Impact on child acute malnutrition of integrating a preventive nutrition package into facility-based screening for acute malnutrition during well-baby consultation: A cluster-randomized controlled trial in Burkina Faso

Abstract

Background: Community management of acute malnutrition (CMAM) is a highly efficacious approach for treating acute malnutrition (AM) in children who would otherwise be at significantly increased risk of mortality. In program settings, however, CMAM's effectiveness is limited because of low screening coverage of AM, in part because of the lack of perceived benefits for caregivers. In Burkina Faso, monthly screening for AM of children <2 years of age is conducted during well-baby consultations (consultation du nourrisson sain [CNS]) at health centers. We hypothesized that the integration of a preventive package including age-appropriate behavior change communication (BCC) on nutrition, health, and hygiene practices and a monthly supply of small-quantity lipid-based nutrient supplements (SQ-LNSs) to the monthly screening would increase AM screening and treatment coverage and decrease the incidence and prevalence of AM.

Methods and findings: We used a cluster-randomized controlled trial and allocated 16 health centers to the intervention group and 16 to a comparison group. Both groups had access to standard CMAM and CNS services; caregivers in the intervention group also received age-appropriate monthly BCC and SQ-LNS for children >6 months of age. We used two study designs: (1) a repeated cross-sectional study of children 0-17 months old (n = 2,318 and 2,317 at baseline and endline 2 years later) to assess impacts on AM screening coverage, treatment coverage, and prevalence; (2) a longitudinal study of 2,113 children enrolled soon after birth and followed up monthly for 18 months to assess impacts on AM screening coverage, treatment coverage, and incidence. Data were analyzed as intent to treat. Level of significance for primary outcomes was α = 0.016 after adjustment for multiple testing. Children's average age was 8.8 ± 4.9 months in the intervention group and 8.9 ± 5.0 months in the comparison group at baseline and, respectively, 0.66 ± 0.32 and 0.67 ± 0.33 months at enrollment in the longitudinal study. Relative to the comparison group, the intervention group had significantly higher monthly AM screening coverage (cross-sectional study: +18 percentage points [pp], 95% CI 10-26, P < 0.001; longitudinal study: +23 pp, 95% CI 17-29, P < 0.001). There were no impacts on either AM treatment coverage (cross-sectional study: +8.0 pp, 95% CI 0.09-16, P = 0.047; longitudinal study: +7.7 pp, 95% CI -1.2 to 17, P = 0.090), AM incidence (longitudinal study: incidence rate ratio = 0.98, 95% CI 0.75-1.3, P = 0.88), or AM prevalence (cross-sectional study: -0.46 pp, 95% CI -4.4 to 3.5, P = 0.82). A study limitation is the referral of AM cases (for ethical reasons) by study enumerators as part of the monthly measurement in the longitudinal study that may have attenuated the detectable impact on AM treatment coverage.

Conclusions: Adding a preventive package to CMAM delivered at health facilities in Burkina Faso increased participation in monthly AM screening, thus overcoming a major impediment to CMAM effectiveness. The lack of impact on AM treatment coverage and on AM prevalence and incidence calls for research to address the remaining barriers to uptake of preventive and treatment services at the health center and to identify and test complementary approaches to bring integrated preventive and CMAM services closer to the community while ensuring high-quality implementation and service delivery.

Trial registration: ClinicalTrials.gov NCT02245152.

Fig 1. Theory of change of the integration of a preventive package at the well-baby visit.

Hypothesized impacts of the intervention presented in this paper are shown in blue. AM, acute malnutrition; BCC, behavior change communication; IYCF, infant and young child feeding; PROMIS, Innovative Approaches for the Prevention of Childhood Undernutrition; SN-LNS, small-quantity lipid-based nutrition supplement.

Fig 2. Trial profile for repeated cross-sectional study and longitudinal study.

AM was defined as WLZ < −2 at enrollment and at the first follow-up. AM, acute malnutrition; BCC, behavior change communication; HC, health center; SQ-LNS, small-quantity lipid-based nutrient supplement; WLZ, weight-for-length z-score.

Fig 3. AM screening coverage by any actor (A) and through CNS (B) in the longitudinal study by age and by study group.

Monthly screening was defined as MUAC measured (≥6 months old) and/or weight and length measured (all ages) in the past month, as per caregiver recall. The orange solid line represents fitted values for the comparison group. The blue dashed line represents fitted values for the intervention group. Gray areas represent 95% confidence bands of kernel-weighted local polynomial smoothened values by study group using the observed data. Both analyses were based on n = 18,757 child visits in the comparison group and n = 17,867 child visits in the intervention group. Mixed-effects regression models with restricted cubic splines (seven knots automatically generated) were used, with health center catchment area and child as random intercepts and month of inclusion, child sex, age splines, whether the child was a first live birth, and intervention as fixed effects. A chunk Wald test was used to test the “age splines × intervention” interaction terms (P values shown). AM, acute malnutrition; MUAC, mid-upper arm circumference.