As part of the Edge 50 list, Apolitical is publishing a series of articles exploring a selection of these bold initiatives to learn more about the stories and strategies behind them.

Created by Apolitical in partnership with the Mohammed Bin Rashid Centre for Government Innovation, this list is an invitation to look beyond the usual playbook and ask what becomes possible when governments are willing to think boldly about today's biggest problems.

Apolitical spoke to Dr Esther Tumbare, Chief Executive Officer of Friendship Bench Zimbabwe, an NGO that trains grandmothers to provide basic therapy on wooden benches in local neighbourhoods.

Zimbabwe has one mental health professional for every one million people. Among them was Professor Dixon Chibanda, a psychiatrist and founder of Friendship Bench, whose experience of losing a patient to suicide because her family couldn't afford transport to the hospital prompted him to ask a different question: what if communities themselves could provide care? His answer was to train grandmothers to deliver problem-solving therapy in everyday spaces where people already feel safe. The results have been striking: participants report a 78% reduction in depression and suicidal ideation. The model has now reached over one million people across 12 countries.

What challenge was Friendship Bench originally trying to address, and how did your understanding of that challenge change once the work began?

Friendship Bench was trying to address a very practical challenge: the country could not meet the treatment gap for mental health conditions. In Zimbabwe, there were very, very few psychiatrists and other trained professionals available to provide mental health services for very common conditions such as depression and anxiety.

As the initiative evolved, the question became not just about having professionals to provide services, but also about making those services easily accessible to the communities being served. These services needed to be delivered in ways that were culturally sensitive, that took language and local context into account, and that created an environment where people felt comfortable coming forward to talk about the issues they were experiencing.

Often, when we think about 'innovation', we think about cutting-edge technologies and AI. This approach is the opposite; it's about speaking to real humans on a bench. What inspired you to develop a solution like this?

The founder of the programme is a psychiatrist from Zimbabwe, Professor Dixon Chibanda, who was working in a government hospital. He treated a young woman with depression who lived far from the capital. Her family couldn't afford transport for her to return to the hospital for follow-up care, and tragically, she died by suicide.

That prompted Dixon to question how local communities could provide support beyond doctors and nurses.

What are some assumptions about how governments or public services usually work that Friendship Bench had to question?

The first assumption was that mental health services can only be provided by professionals who hold medical degrees or diplomas. Another was that if we were to introduce an innovation to address this problem, it would have to operate as a parallel, separate system rather than being integrated into existing systems.

So we knew we had to challenge both the idea that only doctors could provide this kind of care, and also the idea that any new solution couldn't make use of resources already available in communities.

Did you face resistance or challenges?

Yes, there was a lot of resistance. Even today, when the model is introduced in places where it isn't well known, people question whether lay individuals can really provide mental health support. As I said, many people believe that this type of care can only be delivered by trained professionals. We also saw a lot of resistance to the idea that you can provide support outside clinics and hospitals. People strongly believe these services can only be provided there.

We challenged this resistance by conducting research and generating evidence to demonstrate that the opposite was true. We saw that, actually, if you identify trusted community members, in our case, grandmothers, who are respected members of the community with lived experience, they can actually be trained using a structured curriculum to provide problem-solving therapy.

Professor Dixon Chibanda conducted randomised clinical trials to evaluate whether these trained lay people could deliver the same outcomes as medical professionals. The results were striking: participants reported a 78% reduction in depression and suicidal ideation. We saw that quality of care was maintained and, in some cases, even improved. People actually felt more comfortable speaking with individuals who were not wearing nurses' uniforms or doctors' coats. As long as the grandmothers were properly trained, the quality of care remained strong.

These conversations can happen in any safe space. In our case, it was on a bench, which could be outside a health facility, but could also be in a church, on a football field, under a tree, sitting on a rock, or at a school. The 'bench' is symbolic. It represents any safe space where people can talk: under a tree, outside a church, at a school, or even by the roadside.

It's a really cost-effective innovation too, compared with training large numbers of doctors and psychiatrists.

Who needed to be involved in the project for it to work, and why?

The project required involvement at multiple levels. First, we needed community buy-in for it to work. The communities themselves had to accept the programme and take ownership of it. Second, we needed nurses to supervise the grandmothers and to strengthen referral pathways for any cases that require more structured or intensive care from medical professionals. Third, government involvement was essential to incorporate into policies and guidelines for the provision of care. And finally, of course, the people receiving the services themselves had to want to be participants.

What specifically makes grandmothers suited for this role?

Grandmothers are trusted members of the community. In many environments, especially in Zimbabwe, grandmothers are seen as people with lived experience who are easy to talk to. They're people you can open up to in confidence, feel listened to, and not judged. This makes them ideal service providers in our context. Friendship Bench began with fourteen grandmothers at a clinic in Harare and has since expanded across the country and internationally. Today, the programme has reached over one million people.

With so many people involved (communities, health systems, government and programme organisers), how were decision-making and responsibilities shared?

Decision-making was shared between the public health system and the communities themselves. There's always joint supportive supervision between us at the Friendship Bench, governments and nurses at the district level. Communities themselves also play a role in decision-making, especially through feedback mechanisms from beneficiaries and the grandmothers who provide the services. This allows us to constantly learn, adapt and ensure the services continue to meet community needs. So decision-making is co-owned by communities, alongside government and service providers.

We are now seeing more involvement from people with lived experience of mental health challenges, contributing towards how the services are provided. Our hope is for governments to become the primary implementers and funders of the programme at scale. At the same time, we continue providing technical assistance and generating evidence to support the work both nationally and globally.

How have you already seen Friendship Bench influence how governments work?

So far, it's shown governments that things can be done differently, that communities themselves can play an important role in delivering services to other community members. It's demonstrated that task-shifting, when done properly with training, supervision, strong monitoring systems and confidentiality, can work effectively. It's also shown to governments that mental health services don't have to be delivered exclusively in clinics or hospitals. They can be provided in communities and still significantly reduce depression, anxiety and other mental health conditions.

The model has now been adopted in several countries. Which elements are more universal, and which need to adapt to local contexts?

Several elements are universal: task-shifting, community-based care, and data collection for evidence generation. However, the identity of the service provider might vary. In Zimbabwe, grandmothers were the most trusted, non-judgemental group. In other places, for example, it might be peers supporting peers or men supporting other men.

Language and cultural context also matter. In Zimbabwe, we found that clinical terms, like depression and schizophrenia, didn't encourage people to come forward. Instead, we used simple language communities related to, like a word that means overthinking. Everyone overthinks sometimes, and everyone gets stressed sometimes, so it feels more acceptable. This sort of thing would need to be adjusted depending on the community you're working in.

Looking back, what has surprised you the most about the impact of the project?

When we first started, we didn't anticipate how much Friendship Bench would strengthen community cohesion and resilience. The programme not only benefits individuals but brings communities together. The grandmothers themselves also benefit. Many of them say they've found a renewed sense of purpose and importance. You might wonder if they feel overwhelmed, constantly hearing about people's challenges, but actually, they often form their own support groups to talk about how they're feeling, which is nice to see. They chat about difficult cases too, confidentially, of course. This powerful coming together of communities is something we didn't anticipate.

Another surprise has been the global interest in what began as a local solution designed to help Zimbabwean communities. It's proven to be really adaptable.

What advice would you give to public servants who want to replicate something similar?

"First, listen to your communities. Meet them where they are and find out what their needs are. Too often, we design solutions without involving the people they are meant to serve, but they know what's best for them.

Second, work with government. If you're going to scale services and want to reach everyone, it's always better to look at the systems and institutions that are already available in your context and try to work with them. Third, continuously learn. Have strong monitoring and evaluation systems so you can understand the impact of what you're doing and who you're reaching. For example, are there any subpopulations you're leaving behind? Is there anything changing within the context you're working in that might affect the work you're doing? Continuously learning about these things is key to remaining relevant and effective.

Lastly, what has been the most rewarding part of being involved in this work?

Working with the grandmothers and seeing them in action. I love going out into the communities, talking to people who have benefitted from these services and hearing their stories. People often say they were on the brink of ending their lives until they sat on a bench and spoke with a grandmother.

We also run support groups, so the people who sit on benches can provide ongoing support to each other. Many people come to our benches because they are worried about providing for their families, so we also run income-generating projects that support people in activities like keeping pigs, selling pork, gardening, and crocheting or knitting, which help them generate income. But, beyond that, the most important benefit people describe is a sense of belonging. They feel supported and less alone.

In a world where loneliness is increasing, that sense of connection is incredibly meaningful. I love my job because of that.