Socio-economic status and the double burden of malnutrition in Cambodia between 2000 and 2014: overweight mothers and stunted children

Abstract

Objective: The Cambodian population has experienced an increase in the proportion of stunted children who have overweight mothers during a period of rapid social and economic growth. We aimed to identify socio-economic factors associated with this household-level double burden over time.

Design: We used data from four Cambodia Demographic and Health Surveys from 2000 to 2014 to study the impact of socio-economic status (SES) on the link between child stunting and overweight mothers in two periods 2000-2005 v. 2010-2014. We hypothesised that SES would be a primary factor associated with this phenomenon.

Participants: We included 14 988 children under the age of 5 years, among non-pregnant mothers aged 15-49 years of age and conducted analysis on a subsample of 1572 children with overweight mothers.

Setting: Nationally representative household survey across all regions.

Results: SES factors, specifically household wealth and maternal employment in service or manual occupations (in 2010-2014), are the main drivers of stunting among children of overweight mothers. Children with overweight mothers in the poorest households are more than twice as likely to be stunted than in the richest in both periods (2000-2005: adjusted OR (aOR) = 2·53, 95 % CI: 1·25, 5·13; 2010-2014: aOR = 2·61, 95 % CI: 1·43, 4·77), adjusting for other SES factors, indicating that despite decreasing income inequality, the poorest continue to bear excess risk of a double burden of malnutrition. Maternal short stature also doubled the likelihood of child stunting in both periods, which suggests intergenerational transmission of adversity and physical underdevelopment.

Conclusions: Socio-economic inequalities should be addressed to reduce disparities in the household-level double burden of malnutrition.

Keywords: Cambodia; Child stunting; Maternal nutrition; Socio-economic factors.

Fig. 1

Distribution of wealth indices by wealth quintiles and period among (a) the full sample of children. Period 2000–2005; 2010–2014 and (b) the subsample of children of overweight mothers. , 2000–2005; , 2010–2014

Fig. 2

Predicted probabilities of child stunting by wealth quintile, maternal overweight status and period among the full sample (period 1, n 7033; period 2, n 7955)¹. , 2000–2005; , 2010–2014; , mother not overweight; , mother overweight. ¹Predicted probabilities of stunting using coefficient estimates from Table 3, Model 4 among children of mothers aged 30 years, with three total children, non-smokers, not short, with no education, employed in an agricultural, service or manual occupation, rural residence, and not born during the Khmer Rouge famine. Characteristics were selected based on modal values (e.g. smoking status, occupation, etc.) or values close to sample means (e.g. age, number of children). Child’s age and sex were taken at their means. Points were horizontally dodged to show contrast