Effects of persistent undernutrition during adolescence on learning skills: Findings from a longitudinal study of 16,000 adolescents in India

Non-PR
IFPRI5; CRP4
PHND; A4NH
CGIAR Research Programs on Agriculture for Nutrition and Health (A4NH)
Objectives: Limited evidence exists on longitudinal impacts of undernutrition on learning skills as adolescents mature. This study aimed to examine the effect of undernutrition in early adolescence on learning skills in late adolescence. Methods: We used longitudinal data from a project called UDAYA that surveyed 16,929 adolescents aged 10–19 years from Bihar and Uttar Pradesh, India in 2015–16 (wave 1) and again in 2018–19 (wave 2). Adolescents’ ability to read a story and solve division problems at age 19–22 years was assessed using the Annual Status of Education Report tools. We characterized adolescents’ nutrition status into 4 groups: never undernourished (e.g., not stunted in wave 1 and 2), recovered (e.g., stunted in in wave 1 but not stunted in wave 2), faltered (e.g., not stunted in wave 1 but stunted in wave 2), and persistent (e.g., stunted in both waves). These 4 groups were created for stunting, thinness, and anemia. We used multivariable logistic regression models adjusted for key demographics, environmental factors, and sampling design. Results: Undernutriton was high (51% were anemic, 19% thin, and 36% stunted) in both survey rounds. Nearly one third of adolescents had persistent stunting or anemia, and 11% had persistent thinness. Compared to those who were never stunted, adolescents with persistent stunting had poorer reading (adjusted odds ratio: 0.66, 95% confidence interval: 0.53–0.83) and math (0.70, 0.56–0.89) skills. Reading ability was also poorer in adolescents who experienced growth faltering (0.61, 0.42–0.90) or who were stunted in wave 1 but not stunted in wave 2 (0.64, 0.44–0.93). Persistent thinness was negatively associated with reading (0.71, 0.54–0.93) but not math skills. Persistent anemia was also negatively associated with reading skills, but the association did not remain significant in the fully adjusted model. Conclusions: In a high poverty sample of Indian adolescents, persistent undernutrition during adolescence was associat