Household food insecurity and child health outcomes: a rapid review of mechanisms and associations

BACKGROUND:Household food insecurity (HHFI) has detrimental impact on child health outcomes. The primary aim of this work was to describe the mechanisms, defined as statistically quantified processes or systems, that explain the association between HHFI and poor child health outcomes. The secondary aim was to identify which child health outcomes were significantly associated with HHFI. METHODS:In this rapid review, we included real-world evidence studies in children and adolescents aged 3-24 years, from high-income countries. We searched peer-reviewed literature through the databases Medline, Embase, Web of Science and The Cochrane Library on March 10, 2022, with a 15-year time restriction. Key search terms included "food insecurity", "food poverty", "statistics", "structural equation model*", "regression", "conceptual model*", and "theoretical model*". We identified additional studies by searching citations and references. Papers were selected using pre-defined criteria and non-English Language studies were excluded. We extracted summary data of published studies using the Cochrane data extraction template as a guide. We included a broad range of health outcomes to identify what aspects of health are affected by HHFI, including metabolic risk factors (eg, BMI), conditions (eg, depression), biological processes (eg, sleep) and parent-child interactions (eg, feeding styles). We performed a narrative synthesis, as heterogeneity precluded statistical synthesis. We used the Quality in Prognosis Studies tool to assess the mechanism studies. FINDINGS:We included 70 systematic reviews (six mechanism and 64 association reviews) from the UK, Canada, USA, Australia, and Ireland (including 18 887 mechanism studies and 986 759 association studies). Low-to-moderate bias was detected in mechanism studies, and there was variability in HHFI measures. Maternal depression and parenting stress during childhood played a mechanistic role between HHFI and adolescent depression (b=0·008, 95% CI 0·002-0·016) and anxiety (b=0·012, 0·002-0·026). Maternal stress mediated the association between HHFI and child overweight (odds ratio [OR] 1·79, 0·82-3·92, p<0·01), via restrictive feeding styles (OR 1·81, 1·15-2·85, p=0·010) that had negative impact on diet (p<0·01). Ethnicity, income, employment, and education moderated the relationship between HHFI and child health outcomes (p<0·001). INTERPRETATION:Two key mechanistic pathways between HHFI and child health outcomes were diet and mental health of both child and parent, which appeared interrelated. Interventions targeting HHFI in children, should consider these pathways alongside sociodemographic factors. A strength of this review was that it was the first literature summary explaining the mechanisms between HHFI and child health, using evidence from high-income countries. A limitation was that HHFI measures were inconsistent between studies and countries, disallowing study comparisons. FUNDING:University of Sheffield.