Professor of Economics and Finance, Frederic Esser Nemmers Chair, Co-Director of the Global Poverty Research Lab at Kellogg
Robert E. and Emily King Professor of Economics; Co-Director of the Global Poverty Research Lab
Chris Udry admits that he was quite skeptical when he and fellow Northwestern University professor Dean Karlan set out to test the impact of psychotherapy in poor communities in Ghana.
“I was convinced this was going to be a complete debacle,” says Udry, a professor of economics. “So we’re going to get very poor people together and talk about their problems and somehow change their lives?”
Still, the researchers, who co-direct Kellogg’s Global Poverty Research Lab, recognized that since mental-health care was largely missing in efforts to combat poverty, perhaps it was worth investigating. “So much of development policy focuses on job training and skills training and market access,” says Karlan, a professor of finance at Kellogg. “Mental health is thought of almost as a luxury problem that is just not dealt with in many developing-country contexts.”
With this in mind, the Kellogg researchers and collaborators Nathan Barker at Yale, Gharad Bryan at the London School of Economics, and Angela Ofori-Atta at the University of Ghana designed a study to implement cognitive-behavioral therapy (CBT) in low-income households in Ghana. CBT is an established psychotherapeutic approach to depression, anxiety, and other conditions that involves helping people recognize and address “cognitive distortions” that influence how they interpret and react to events, along with taking strategic actions to solve problems.
The conditions of chronic poverty, where people are “constantly presented with stimuli regarding their own low status,” may yield negative self-beliefs about one’s talent, worth, or future prospects, the researchers write. The potential results: poor mental and physical health, which could adversely affect economic outcomes. These sorts of compounding difficulties are exactly what those living in poverty don’t need.
To assess the impacts of CBT, the researchers ran a large-scale randomized controlled trial in rural Ghana, involving thousands of participants living in hundreds of communities. Survey-based measures showed that CBT improved both mental and self-reported physical health, as well as economic outcomes and cognitive skills.
Critically, the impacts occurred irrespective of whether a participant was identified as depressed before treatment began. This has important implications: even though CBT is designed for depression, it can also be beneficial for those struggling with the stress of living in poverty, which for some will lead to depression. “There’s a lot of movement in and out of depression when you’re living with dire trade-offs: What need are you going to put last? Is it going to be food? Is it going to be medicine? Is it going to be education? These are stressful, potentially depression-inducing, decisions to face,” Karlan says.
Those receiving CBT reported missing fewer days of work due to health issues than did the control group.
For those hoping to make an impact on mental health in these contexts, screening for poverty thus makes more sense than screening for depression. Otherwise, care providers and researchers might miss members of the population who would benefit from treatment, the researchers say.
The researchers worked in over 250 Ghanian communities, which they divided into either control communities or those receiving CBT.
In general, this is a population facing significant stress: at the study’s start, 55 percent of participants reported some form of psychological distress, with 15 percent reporting severe levels—much higher than rates observed in the U.S., the authors note.
The CBT was delivered in 2016 in 12 weekly 90-minute sessions to 10-person groups. Therapy was administered by recent college graduates hired and trained by Innovations for Poverty Action and the University of Ghana. The CBT leaders received two weeks of classroom training and completed one week of pilot CBT work before delivering the intervention.
The program included modules on promoting healthy thinking, solving problems at home and work, managing relationships, and goal setting. For example, participants learned how not to dwell on specific issues or catastrophize events by thinking, for instance, “A week without rain means all my crops will fail this year.” Participants also learned to recognize and mitigate “should statements,” such as “I should be doing better in life than I am.”
The researchers collected data before and after the interventions from individuals in both treatment and control villages via surveys that measured mental health, physical health, socioemotional skills, cognitive skills, and economic outcomes. They then compared results for the CBT group with those of the control group.
Watching the CBT sessions, Udry found himself rethinking his initial skepticism: “People were so interested in the training sessions. They loved it.” As an example, he observed that participants “seemed excited to talk about problems they had, and that other people in the group really seemed to be paying attention—they were really listening to each other.” Afterward, participants compared the experience to “talking with their pastor,” Udry says.
For instance, participants discussed challenges they faced with their children. “They talked about strategies for when their kids were not in school and didn’t have a job yet—like if a kid didn’t pass their high-school admissions exams,” Udry says. “They had a lively discussion and made a list of strategies on a whiteboard.”
The CBT group did better than the control group on multiple key measures.
For example, those receiving CBT reported missing fewer days of work due to health issues than did the control group. “That translates to real economic changes because of improved mental health,” Karlan says.
Additionally, those receiving CBT were 10 percent less likely to have any psychological distress and 24 percent less likely to have severe psychological distress than control-group peers. They also reported having 11 percent fewer days per month with poor mental health and 20 percent fewer days with poor physical health than the control group.
“We were so glad to see that people had lower stress levels and symptoms of depression,” Udry says. The study also revealed an unusually large improvement for the treatment group compared with the control group in socioemotional skills, such as those related to self-control.
And benefits of CBT went well beyond mental health and socioemotional functioning. For example, the intervention improved people’s cognitive performance as well, such as their performance on a memory test of 10-digit numbers. Importantly, the results for health and cognitive skills again did not differ by baseline mental health, suggesting the program’s effectiveness for those with and without diagnosable conditions.
Based on the results, Karlan and Udry conclude CBT works through two pathways in the communities that the intervention targeted: by reducing the likelihood of experiencing poor mental health and by directly improving bandwidth for cognitive tasks, including how to allocate mental resources toward solving specific problems, such as household finances.
The researchers plan to analyze more specific economic impacts of the intervention and to use measures of income, investment, and consumption collected during a follow-up survey.
Access to mental health therapy is not always at the center of antipoverty efforts. But the researchers say the results of this study are a call for a major shift in attitude. They conclude, “increasing access to mental health therapy in low-income countries should be seen as a core means of increasing human capital in the general population, with relevance far beyond treating those with a diagnosable mental health condition.”
Sachin Waikar is a freelance writer based in Evanston, Illinois.
Barker, Nathan, Bryan Gharad, Dean Karlan, Angela L. Ofori-Atta, and Christopher Udry. 2022. “Cognitive Behavioral Therapy among Ghana’s Rural Poor Is Effective Regardless of Baseline Mental Distress.” American Economic Review: Insights. ISSN 2640-2068. In Press.