Lab Researcher Spotlight: Julia R Pozuelo

Julia R. Pozuelo, PhD, is a Postdoctoral Research Fellow in the Department of Global Health and Social Medicine at Harvard Medical School. Julia’s research interests lie at the intersection of psychiatry and development economics. Her work focuses on developing and implementing cost-effective and scalable interventions to treat depression in low-resource settings.  Her recent work leverages AI to optimize treatment selection and improve therapy quality.

Let’s hear from Julia about her career, interdisciplinary work in mental health, and advice for the next generation of researchers.  

Q&A With Julia R. Pozuelo

Path, Projects, and Impact of Mental Health Research

Q: Can you tell us about your background and what inspired you to pursue mental health research?

I studied economics during my undergraduate and master’s degrees before pivoting to a PhD in psychiatry at the University of Oxford. This trajectory has led me to work at the intersection of both fields, where I focus on creating affordable treatments for depression in settings where resources are scarce. The interventions I develop include training non-specialists to deliver care, using digital tools to extend access, and more recently, leveraging AI to improve therapy quality. Given my background in development economics, I have prioritized working in low-income settings, applying these approaches across diverse contexts in Latin America, Sub-Saharan Africa, and South Asia.

My interest in mental health began while working on a digital mental health intervention just after my master’s degree. I initially joined the economics team to work on data analysis and measurement, but I soon became interested in two things: the knock-on effects that depression had on outcomes I cared about (such as education, income, decision-making, and risk-taking) and the question of how treatments need to be adapted to different cultural contexts to be effective. I then realized that very few economists were focusing on mental health at the time, despite its profound influence. That insight, together with my training and guidance from mentors like Profs. Alan Stein and Vikram Patel inspired me to build a career in global mental health.  

Q: What was it like transitioning from economics to mental health research?

At first, it was challenging. I remember going to psychiatry seminars and feeling overwhelmed by the number of acronyms and technical language. But it was also exciting, as I could see an opportunity to learn and grow. Over time (and maybe with age!), I’ve become more comfortable knowing what I need to learn, and what I can leave aside as outside my expertise. I also began to see complementarities between the fields. For example, economists, with their focus on establishing causality, can bring rigorous methodology to psychiatry. And psychiatry, in turn, brings rich clinical insights and direct patient engagement that help ground economic models in lived human experience. Global mental health, in particular, invites this kind of interdisciplinary collaboration. The field is defined by complex problems that no single discipline can solve alone. That openness has made it a very exciting and welcoming space for me to combine my economics background with psychiatry.

Q: Can you tell us about your role with the Mental Health for All Lab and the focus of your research?

I am currently a postdoctoral research fellow, and I dedicate most of my time to the OptimizeD trial in India. This is a large precision mental health trial based in primary healthcare clinics that aims to predict whether patients with depression will respond better to psychotherapy or to antidepressants. This question is important because, while effective treatments for depression exist, patients differ widely in how they respond. As a result, many people undergo a difficult, costly, and inefficient trial-and-error process before finding a treatment that works for them. If we could identify the optimal treatment from the outset, patients would recover faster, with less suffering, and in a far more cost-effective way.

Beyond OptimizeD, my research spans three main areas:

• Global epidemiology of mental health: I work with large-scale surveys to estimate how many people live with mental illness, and to identify treatment gaps and barriers to care. I have worked on nationally representative surveys among refugees in East Africa and am now collaborating with Prof. Ron Kessler on a series of papers that rely on data from the World Mental Health Survey.
• Scalable interventions: I develop and evaluate cost-effective and scalable interventions, including app-based therapies in South Africa and Uganda, a chatbot to support patient engagement with treatment, and task-sharing with non-specialists in areas where clinicians are unavailable.
• Optimizing treatment: I apply AI and machine learning methods to understand what works best for whom, and develop AI tools to assess therapy quality and support supervision so that effective treatments can be delivered at scale.
   

Q: Could you introduce the methodologies and significant findings from the OptimizeD study?

The OptimizeD trial is a 4-year study, and we are now entering the final year. Our target sample size is 1,500 participants, and we have recruited ~1,200 participants to date. The plan is to use data from the first ~1,000 participants to develop an algorithm that predicts the optimal treatment for a given patient based on a wide range of baseline characteristics. We will then validate the algorithm using data from the final 500 participants.

While it is too early to say whether the algorithm will succeed in predicting what works for whom, we can already see that the treatments themselves are effective. So far, we are observing remission rates of around 50% and response rates of around 70%. These early results are encouraging and show that providing evidence-based treatments for depression in primary care clinics is possible and effective.  

Q: Would you like to share some success stories in your work?

One of the most exciting aspects of my work has been seeing people benefit from treatments; for example, the improvements we are already observing among participants in the OptimizeD trial.

Another highlight has been developing interventions that are engaging and useful, such as a gamified mental health app for adolescents in rural South Africa and Uganda. Using a user-centered design approach, we achieved very high engagement, with most participants completing treatment and one-third continuing beyond the trial – far above typical digital mental health tools and showing real promise for integrating digital mental health into rural care.

More recently, one of my projects was selected for the MEXA Generative AI for Mental Health Research Accelerator, funded by the Wellcome Trust. This accelerator phase is allowing us to refine an idea for an AI tool that could provide feedback to non-specialist counselors in India, with the aim of helping them deliver higher-quality therapy at scale.

Q: What are some of the biggest challenges you face in your research?

Sustainable funding is a constant challenge. Grants are often short-term and fragmented, which means each project builds its own infrastructure from scratch, and when the funding ends, that infrastructure often disappears, even if the science or partnerships are still needed.  

Another challenge is the time it takes to run trials. Our India study, for instance, will take five years before we can draw any conclusions. This creates pressure to diversify across multiple projects, not only as a safeguard in case one doesn’t work out, but also to maintain the steady flow of publications needed to progress in my career.

Finally, there’s the issue of impact and communication. Academic incentives reward publishing papers, but these are rarely picked up by policymakers, and even when they are, they’re often full of jargon. Very few grants provide resources to translate research into concrete policies or practice, so the potential impact is often lost.

Q: Where do you see the future of global mental health research headed?

I think AI will play a central role in closing the treatment gap. One promising use is to strengthen the mental health workforce. AI is already being applied to scale training for non-specialists and support their supervision. Looking ahead, large language models could help monitor therapy quality in real time and provide tailored feedback, opening up new possibilities for making effective care more accessible and scalable.  

AI could also be used directly to support patients. While I don’t think it will ever replace the skills or human connection of a clinician, the reality is that in many parts of the world there simply are no clinicians. In those contexts, AI could provide supplemental care or structured support where otherwise there would be none.

That said, much of the innovation so far has come from high-income countries and in English-language contexts. An exciting frontier for the field will be adapting and testing these tools in low-resource settings and diverse languages, where the need is greatest and where AI could make the biggest difference.

Personal Insights

Q: What keeps you motivated in your research work, given the long turnaround time in your trials?

What keeps me going is knowing that I’m working on a problem that is both important and often overlooked. Depression is very common in low-income settings, yet most people who need treatment never receive it. At the same time, we now have interventions and tools that could help close this gap, making it not just an urgent problem, but one we can do something about.  

Overall, I stay motivated by knowing that I’m spending my days on work that matters, and by seeing that (even at a small scale) participants in my studies are already showing improvements in their symptoms. It’s also inspiring to think that the solutions we’re developing could one day reach millions of people who currently have no access to care.
 

Q: How do you balance your time as a researcher？

Honestly, it’s challenging, especially when working with teams across the world in very different time zones, which makes it hard to keep clear boundaries between work and personal time. I try to prioritize carefully and make sure I get enough sleep, exercise, and fun time too (otherwise I’m useless!)  

What really makes the difference, though, is having a strong support system. On a personal level, I rely on family and friends to keep me grounded. Professionally, I’ve been fortunate to work with brilliant colleagues in India, South Africa, Uganda, the UK, and the US who have my back when I’m overstretched. Knowing I’m not doing this work alone makes it much easier to find balance and keep going.

Q: Do you have any advice for aspiring mental health researchers?

I would start by encouraging them to spend time thinking carefully about the problem they want to focus on. It’s tempting to say yes to every opportunity (I’ve been guilty of this myself!), but you can’t do everything well. It’s better to choose one problem with the potential for big impact and do it as well as you can.

I also recommend surrounding yourself with good collaborators and building strong partnerships. My own work wouldn’t exist without collaborators at Sangath (India), MRC/Wits Agincourt Unit (South Africa), BRAC (Uganda), Oxford, and many others. Research in global mental health is always a team effort.

Finally, be creative in how you build your career – there are many ways to do research that matters. My move from economics into psychiatry was sometimes met with skepticism, but it also brought critical insights into my work. Diverse experiences can be a strength.  

Julia’s Contacts

Mental health research thrives when we combine diverse expertise with deep community partnerships. There are many opportunities to get involved with Julia’s current and upcoming work, and she would love to hop on a call to discuss future synergies. If you’re interested in collaborating with Julia, please reach out at  Julia_RuizPozuelo@hms.harvard.edu.