BACKGROUND

Childhood overweight and obesity (OW/OB) is increasingly centered in low- and middle-income countries (LMICs) as rural populations experience market integration and lifeway change. Most explanatory studies have relied on imprecise estimates of children's energy expenditure, restricting understanding of the relative effects of changes in diet and energy expenditure on the development of OW/OB in transitioning contexts.

OBJECTIVES

This study used gold-standard measurements of children's energy expenditure to investigate the changes that underlie OW/OB and the nutrition/epidemiologic transition.

METHODS

Cross-sectional data were collected from "rural" (n = 43) Shuar forager-horticulturalist children and their "peri-urban" (n = 34) Shuar counterparts (age 4-12 y) in Amazonian Ecuador. Doubly labeled water measurements of total energy expenditure (TEE; kcal/d), respirometry measurements of resting energy expenditure (REE; kcal/d), and measures of diet, physical activity, immune activity, and market integration were analyzed primarily using regression models.

RESULTS

Peri-urban children had higher body fat percentage (+8.1%, P < 0.001), greater consumption of market-acquired foods (multiple P < 0.001), lower concentrations of immune activity biomarkers (multiple P < 0.05), and lower REE (-108 kcal/d, P = 0.002) than rural children. Despite these differences, peri-urban children's TEE was indistinguishable from that of rural children (P = 0.499). Moreover, although sample-wide IgG concentrations and household incomes predicted REE (both P < 0.05), no examined household, immune activity, or physical activity measures were related to children's overall TEE (all P > 0.09). Diet and energy expenditure associations with adiposity demonstrate that only reported consumption of market-acquired "protein" and "carbohydrate" foods predicted children's body fat levels (multiple P < 0.05).

CONCLUSIONS

Despite underlying patterns in REE, Shuar children's TEE is not reliably related to market integration and-unlike dietary measures-does not predict adiposity. These findings suggest a leading role of changing dietary intake in transitions to OW/OB in LMICs.