By: Nayantara Arora (DPhil in Clinical Medicine, 2024)
Ex Aula Prize 2026 Runner Up
Mental health conditions such as depression and anxiety are leading causes of disability worldwide, yet are difficult to systematically measure in many low- and middle-income countries.
National health systems routinely collect data on infectious diseases, maternal health, and child mortality, and have done so for several decades, but mental health has often been absent from large-scale surveys and public health monitoring systems. As a result, policymakers lack evidence needed to identify vulnerable or at-risk populations, allocate resources, and design appropriate interventions. My research explores how mental health can be measured more effectively in sub-Saharan Africa, and how social and structural inequalities influence mental health outcomes.
My work probes three interrelated questions: how do we measure mental health, who experiences the greatest burden of mental ill-health, and why some groups are more vulnerable than others? To address these questions, I have combined evidence from systematic reviews with analyses of nationally representative survey data.
The first part of my research investigates whether one of the world’s most widely used depression screening tools, the Patient Health Questionnaire-9 (PHQ-9), performs reliably in sub-Saharan Africa. Although the PHQ-9 is used extensively in clinical practice and research, it was primarily developed and validated outside the region. Through a meta-analysis of 59 studies, I evaluated its reliability and diagnostic performance in diverse populations and settings, including different countries, urban vs. rural settings, and patient groups. The results suggested that the PHQ-9 performs well, demonstrating good mathematical reliability and accuracy. However, its performance varied across contexts, including between clinical and community-based populations. These findings illustrated the importance of validating screening tools locally, rather than assuming that instruments developed in one setting will perform identically elsewhere, and set the precedent for my use of the PHQ-9 to capture depressive symptoms later in my research.
Having examined the validity of mental health measurement, I then turned to the question of who is most affected. Using data from the 2022-23 Mozambique Demographic and Health Survey, one of the first nationally representative surveys in the region to include detailed mental health measures, I analysed patterns of depression, anxiety, and suicidal ideation among more than 9,000 women. The survey included PHQ-9 responses to capture depression, the Generalized Anxiety Disorder scale (GAD-7) to measure anxiety, and four questions regarding one’s plans and/or attempts to commit suicide.
I focused particularly on women living with HIV because they experience an unfortunate convergence of health and social challenges. HIV can affect mental health directly through its impact on physical well-being, but it can also expose individuals to stigma, discrimination, and economic insecurity. In my analyses, women living with HIV reported higher levels of psychological distress than women without HIV. Poor mental health was also associated with socioeconomic disadvantage, poorer self-reported health, and a history of sexually transmitted infections, suggesting that mental health outcomes are linked to patterns of inequality.
Intrigued to discover further potential mechanisms underlying these disparities, I chose to delve into HIV-related stigma. Using survey data from nearly 1,000 women living with HIV in Mozambique, I examined several forms of stigma, including internalised stigma, such as feelings of shame about one’s HIV status, and experienced or external stigma, such as discrimination or harassment from others. The findings revealed that different forms of stigma were associated with different mental health outcomes. Internalised stigma was strongly associated with symptoms of depression and anxiety, even after accounting for socioeconomic factors and healthcare access. Experienced stigma, by contrast, was more closely associated with suicidal ideation. These results suggest that stigma is a complex social process that can affect mental health and how it is experienced through multiple pathways. For example, could it be that external stigma drives individuals to acute, extreme mental distress, whereas internalized stigma chronically sits within individuals in the form of depression and anxiety?
The overarching priority of my research is to understand the relationship between measurement and visibility. Alongside these empirical studies, I have examined how mental health is incorporated into large international surveys such as the Demographic and Health Surveys. Different countries have adopted different approaches to measuring mental health, ranging from symptom-based screening scales (the PHQ-9, the GAD-7) to indicators focused on formal diagnosis and help-seeking behaviour. These choices influence the estimates researchers produce, as well as the questions that can be asked and the populations whose needs become visible. In this sense, measurement is not simply a technical issue; it shapes the evidence available to policymakers and influences how health priorities are defined.
These projects contribute to a burgeoning effort to strengthen mental health evidence in sub-Saharan Africa. Improving mental health care requires robust evidence, culturally appropriate measurement, expansion of treatment services, and acknowledgment of the social conditions that influence psychological well-being. Generating and synthesizing this evidence is a crucial step towards building more equitable and effective healthcare systems.