Abstract
Objectives
Northern coastal Ecuador has seen rapid market integration since the initiation of a road system in the late 1990 s connecting the region to the Andean mountains. We previously demonstrated that, between 2003 and 2013, living in a relatively remote (less accessible) community was protective against stunting in children under five. Our objectives were to characterize trends in BMI among rural adults over the same period, and to examine the co-occurrence of adult overweight or obesity (OB/OW) and child stunting.
Methods
Adult anthropometry was collected through a series of case-control studies repeated over a decade across thirty communities (N = 2053). Anthropometry was measured for all children under five in every community (N = 4170). Logistic and mixed-effect Poisson models, weighted to account for the case-control sampling design, were used to estimate the prevalence of OW/OB over time, to examine individual- and household-level risk factors associated with adult OW/OB, and to examine the co-occurrence of adult OW/OB and under-five stunting within households.
Results
Among men, OW/OB increased over time (35.0% in 2003 to 52.9% by 2013) and was most common in the least remote versus the most remote communities (46.2% versus 26.9% across the study period). Among women, OW/OB was more prevalent (65.4%), but was not associated with community remoteness and did not increase over time. At the community level, the prevalence of under-five stunting was correlated with the prevalence of OW/OB among men (Spearman's ρ = 0.56, P = 0.002) but not women (Spearman's ρ = 0.16, P = 0.404). However, the presence of a stunted child under five within the household was not associated with adult OW/OB (odds ratio 0.87, P = 0.151).
Conclusions
Changes in food environments and livelihoods due to the presence of roads may have driven increasing OW/OB among men over the study period. The positive association between adult male OW/OB and child stunting at the community level, but not at the household level, suggests that strategies to address the double burden of malnutrition that target the community (e.g., the food environment) are more likely to be effective than interventions targeted to households alone.
Funding Sources
This project was supported by the National Institutes of Health [R01-AI050038].