The case for funding psychotherapy in the developing world

It’s hard enough to access mental health care in rich countries. Now try it in a poor country.

When people think of ways to help the world’s poor, a few obvious ideas come to mind: giving them cash; preventing diseases like malaria through the distribution of bed nets and pills; treating HIV/AIDS in areas ravaged by those conditions; and other tactics that take aim at economic privation and infectious diseases.

That focus is understandable and necessary — but what if it elides a different way of thinking about easing suffering in the world? What if there was a real opportunity to improve the lives of low-income people by devoting resources toward their mental well-being, too?

A new report raises that intriguing prospect. Written by Michael Plant, Joel McGuire, and Barry Grimes of the Happier Lives Institute, a research center that aims to find evidence-based ways to improve happiness worldwide, the study looks at the role therapy — particularly cognitive behavioral therapy (CBT) — can play in improving lives in the developing world.

CBT has become the dominant paradigm in talk therapy in recent decades, and in that time a staggering body of evidence has grown showing its effectiveness in improving subjective well-being. (As someone who has had CBT treatment on and off for years, I can attest to this.) The idea behind CBT is to train people to identify and combat unhelpful patterns of thought or behavior, and to learn more useful patterns that can help them avoid anxiety, depression, and other problems.

To date, global health efforts have mostly focused on illnesses of the body: malaria, vitamin deficiency, HIV/AIDS prevention, tuberculosis. Obviously, such diseases can affect the mind, and canonically “mental” illnesses like depression can take a physical toll. But historically, mental well-being has simply never gotten equal billing. Until 2015, the UN’s Sustainable Development Goals didn’t even include benchmarks for mental health, even as they focused heavily on infectious diseases and markers of physical health.

Through the increased use of tools like randomized controlled trials, policymakers have gotten better at understanding what really works in raising incomes and treating diseases among the world’s poorest people, and what doesn’t. That’s great, but it also may have led to some complacency — the idea that we already know what works.

The world’s a vast place, and there are surely dozens or hundreds of other interventions that could work and aren’t being adequately funded. Organizations that focus on mental health could be among them.

Cognitive behavioral therapy, briefly explained

But first, what exactly is cognitive behavioral therapy?

You might not have gone to therapy, but you probably have heard variations on the following:

• Pay attention to your thought patterns, especially when you might be engaging in “cognitive distortions.”
• Recognize when you jump right to worst-case scenarios, and remind yourself that the worst-case scenario might not happen, and you can survive it if it does.
• Identify fears you have, and try to expose yourself to the experiences you fear so they lose their power.
• Notice when you are filtering out positive feedback and only recognizing negative feedback, and remember the former should count just as much as the latter.
• Try journaling so you can recognize your own thought patterns, and notice when they’re serving you well and when they aren’t.

This litany of advice and others like it are the bread and butter of CBT.

Among the reasons CBT has become so dominant is that a formidable evidence base has been amassed attesting to its effectiveness — not just in combating depression and anxiety, but also certain other goals, such as preventing violence among at-risk teens.

The evidence base is so large, a recent paper by Oxford researchers reviewing it states that “the majority of psychological treatment research is dedicated to investigating the effectiveness of cognitive behavioral therapy (CBT) across different conditions, population and contexts.” (Emphasis mine.) That is, most research on psychological treatments is about this one technique, hammering out how well it works, to treat what, among whom.

In general, as that paper and others like it conclude, the research suggests that CBT works fairly well, but that it’s not a miracle cure. We still don’t know precisely how well it works for everything (or if it’s less effective now than it was when it was invented). But the basic idea of teaching people to observe their own thoughts at a distance, and learn to recognize and respond when the thoughts they think automatically are hurting them, is powerful and effective.

“I think of it as one of the greatest inventions of the 20th century,” Chris Blattman, a professor of economics and political science at University of Chicago who’s tested CBT as a therapy for young men in Liberia with criminal backgrounds, told me. “A lot of CBT is just habits for overcoming automatic thoughts and behaviors that we all would love to avoid, everyday anxiety or anger or stress. We can all learn something from that.”

But CBT might be especially useful in a developing country context. Angela Ofori-Atta, a professor of psychiatry at the University of Ghana and founder of Psych Corps, a CBT program based in Ghana, said in an email, “Our goal was to build resilience among people who were at risk for more knocks from life than usual. We reckoned that people who were poor and had few buffers would need extra skills to deal with unfortunate events such as poor harvests, loss of jobs, death in the family and so on.”

How to measure the happiness impact of charity

The takeaway of the new Happier Lives Institute report is simple: It makes the case that charities and foreign aid agencies, especially those concerned with maximizing their impact per dollar, should consider prioritizing expanded access to CBT.

The study focused on StrongMinds, which works on improving mental health care in low-income countries in sub-Saharan Africa, specifically among women. It’s existed since 2013, and was inspired by a promising randomized controlled trial from about a decade earlier evaluating a group therapy program in Uganda.

That study had participants meet for 90 minutes each week, for 16 weeks, in groups of 8-10 people to talk about their depressive symptoms, and then (in typical CBT fashion) to recount the past week’s events and try to see how their symptoms related to those events and their thought patterns. Researchers found a substantial reduction in severe depression among the people treated, and a follow-up analysis found that the benefits persisted six months later.

The Happier Lives Institute report compares the effectiveness of Strong Minds to a foreign aid intervention with lots of evidence behind it: just giving out cash. In global development, cash transfers have emerged as something of a yardstick for evaluating many other interventions. Cash seems to help, at least a little bit, with just about everything, whether it’s promoting healthy behaviors, or reducing hunger and malnutrition, or building up assets like homes and livestock. So increasingly, development economists have adopted an approach called “cash benchmarking.” To test how good an intervention is, they compare it to just giving out the equivalent amount in cash.

The report specifically compares cash versus CBT when it comes to “subjective well-being.” It measures that in several subtly different ways. The studies of GiveDirectly’s and similar cash programs, for instance, often include survey questions that measure both happiness in the moment — “how do you feel right now/today?” — and studies of what psychologists call “life satisfaction,” or how well the respondent thinks their life is going all-told, apart from their moment-to-moment feelings.

In the studies of StrongMinds, by contrast, “subjective well-being” was measured through the prevalence of symptoms of happiness-related disorders like depression.

The Happier Life Institute researchers attempt to aggregate and compare the effects of cash and therapy across these different measures, to get an estimate of their effect on “subjective well-being” in general.

“There is quite a bit of precedent for aggregating various psychological measures, including by some Nobel prize winners,” Plant, cofounder of the Happier Lives Institute and a coauthor of the analysis, told me in an email. “This seems reasonable to us: mental health measures tend to have lots of questions about psychological states, so are picking up the same sorts of things.”

Chris Madden/Getty Images

To compare apples to apples, the Happier Life Institute researchers estimated changes in well-being in terms of standard deviations (SDs), which is common in evidence reviews like this one.

This can get wonky, so let me break this down: How to interpret the size of the effect of a given intervention is a huge and contentious question, but a common, very rough rule of thumb devised by the late statistician Jacob Cohen is that an effect size of 0.2 is small, 0.5 is medium, and 0.8 is large.

A lump-sum cash transfer from GiveDirectly of $1,000, the analysis concludes, increases well-being by about one standard deviation — 0.92 — which is a large effect. That makes sense: An extra $1,000 in cash in a rural village in the Lakes region of Africa might double your income, so it would be surprising if it didn’t tend to make you happier. A monthly cash transfer of a smaller amount has a smaller effect on happiness: about 0.4 SDs per $1,000 spent.

In its direct comparison between the GiveDirectly and StrongMinds programs, the HLI report estimates that a StrongMinds intervention costs about $128 per person, and increases subjective well-being by about 1.7 SDs. So its effect size, per $1,000, is an astonishing 11.8 SDs, because $1,000 can fund therapy for about seven people, each of whom gets a positive effect of 1.7 SDs.

That figure — 11.8 SDs per $1,000, compared to 0.92 for cash transfers — is a massive effect, if it holds up. The report does the division and concludes that therapy is likely around 12 times more effective, in terms of improving human subjective well-being, than cash transfers.

I reached out to Joe Huston, managing director of GiveDirectly, for a friendly skeptic’s view on these results. He was mostly pleased that Happier Lives Institute did cash benchmarking: Getting other groups to compare programs to cash is a key goal of GiveDirectly.

But he worried that the analysis undercounted the effect of cash on happiness by ignoring how cash can make not just its recipient, but the recipient’s family and the broader community happier. “In most cases I think you should also expect to see large, positive intra-community spillovers,” Huston said in an email.

Should promoting happiness be the goal of development?

The bigger, more philosophical question the research raises is: How important is subjective well-being relative to everything else that development agencies and global health charities want to promote?

Charities like GiveDirectly, StrongMinds, and larger institutions like Doctors Without Borders and UNICEF target a variety of different metrics in their work. They try to reduce deaths among children under the age of 5. They try to reduce rates of infection from specific diseases like HIV or malaria. They target economic outcomes, like income or consumption.

Making comparisons across these different goals is incredibly hard. How much better is it to save a child’s life than to double their parent’s income? How valuable is preventing a lifelong worm infection, compared to reducing vitamin A deficiency? GiveWell, the charity evaluator that attempts to find the most effective per-dollar global charities, has spreadsheets where staff spell out different views on these trade-offs, and see how different ways of comparing causes affect their judgment on what charities to recommend. (Disclosure: I donate to GiveWell’s top charities.)

Part of what the Happier Lives team is trying to do is submit subjective well-being as a common currency in comparing development programs. And there’s a long philosophical tradition, going back at least to Jeremy Bentham and the early English utilitarians, arguing that happiness or some other measure of well-being should be our common currency in settling moral questions. It provides a way to compare between money and years of life, between preventing sickness and enjoying better health, between all manner of usually incomparable goods.

HLI is quite explicit that this is their goal. “In order to do as much good as possible, we need to compare how much good different things do in a single ‘currency,’” they write in their report. “[W]e believe the best approach is to measure the effects of different interventions in terms of ‘units’ of subjective well-being.”

I find much of this view attractive, personally — but it’s profoundly arguable. Plato famously wrote in Philebus that if you lived life merely to maximize pleasure, “your life would be the life, not of a man, but of an oyster.”

And even if one did think the life of an oyster is a fine life, questions would remain about how well humans are currently able to measure subjective well-being. The proper way to ask about happiness or life satisfaction remains profoundly controversial within psychology and economics, as does the relative importance of those two things if they differ from each other.

Is it more important to ask about people’s mood in the moment or over the past week, compared to asking how well they think their life is going overall? How meaningful are answers to these questions? How do they compare to measures of illnesses like depression, which HLI includes in its analysis as another measure of well-being?

These may seem like nitpicks, but when you’re allocating scarce dollars among charities, getting these details right can be crucial.

Whether or not CBT is the best health program to fund in poor countries, or even better than giving out cash, there seems to be a strong case that it should be on the menu. Plant estimates that StrongMinds has the capacity to spend about $6.6 million more next year, if enough people donate, and more like $30 million over the next three years; that seems like a reasonable amount to complement the billions spent on the tried-and-true targets of development spending every year.

I’ve certainly benefited from CBT. People around the world could benefit, too.