**Background:** Food insecurity (FI) is a major global health challenge associated with both physical and mental health outcomes. Youth aged 15–24 are particularly vulnerable to FI and mental health concerns, with approximately 86 million youth aged 15–19 living with a mental disorder in 2019. Climate-related disasters are a key determinant of both FI and poor mental health, yet the three-way relationship among these factors has not been well studied. Only one prior study (Friel et al., using an Australian sample) examined whether drought modifies the FI-mental health association. The present study aimed to determine whether climate-related disaster severity intensifies the relationship between FI and poor mental health among youth globally.
**Methods:** The study used a cross-sectional design with data from two secondary sources. Individual-level data on FI, mental health, and confounders came from the 2017 Gallup World Poll (GWP), a nationally representative survey. The final sample included 28,292 youth aged 15–24 from 142 countries. FI was measured using the 8-item Food Insecurity Experiences Scale (FIES), categorized as none/mild (score 0–3), moderate (4–6), or severe (7–8). Poor mental health was measured using the Daily Experience Index (DEI), a 10-item count score (0–10, higher = worse). Country-level climate-related disaster data (2015–2017) came from the International Disaster Database (EM-DAT). Disaster severity was defined as total number of people killed per 1,000,000 population, categorized as high (top 10% of countries) vs. low (bottom 90%). Multilevel negative binomial regression with robust standard errors was used, with a random intercept for country. Models adjusted for age, gender, urbanicity, number of children in household, marital status, education, employment, and log annual household income.
**Key Results:** Among the sample, 77.8% reported none/mild FI, 11.2% moderate FI, and 11.0% severe FI. Mean poor mental health score was 2.29 (SEM 0.06) out of 10. During 2015–2017, 828 climate-related disasters occurred across the 142 countries, resulting in 27,508 deaths (47.7% floods, 33.2% storms). Youth with moderate FI were significantly more likely to report poor mental health compared to those with none/mild FI (RR 1.37, 95% CI 1.32–1.41), and those with severe FI showed an even stronger association (RR 1.60, 95% CI 1.54–1.66). FI produced the strongest effect of all predictors. Other significant predictors included older age (RR 1.02, 95% CI 1.02–1.03), greater number of children in household (RR 1.01, 95% CI 1.00–1.01), lower income (RR 0.96, 95% CI 0.95–0.97), elementary education or less (RR 1.06, 95% CI 1.00–1.13), and being widowed (RR 1.34, 95% CI 1.13–1.58), separated (RR 1.14, 95% CI 1.02–1.28), or married (RR 1.06, 95% CI 1.01–1.10). The FI × disaster severity interaction was not significant when disaster severity was dichotomous (p=0.5362). However, in sensitivity analyses treating disaster severity as a continuous measure of deaths per 1,000,000 population, a weak but significant interaction was observed (p<0.0001), driven by severe FI (p=0.0063) but not moderate FI (p=0.1817). The association between FI and poor mental health was consistent across all country income levels and all 19 UN sub-regions, with stronger associations concentrated in Western countries.
**Clinical Implications:** The findings demonstrate a robust, dose-response relationship between FI and poor mental health among youth that persists across geographic and cultural contexts. The association remained significant after controlling for poverty indicators, suggesting psychosocial and biological mechanisms beyond material deprivation (e.g., toxic stress, malnutrition effects on brain function). The weak interaction with disaster severity suggests that climate-related disasters may compound the mental health burden of FI, though the effect is small. Policies aimed at reducing FI—such as diversifying food production, improving food transport infrastructure, promoting food fortification, and reducing poverty—may also improve youth mental health globally. The authors recommend that mental health be considered in national climate change adaptation actions and that additional FI-related supports may help improve youth mental health. Limitations include the cross-sectional design (preventing causal inference), potential unmeasured confounders (e.g., pre-existing mental health conditions), self-reported measures, and the use of country-level disaster data that may not reflect individual-level exposure.