Ultra-processed foods in a rural Ecuadorian community: associations with child anthropometry and bone maturation

Abstract

Frequent ultra-processed food (UPF) consumption is consistently associated with poor health outcomes. Little is known about UPF intake during early childhood and its effects on growth. We assessed UPF in relation to child anthropometry, bone maturation, and their nutrition profiles in a rural Ecuadorian community. Covariate-adjusted regression models estimated relationships between UPF intake from a 24-hour Food Frequency Questionnaire and three outcomes: linear growth, weight status and bone maturation. Nutrient Profiling Models (NPM) evaluated a convenience sample of UPF (n 28) consumed by children in the community. In this cohort (n 125; mean age = 33·92 (sd 1·75) months), 92·8 % consumed some form of UPF the previous day. On average, children consuming UPF four to twelve times per day (highest tertile) had lower height-for-age z-scores than those with none or a single instance of UPF intake (lowest tertile) (β = -0·43 [se 0·18]; P = 0·02). Adjusted stunting odds were significantly higher in the highest tertile relative to the lowest tertile (OR: 3·07, 95 % CI 1·11, 9·09). Children in the highest tertile had significantly higher bone age z-scores (BAZ) on average compared with the lowest tertile (β = 0·58 [se 0·25]; P = 0·03). Intake of savoury UPF was negatively associated with weight-for-height z-scores (β = -0·30 [se 0·14]; P = 0·04) but positively associated with BAZ (β = 0·77 [se 0·23]; P < 0·001). NPM indicated the availability of unhealthy UPF to children, with excessive amounts of saturated fats, free sugars and sodium. Findings suggest that frequent UPF intake during early childhood may be linked to stunted growth (after controlling for bone age and additional covariates), despite paradoxical associations with bone maturation.

Keywords: Bone age; Children; Ecuador; NOVA classification; Stunting; Ultra-processed foods; Weight status.

Fig. 1.

Mediation model: total effect: effect of UPF consumption (tertile 2 (c₁) and tertile 3 (c₂)) on HAZ or stunting. Direct effect: effect of UPF consumption (tertile 2 (c'₁) and tertile 3 (c'₂)) on HAZ or stunting, after controlling for BAZ. Indirect effect: effect of UPF consumption (tertile 2 (a₁ × b) and tertile 3 (a₂ × b) on HAZ or stunting, mediated by BAZ. tertile 1: consumption of UPF 0–1 times per day (reference group, not shown in model); tertile 2: consumption of UPF 2–3 times per day; tertile 3: consumption of UPF 4–12 times per day. UPF, ultra-processed food; HAZ, height-for-age z-score; BAZ, bone age z-scores.

Fig. 2.

Consumption prevalence for ultra-processed food groups: bars in grey on the farthest left of each ultra-processed food category (a) salty snacks; (b) sugary foods; (c) sugar-sweetened beverages represent the proportion of children within the cohort studied (Overall) (n 125) who consumed food from a given group the day prior. The three additional bars represent the proportion of children within each tertile (Tertile 1 (n 42); Tertile 2 (n 42), Tertile 3 (n 41)) who consumed food from a given group the day prior.

Fig. 3.

Association between ultra-processed food consumption and stunting (HAZ < –2): adjusted OR and 95 % CI showing the association between different UPF exposures and stunting (HAZ < –2), determined by logistic regression models. Each model was adjusted for child BAZ, child sex; child age; parental/caregiver socio-economic characteristics: employment status, involvement in food production, livestock ownership; meeting minimum dietary diversity, and group assignment from the Lulun project. ^aModel where frequency of UPF consumption according to tertiles is the primary exposure; reference group – tertile 1(consumption 0–1 time a day). ^bModel where any salty snack consumption is the primary exposure; reference group – no salty snack consumption. ^cModel where any sugary food consumption is the primary exposure; reference group – no sugary food consumption. ^dModel where any sugar-sweetened beverage consumption is the primary exposure; reference group – no sugar-sweetened beverage consumption. *indicates statistical significance (P < 0·05). HAZ, height-for-age z-score; UPF, ultra-processed food; BAZ, bone age z-score.

Fig. 4.

Association between ultra-processed food consumption and low bone age (BAZ < –2): adjusted OR and 95 % CI showing the association between different UPF exposures and low bone age (BAZ < –2), determined by logistic regression models. Each model was adjusted for child sex; child age; parental/caregiver socio-economic characteristics: employment status, involvement in food production, livestock ownership; meeting minimum dietary diversity, and group assignment from the Lulun project. ^aModel where frequency of UPF consumption according to tertiles is the primary exposure; reference group – tertile 1(consumption 0–1 time a day). ^bModel where any salty snack consumption is the primary exposure; reference group – no salty snack consumption. ^cModel where any sugary food consumption is the primary exposure; reference group – no sugary food consumption. ^dModel where any sugar-sweetened beverage consumption is the primary exposure; reference group – no sugar-sweetened beverage consumption. BAZ, bone age z-score; UPF, ultra-processed food.