Prevalence and Sociodemographic Correlates of Depression and Thoughts of Self-Harm Among Adolescents in Five Low- and Middle-Income Countries: Evidence from the Demographic and Health Survey Program
Year: 2026
Type: preprint
Abstract: Background Adolescence is a critical period for the onset of depression and self-harm, yet comparable, nationally representative estimates from low- and middle-income countries (LMICs) remain scarce. Existing evidence relies heavily on school-based samples, excluding out-of-school adolescents. The recently available data from Demographic and Health Surveys (DHS) Mental Health Module offers an opportunity to address this gap. Methods and findings We analysed data from 16,269 adolescents aged 15–19 years across five DHS surveys: Bangladesh (2022), Lesotho (2023–24), Mozambique (2022–23), Nepal (2022), and Zambia (2024). Probable depression was defined as a PHQ-9 score ≥10, and thoughts of death or self-harm as any endorsement of PHQ-9 item 9. We estimated survey-weighted prevalence by country, sex, and age, examined symptom profiles, and used sex- and country-specific modified Poisson regression to identify sociodemographic correlates. The country-normalised pooled prevalence was 3.9% (95% CI 3.4–4.5) for probable depression, ranging from 0.6% in Zambia to 6.5% in Mozambique, and 7.0% (95% CI 6.4–7.7) for thoughts of death or self-harm, ranging from 3.1% to 10.2%, respectively. Girls had higher prevalence than boys for both depression (4.9% vs 1.9%) and thoughts of death or self-harm (8.6% vs 3.9%), although sex differences varied across countries. No consistent age gradient emerged. Associations with household composition, marital and parental status, urban residence, education, and wealth were heterogeneous across settings. Interpretation is limited by the lack of population-specific PHQ-9 validation and cross-country measurement invariance testing, the inability of item 9 to distinguish passive death wishes from suicidal ideation, the restriction to ever-married girls in Bangladesh, and the cross-sectional design. Conclusions Across five LMICs, PHQ-9-defined probable depression was relatively uncommon, while thoughts of death or self-harm affected approximately one in fourteen adolescents, particularly in girls. Substantial cross-country variation highlights the need for context-sensitive adolescent mental health surveillance and prevention, including out-of-school populations, supported by repeated household surveys and locally validated measures. Author summary Why was this study done? Depression and self-harm often begin during adolescence, and suicide is one of the leading causes of death among young people worldwide. In low- and middle-income countries, most of what we know comes from school-based or local studies that use different questionnaires. These studies miss adolescents who are not in school and are hard to compare across countries. The Demographic and Health Survey (DHS) Program recently added a mental health module to its national household surveys. This allows comparable estimates across countries for the first time. What did the researchers do and find? We analysed responses from 16,269 adolescents aged 15–19 years in national household surveys from Bangladesh, Lesotho, Mozambique, Nepal and Zambia. All five surveys used the same questionnaire (PHQ-9) to assess depressive symptoms including suicide ideation and self-harm. About 4 in 100 adolescents had probable depression, ranging from under 1% in Zambia to 6.5% in Mozambique. Thoughts of death or self-harm were more common, affecting about 7 in 100 adolescents, and were more frequent than probable depression in every country. Girls reported more symptoms than boys in all countries, but the size of this gap varied widely between countries. Age, place of residence, education, wealth, marriage and parenthood showed no consistent pattern across countries. What do these findings mean? Many adolescents in these countries have thoughts of death or self-harm, including some who do not meet the threshold for probable depression. Overall, girls report higher rates than boys. Screening based only on depression scores may miss them. Risk factors differ between countries, so prevention and services should be designed for each local context. They should also reach adolescents who are not in school.
Source: medRxiv
Language: English
Subfield: Social Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7224990087