Examining Transactional Associations between Food Insecurity and Parental Depressive Symptoms

Background and purpose: Food insecurity is a pervasive and serious social problem in the United States, particularly within households with young children. Prior research has outlined the potential bidirectional relationships between food insecurity and parental depressive symptoms. However, longitudinal investigations in this area have been limited, with most studies heavily emphasizing the effects of food insecurity on parental depressive symptoms rather than exploring the reverse relationships. To untangle the directional relations, this study examined the longitudinal transactional relations between food insecurity and parental depressive symptoms, separating within-person from between-person effects to yield less biased estimates of transactional relations.

Methods: This study analyzed data from the Early Childhood Longitudinal Study: Kindergarten Class of 2010-11 (ECLS-K: 2011), a nationally representative cohort study of U.S. children. The final sample consisted of parents who completed at least one of the main variables (N =15,102, M_age = 34.5 years, SD = 6.7; 55.2% White; 70.7% married). Parents reported food insecurity, which was measured using 10 items from the U.S. Department of Agriculture. This study used a continuous measure of food insecurity provided by the ECLS-K: 2011 based on raw scores. Parental self-reported depressive symptoms were assessed using a 12-item 4-point Likert scale adapted from the Center for Epidemiological Studies Depression Scale. All variables were assessed at four time points: spring 2010 (T1), spring 2013 (T2), spring 2014 (T3), and spring 2015 (T4), spanning from kindergarten to fifth grade for a child of the parents. Random-intercept cross-lagged panel modeling (RI-CLPM) was utilized to separate between- and within-person effects, using Mplus 8.3.

Results: The final RI-CLPM had an acceptable model fit: CFI = .994, RMSEA = .019 (90% CI = .015 to .024), and SRMR = .016. This study partially observed transactional relations at the within-person level. Higher levels of parental depressive symptoms at T1 were significantly associated with higher levels of food insecurity at T2 (β = .044, SE = .022, p < .05), which in turn led to greater depressive symptoms at T3 (β = .062, SE = .025, p < .05). In addition, the between-person correlation between food insecurity and depressive symptoms was significant (β = .516, SE = .019, p <.001).

Conclusions and implications: Using a rigorous analytical technique, this study contributes to the literature by demonstrating the potential transactional relations between food insecurity and parental depressive symptoms. The transactional relations at the within-person level suggest the need for interventions and policies addressing both food insecurity and parental depressive symptoms simultaneously. Such efforts may ultimately contribute to promoting a child’s healthy developmental outcomes. It is worth noting that this study only observed transactional relations at three time points. Therefore, findings should be replicated with other samples and further verified in future research.