Branching Out: Annie Hartley Explores D-tree Collaborations in Zanzibar

After 30 hours of transit, Annie Hartley has arrived on the archipelago of Zanzibar just off the east coast of Tanzania.

Born and raised in South Africa, Hartley is an Assistant Professor of Biomedical Informatics & Data Science at Yale School of Medicine. After completing her undergraduate degrees in Pretoria and Cape Town, she worked in Sierra Leone during the Ebola epidemic, studied at the London School of Hygiene and Tropical Medicine, and lived for several years in Switzerland. These experiences have taught her to value global approaches to medical and scientific research. Hartley's lab, LiGHT (Laboratory for Intelligent Global Health Technologies) puts these approaches into practice as a "global lab for global impact."

Hartley also serves in a voluntary capacity on the board of a global health nonprofit D-tree, a role that has brought her to Zanzibar to conduct a field visit and explore potential partnerships with researchers, students, and health workers. D-tree augments health systems with digital technologies that support improved (D)ecision-making deep within the branches of the communities they serve. It's an exciting time for D-Tree, as they have recently scaled a solution for women's and children's health across the archipelago of over 2 million residents—a project robustly supported by Zanzibar's forward-thinking Ministry of Health. “This is such an inspiring success story of an integrated approach to global health interventions, that it has to be shared with our students!” says Hartley. She is spending two days with Peace Mathew Kipuyo, a videographer based in Tanzania, to film an video highlighting D-tree's partnership with local community health workers. Up next is a meeting with workers who have just left the office of the Sheha—a local leader who guides community health initiatives.

D-tree was founded in 2004 by renowned pediatrician Marc Mitchell, who believed in empowering communities to make better decisions with "the right information, at the right time, in the right place," Hartley explains. Since then, it has launched a volley of efforts to increase access to quality healthcare in Tanzania and other neighboring countries. D-tree’s flagship program, Jamii ni Afya, formalized the role of community health workers in Zanzibar as part of the first-ever community health program to be scaled nationally through a state healthcare system. This work could not be more urgent: the World Health Organization has forecast a shortage of 10 million health workers by 2030, a deficit researchers warn will disproportionately impact the global south. Through Jamii ni Afya, D-tree has trained health workers to use a new digital system that standardizes care and connects clinicians and patients to primary care facilities. Partnering with the Ministry of Health in Zanzibar, this program reached national scale in 2021, and enrollment now encompasses 90% of the national population. D-tree has conducted over 3 million healthcare visits and trained over 10,000 community health workers.

"D-tree is so well accepted by the community because it is part of the community," Hartley says. "Every day, thousands of community health volunteers across Tanzania, Zambia, and Zanzibar go door-to-door to help patients make better decisions about their health." They provide validated health advice, remind parents and mothers-to-be about their check-up and vaccination appointments, and are trained to recognize and respond to emergencies. As a data scientist, Hartley believes that the massive datasets emerging from programs like Jamii ni Afya represent an opportunity for research that could further improve the decision-making power in communities through data-informed predictions. In other words, D-tree’s extraordinary work could be supercharged through the power of academia. Researchers could respond to the needs identified by the community (for example, a need for an algorithm that better identifies pregnancies at risk of poor outcomes), and integrate their solutions within the extraordinary D-Tree network. "It's a win-win," says Hartley. "The community benefits from the rigor of academic approaches to responsible evidence-based implementation—and the researchers will jump at the chance to respond to the needs of the community."

"This kind of collaboration has already begun," says Riccardo Lampareillo, CEO of D-tree. In 2022, D-tree partnered with researchers from Harvard School of Medicine to investigate if machine learning could predict whether a pregnant person is likely to deliver their child in a health facility or at home. “Because the researchers don’t have to bear the enormous costs of setting up the training and network, [these partnerships] optimize funding to maximize impact for everyone,” Lampariello explains.

At the same time, it is important to understand the social, economic, and cultural contexts from which the data originates. Lampareillo emphasizes the importance of developing digital health technology for under-resourced communities. "We always act on behalf of those left behind, and in this case, those who are underrepresented in AI systems," he says. "It is our priority to advocate for their inclusion, to ensure access not only to quality of care but also to technology, such as digital health. Technology can narrow or exacerbate the equity gap: its governance is key."

For Hartley, this means the creation of equitable and responsive partnerships, where resources are pooled around the need and the impact rather than a specific institute. At times, global health research and NGOs more generally have faced criticism for being one-sided, patronizing or even extractive. Phenomena like philanthrocapitalism and 'trauma tourism' complicate the systems within which organizations operate in the global south. "Bilateral knowledge exchange is key to ensuring a needs-based and impact-centered design of technology," Hartley says. An exchange program, where students from East Africa can swap places with students at Yale, would enact these principles of reciprocity, Hartley believes. "These kinds of visits allow us to appreciate the needs and challenges of the end users and form meaningful long-term partnerships," she says.

Kick starting this initiative this summer, four students from East Africa will take part in a 3-month internship at Yale School of Medicine's Section of Biomedical Informatics & Data Science, where they will perform "AI4health" research. The video that was filmed with health workers in Zanzibar will also be incorporated into Yale’s online Certificate in Medical Software and Medical AI, which is launching in March 2024. Students enrolled from around the world will be briefly immersed in the real-life applications of digital health technology. And—Hartley hopes—this might just inspire them to go further with their research and join D-tree’s vision of quality healthcare for all.

Watch the video below: