Search

ObjectiveHousehold food insecurity is positively associated with depression; however, the association among adolescents is not well known. We examined the association between household food insecurity and depression among adolescent girls in Northern Ghana.MethodsWe analysed data from the Ten2Twenty-Ghana randomised controlled trial end-line for adolescent girls aged 10–17 years (n=582). The girls were chosen at random from 19 primary schools in the Mion District of Ghana’s northern region. The children’s depression inventory and the Food Insecurity Experience Scale were used in face-to-face interviews to assess depression and household food insecurity. Hierarchical survey binary logistic regression and linear mixed models were used to examine the association between household food insecurity and depressive symptoms. We took into account a number of potential confounders in the analysis, such as life satisfaction, self-efficacy, self-esteem, health complaints, child’s age, menarche status, pubertal development, anaemia, stunting, frequency of consuming fruits and vegetables, frequency of consuming animal-sourced foods, maternal age, household wealth index and size, and the intervention group the girl was assigned to in the trial.ResultsAbout 20.1% of adolescent girls were classified as likely depressed, and 70.3% of their households were food insecure, with 22.9% and 18.0% being moderately and severely food insecure, respectively. Compared with girls from food-secure households, those from moderately (adjusted OR (AOR) 2.63, 95% CI (1.35 to 5.12)) and severely (AOR 3.28, 95% CI (1.66 to 6.49)) food insecure households had about three times the odds of being classified as depressed, after controlling for potential confounders. The odds of being likely depressed were about twice for adolescent girls from food-insecure households compared with their peers from food-secure households in both the crude and final adjusted model.ConclusionThe study discovered high levels of household food insecurity and depression in adolescent girls in Northern Ghana, with a dose-response association between the two.