Skip to Main Content

Nancy had been coughing for months. When she started experiencing chest pain, this bubbly mother of three and very proud grandmother went to see a doctor at her local clinic in Thika, about 20 miles northeast of Nairobi. He delivered a crushing diagnosis: She had contracted a drug-resistant strain of tuberculosis. That was in June 2016.

For the next eight months, Nancy went to the clinic daily to receive an injection of a strong antibiotic and take a cocktail of 15 pills that were also antibiotics. She became so weakened by the disease and her medications that she couldn’t walk. Her children carried her to the clinic for her daily visits and provided constant support and encouragement, but she still felt she was alone — she wasn’t working, and her friends avoided her out of fear of being infected by the disease.


Nancy is one of roughly 10 million people worldwide who develop tuberculosis each year. Once on the decline, TB has again become the world’s deadliest infectious disease, killing nearly 2 million people a year, more than malaria and HIV combined.

The cause of TB’s resurgence is not medical; a highly effective though burdensome treatment has existed for the disease since the mid-1940s. Instead, the cause is mostly behavioral: Faced with the prospect of extended treatment and isolating stigma, many people are slow to seek treatment or quit partway through. This fuels the tuberculosis epidemic by giving the disease ample opportunities to spread and mutate into drug-resistant strains like the one that infected Nancy.

If the fuel is behavioral, then the solution should be as well.


Partway through her treatment, Nancy took part in a trial of a mobile phone platform that we — as part of a team of behavioral scientists working with a mobile health startup — developed to help people like her. Keheala — a homophone of the Hebrew word for community — is a platform designed to address the behavioral causes of not completing treatment by providing motivation and support throughout the arduous course of therapy.

Each morning, Keheala sent Nancy a text, asking her to log into the platform and verify that she’d taken her medications. On the days she’d forget to do that, the platform would send her more reminders — up to three in all — and if she still didn’t verify, she’d get a text or phone call from one of her “support sponsors,” checking in to see how she was doing. The first time she got such a call, Nancy told us later, she thought, “Wow, I wish I had this all along.” She said the calls made her feel like she wasn’t fighting TB on her own.

To make the platform as motivating as possible, we incorporated principles from behavioral science, the same principles that have been successfully employed to promote charitable giving, volunteering, and resource conservation. It’s why, for instance, we decided against a one-way system that just sent people with tuberculosis a reminder each day in favor of a two-way system that requires them to verify they have taken their medications. This approach ensures adherence and eliminates plausible excuses like, “I didn’t get the text,” or “I got the text but put the phone down and forgot about it.”

To make Keheala accessible, we built it to work on the feature phones (what some people call dumb phones) that are ubiquitous in areas where TB is common, like sub-Saharan Africa, Southeast Asia, and Eastern Europe. These phones lack the functionality of smartphones but can be used to make calls, send text messages, and access some simple internet features through a text-based interface. Using Keheala requires no downloads or previous experience.

The trial Nancy is participating in is still underway, but we recently published with several colleagues a letter in the New England Journal of Medicine summarizing the results of a previous study of interactive messaging with mobile phones that included slightly more than 1,100 people with tuberculosis in Nairobi. In that trial, only 4.2% of people who received messages failed to complete their courses of treatment, two-thirds lower than the 13.1% who didn’t receive messages. The percentage in the messaging group was well within the target the World Health Organization hopes to reach by 2035. This is, to our knowledge, the largest impact of any such intervention to promote the completion of tuberculosis treatment to date.

Nancy successfully finished her treatment. Today she’s healthy, spending as much time as she can with her family and friends. She’s also an advocate for TB awareness, participating in activities in Thika and occasionally giving talks about her experience.

This work shows what people can achieve with some carefully designed motivation and support, and it suggests the power of behavioral science for helping with that design. It also shows what can be done right now, without waiting for the development of faster treatments or for countries to reform their existing health care systems — efforts that currently receive top billing (and funding) in the medical community.

With millions of dollars being spent on medical solutions, our results suggest that we should pause and consider allocating more of that budget to behavioral solutions, so all that good medical research doesn’t go to waste.

Erez Yoeli is a research scientist at the MIT Sloan School of Management and co-director of MIT’s Applied Cooperation Team. Jon Rathauser is the founder and CEO of Keheala. David Rand is associate professor of management science and brain and cognitive sciences at the Sloan School, director of MIT’s Human Cooperation Laboratory and co-director of the Applied Cooperation Team.

Comments are closed.