A social-ecological trap theory-informed investigation of dietary patterns in southwestern Madagascar

Heather Kelahan^1*Stephanie M. Wu²Hervet Randriamady¹Aroniaina Falinirina³Madeleine Rasoanirina³Frédéric Déclerque³Marc Y. Solofoarimanana³Jean C. Mahefa³Eric B. Rimm^1,4Jessica Fanzo⁵Aaron C. Hartmann^6,7Emma Gibbons⁸Gildas Georges Boleslas Todinanahary³Sebastien Haneuse⁹Christopher D. Golden¹

• ¹Department of Nutrition, Harvard T.H. Chan School of Public Health, Boston, United States
• ²University College London Division of Psychiatry, London, United Kingdom
• ³Universite de Toliara Institut Halieutique et des Sciences Marines, Toliara, Madagascar
• ⁴Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, United States
• ⁵Columbia University Climate School, New York, United States
• ⁶Department of Organismic and Evolutionary Biology, Harvard University, Cambridge, United States
• ⁷Perry Institute for Marine Science, Waitsfield, United States
• ⁸Reef Doctor, Ifaty, Madagascar
• ⁹Department of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, United States

Social-ecological trap theory highlights the potential for food systems and the social and environmental contexts within which they are situated to 'trap' individuals into a trajectory of specific nutritional outcomes when their physical or financial access to traditional foods is restricted, resulting in less healthy dietary patterns than those traditionally consumed. While social-ecological trap theory literature highlights the potential for these traps to result in four hypothesized dietary patterns, the presence and composition of such dietary patterns have not been explored in southwestern Madagascar. This study employs innovative Weighted Overfit Latent Class Analysis methods to identify dietary patterns among individuals residing in southwestern Madagascar. The study used longitudinal cohort data collected from 2023-2024. Four dietary patterns were identified and characterized as 1) a traditional dietary pattern which included 35.9% (SD 11.1%) of the population, 2) a industrialized-transitioning dietary pattern which included 29.2% (SD 13.5%) of the population, 3) an traditional-undernourishing dietary pattern which included 16.3% (SD 5.2%) of the population, and 4) a industrialized-undernourishing dietary pattern which included 17.8% (SD 11.2%) of the population. The four dietary patterns identified aligned with three of the four patterns hypothesized to result from social-ecological traps. Those in the traditional dietary pattern consumed the most diverse diet and tended to be fishers who also often participated in crop-based agriculture. Those in the industrialized-transitioning dietary pattern consumed a greater proportion of their diet from market-source foods. Those in the traditional-undernourishing dietary pattern consumed the fewest calories and had the lowest level of food security. Lastly, those in the industrialized-undernourishing dietary pattern consumed 63% of their calories from rice and consumed more market-source foods than those in the traditional-undernourishing dietary pattern. Of the four dietary patterns identified in southwestern Madagascar, two are characterized as higher-quality and two as undernourishing dietary patterns. Each dietary pattern comprises individuals of varying demographic and socio-economic status. Understanding dietary patterns and who follows them enables policymakers and public health practitioners to better understand who may be most affected by the impacts of social and ecological change on the food system, thereby improving the targeting of nutritional interventions.