Correlates of Depressive Symptoms Among Mothers With Acutely Ill Children Admitted to Hospitals in Low- And Middle-Income Countries

Abstract

Background: Maternal depression is the second leading cause of maternal disability globally. In addition to causing significant maternal morbidity, maternal depression may also be associated with increased child morbidity and mortality, possibly through a reduction in the ability of mothers to meet children’s care needs. Mothers in low-and-middle income countries (LMICs) have higher rates of depression after childbirth than high-income peers. Despite a well-established body of evidence surrounding maternal depression in high-income countries (HICs), few studies have addressed correlates of maternal depression in LMICs. This study used data from the Childhood Acute Illness & Nutrition Network to investigate potential correlates of maternal depression among caregivers with a child being admitted to hospital in nine LMIC settings. Methods: 2,949 biological mothers were included in this secondary analysis. Demographic and socioeconomic data were collected from caregivers upon the admission of acutely ill children to nine hospitals within six LMICs. The participants were asked to complete the patient health questionnaire-9 (PHQ-9) within 6 hours of their children’s admission, to screen for depressive symptoms. The cumulative PHQ-9 score was coded as a continuous variable throughout our secondary analysis and all analyses were conducted in R. Using univariate, bivariate, and multivariate interaction models, we estimated hypothesized variables as effect modifiers and also examined their direct effects on PHQ-9 scores in bivariate and multivariate analyses and we investigated the correlates of maternal depressive symptoms in our study context. Results: Overall, there were 1671 male children (57%) and 1271 female children (43%) in the sample. No crude relationship between child sex and maternal PHQ-9 score was found (β:0.3, 95% CI -0.1, 0.6, P=0.15), but a significant interaction between the Migori Hospital site (Kenya) and child gender suggests that have a male child at this site was associated with a 1.7 point greater reduction in PHQ-9 than having a male child at other sites (95% CI: -3.2, -0.3, p= 0.02). Mothers who were the child’s primary financial provider also appeared to affect the relationship between child’s sex and maternal PHQ-9 in crude models (β: 1.7, 95% CI -2.4, -1.1 & p <0.001), but this association was not significant in multivariate models. Multiple other exposures of interest were independently associated with PHQ-9 scores in multivariate models. These exposures are sites of: Blantyre (β: -3.4, 95% CI -4.3, -2.6, p <0.0001); Karachi (β: -2.9, 95% CI -3.8, -1.9, p <0.0001); Matlab (β: -4.1, 95% CI -4.7, -3.5, p <0.0001; and Migori (β: 1.0, 95% CI 0.0, 1.9, p=0.04). Also, food insecurity (β: 0.4, 95% CI 0.4, 0.5, p <0.001; marital status (β: -0.8 (95% CI -1.3, -0.3, p <0.001); child’s age at admission (β: 0.3, 95% CI 0.3, 1, p <0.029); and the middle upper arm circumference (MUAC) (β: -0.2, 95% CI -0.3, -0.1, p <0.001). Conclusion: Mothers of admitted acutely ill children to hospitals in LMICs may be at a particularly high risk of depression, both due to higher risk of child illness in these families and due to situational stress during prolonged periods of intense contact with the healthcare system. Multiple strong correlates of maternal mental health were identified at pediatric admission in the context of an acutely ill child being admitted to hospitals. Some of these correlates appeared relevant across study sites, while others, such as gender, appeared restricted to individual sites. Understanding these correlates could improve the detection of maternal depression in low and middle-resource settings and could contribute to the design of target mental health interventions during pediatric hospitalization.