues. Cohen is now testing whether adding an expiration date to the vouchers might make a difference. The preliminary results are encouraging, she says. “We are applying a very simple concept from psychology: deadlines overcome inaction.”
Human behavior is tricky, she adds, but it is tricky in predictable ways. Theories of procrastination, limited attention, and habit formation can be used to design and test countless public health programs and projects, from vaccination campaigns to water purification projects to healthful-eating promotions. Cohen envisions a future of more efficient and effective public health and international development efforts, if only because governments, NGOs, and other organizations are finally taking evidence-based-and-burnished behavioral economic factors into account. To ignore them is to court failure, waste resources, and possibly cost lives. “Sometimes all you need is to nudge people and human behavior takes care of the rest,” she says.
Life on the edge
Although Cohen enjoys full professional support from her husband, Peter Kovalko, a building contractor, she is hyper-busy these days keeping up with motherhood and academic responsibilities. Her trips to eastern Africa are less frequent.
On her shelves is a reminder of what a dangerous part of the world it can be: about 30 binders of paper surveys of a malaria drug adherence project in Uganda, managed by her former doctoral student, Elif Yavuz, who was killed in the September 2013 terror attack on the Westgate mall in Nairobi. As she worked on a paper based on the project, Cohen read hundreds of e-mails she had received from Yavuz. “They have these wonderful personal notes like, ‘Today I taught the children in the village how to do the Thriller dance.’ It’s hard to read them. Really awful.”
Cohen says she has always worried about the people who work for her, and Yavuz’s death made her that much more apprehensive. She insists on safety precautions but struggles with the “mother-mentoring balance.” “I remind myself that they are basically adults and that they make their own decisions. At the same time, I have to admit that if Elif had still been working for me at the time of her death, it might have pushed me to stop this line of research altogether.”
In her own career, Cohen describes herself as teetering on a line between public health and economics. There are practical problems to the balancing act. It is difficult, for example, to know where to publish her work—in an economics journal or a medical or public health one. “It’s not just about publishing,” she explains. “It is a matter of how to design projects that make sense and have an impact in both worlds.”
But Cohen finds it immensely satisfying—and important—to bring her behavioral health perspective to the realms of international development and public health programs. “Sometimes I worry that behavioral economics is viewed as just academic tinkering,” says Cohen. “But it is tinkering that can uncover ways to make public health smarter and more successful. What could be more important than that?”
Peter Wehrwein is a Newton, Massachusetts-based journalist, specializing in health care.
Photo: Kent Dayton/ Harvard Chan