PROCOMIDA, a Food-Assisted Maternal and Child Health and Nutrition Program, Reduces Child Stunting in Guatemala: A Cluster-Randomized Controlled Intervention Trial

Abstract

Background: Food-assisted maternal and child health and nutrition (FA-MCHN) programs may foster child growth during the first 1000 d (pregnancy and the first 2 y of a child's life), but evidence is scant.

Objective: We evaluated the impact of an FA-MCHN program, PROCOMIDA, on linear growth (stunting [length-for-age z score (LAZ) < -2] and length-for-age difference [LAD]) among children aged 1-24 mo. PROCOMIDA was implemented in Guatemala by Mercy Corps and was available to beneficiaries throughout the first 1000 d.

Methods: We used a longitudinal, cluster-randomized controlled trial with groups varying in family ration sizes [full (FFR), reduced (RFR), and none (NFR)] and individual ration types provided to mothers (pregnancy to 6 mo postpartum) and children (6-24 mo of age) [corn-soy blend (CSB), lipid-based nutrient supplement (LNS), micronutrient powder (MNP)]: 1) FFR + CSB (n = 576); 2) RFR + CSB (n = 575); 3) NFR + CSB (n = 542); 4) FFR + LNS (n = 550); 5) FFR + MNP (n = 587); 6) control (n = 574). Program impacts compared with control, and differential impacts between groups varying family ration size or individual ration type, were assessed through the use of linear mixed-effects models and post hoc simple effect tests (significant if P < 0.05).

Results: PROCOMIDA significantly reduced stunting at age 1 mo in FFR + CSB, RFR + CSB, and FFR + MNP groups compared with control [5.05, 4.06, and 3.82 percentage points (pp), respectively]. Stunting impact increased by age 24 mo in FFR + CSB and FFR + MNP relative to control (impact = 11.1 and 6.5 pp at age 24 mo, respectively). For CSB recipients, the FFR compared with RFR or NFR significantly reduced stunting (6.47-9.68 pp). CSB reduced stunting significantly more than LNS at age 24 mo (8.12 pp).

Conclusions: FA-MCHN programs can reduce stunting during the first 1000 d, even in relatively energy/food-secure populations. Large family rations with individual rations of CSB or MNP were most effective. The widening of impact as children age highlights the importance of intervening throughout the full first 1000 d. This trial was registered at clinicaltrials.gov as NCT01072279.

FIGURE 1

Study flow diagram. CC, convergence center; CSB, corn-soy blend; FFR, full family ration; LNS, lipid-based nutrient supplement; MNP, micronutrient powder; NFR, no family ration; RFR, reduced family ration.

FIGURE 2

Proportion of eligible beneficiaries participating in PROCOMIDA at each time point by study group. Participation was assessed by self-report of participation in the PROCOMIDA program. Enrollment into the survey was done during pregnancy and when the child was 1, 4, 6, 9, 12, 18, and 24 mo of age. Among those currently enrolled in PROCOMIDA, participation in the monthly behavior change communication and food distributions was >95% at all time points for all study groups. CSB, corn-soy blend; FFR, full family ration; LNS, lipid-based nutrient supplement; MNP, micronutrient powder; NFR, no family ration; RFR, reduced family ration.

FIGURE 3

Estimated marginal mean prevalence of stunting (length-for-age z score <−2) among children in treatment compared with control groups at 1, 4, 9, 12, 18, and 24 mo of age. Estimated marginal means (unadjusted means are presented in Tables 5 and 6), *P < 0.05, **P < 0.01 for impact on stunting of treatment group [FFR + CSB (n = 576); RFR + CSB (n = 575); NFR + CSB (n = 542); FFR + LNS (n = 550); FFR + MNP (n = 587)] compared with the control group (n = 574) (Table 4). CSB, corn-soy blend; FFR, full family ration; LNS, lipid-based nutrient supplement; MNP, micronutrient powder; NFR, no family ration; RFR, reduced family ration.