Effect of a Complementary Food Supplement on Growth and Morbidity of Ghanaian Infants

The current study is a cluster randomized single blind intervention design study with three study arms that aimed to examine the effect of providing a macro- and micro-nutrient fortified complementary food supplement (KokoPlus) for a period of 12 months (starting at 6 months of age) on growth and nutritional status of Ghanaian infants at 18 months of age. KokoPlus was formulated using linear programming methodology based on formative and market analysis research findings.

The subjects in this cluster-randomized trial are from communities in three districts of the Central region in Ghana with high rates of moderate and severe acute malnutrition. A total of 38 communities will be randomly assigned to one of three groups using block randomization and another 11 (randomly selected) will be followed cross sectionally as part of a fourth/non intervention group (growth monitoring).

The total sample size is 1204. Sample size calculations were based on two outcome measures: expected reduction in diarrheal morbidity and growth (improvements in height-for-age). Sample size estimates for detecting a 0.5 cm change in height in children provided a caloric and non caloric micronutrient supplement using a design effect of 2, power of 0.80, alpha of 0.05 and assuming an attrition rate of 15%, the required sample size per group was 301.

Mother-infant pairs will be recruited between infant age of 0 to 3 months to participate in monthly nutrition educations sessions and to encourage the women to continue exclusive breast feeding across all three groups. At 6 months of age, infants in each of the three groups were enrolled into the intervention study (upon receipt of informed consent). Data collection involves baseline, midline and endline measurements in the infants at 6, 12 and 18 months of age. In addition, participating mother-infant pairs will be visited weekly for delivery of the supplements and for morbidity monitoring and monthly for the measurement of anthropometry. Anthropometric measurements include: length (Infant/Child Shorr Height Board; Weigh and Measure, LLC ), weight (Seca 874 digital scale ), mid-upper arm circumference (MUAC) (Child MUAC Tape; Weigh and Measure, LLC), subscapular and triceps skinfolds (Holtain skinfold caliper ),head and chest circumference.

Data collection at baseline, midline and endline included one venous sample (3ml) from the infant, HemoCue (Model 301) measurement to assess severe anemia (7 <g/dl) with appropriate referrals as mentioned earlier. Ethylenediaminetetracetic acid (EDTA) Vacutainers (BD ; catalog number 368841) for whole blood and plasma analyses and Trace Element Serum Separator Tube Vacutainers (BD; catalog number 368380) for serum analyses will be used for the sample collection with butterfly needles (21 or 23 gauge). Samples will be immediately placed in a super cooler tube rack and transported back to the lab within five hours where they were processed immediately. Questionnaires will be administered to assess socio-economic status, infant and young child feeding practices, morbidity (past week), household food security along with a 24 hour diet recall. Data was uploaded daily through cellular network, stored on Formhub and then Ona organization servers.

The primary outcome of the study is the change in length for age Z-score from 6 to 18 months of age in infants in the KokoPlus group versus Micronutrient powder group and the Nutrition Education group. Data calculations included estimating anthropometric indices using the WHO (World Health Organization) 2006 growth reference charts using the WHO macro in STATA (18), computing the USAID Food and Nutrition Technical Assistance (FANTA) household insecurity access scores (HFIAS) (19) maternal body mass index (BMI), dietary diversity scores (20), and re-coding variables as required to binary and accounting for missing. Infant anthropometric indices calculated include length for age (LAZ) Z- score, weight for age Z-score (WAZ) and weight for length (WLZ score).

All analyses are intent to treat. Descriptive statistics (means, medians, standard deviations and standard errors) were computed. To verify the randomization assumption, any differences in mean values at baseline across three groups were tested using linear mixed effects regression analyses accounting for clustering. The effect of the supplement (KokoPlus, Micronutrients and Nutrition Education group) on different dependent variables across the intervention period will be tested using mixed effects regression analyses accounting for clustering and repeated measures. The dependent variables tested included change in LAZ between baseline and endline (primary outcome), change in LAZ on a monthly basis (primary outcome 2), change in WAZ, WLZ, serum hemoglobin (unadjusted and adjusted for inflammation), serum ferritin (unadjusted and adjusted for inflammation), serum zinc, serum cortisol, serum insulin growth factor-1 (IGF-1), serum retinol binding protein, C-reactive protein and alpha glycoprotein, prevalence of acute and chronic infection.

A cross-sectional assessment (anthropometry) will be conducted at baseline, midline and endline in 301 infants that are randomly selected from another set of communities (to be identified based on the same community selection criteria). These infants will not be followed longitudinally and the only measurements to be collected include weights and heights. Informed consent procedures will be similar to the three intervention arms. This group is a reference group only and cannot be included in any comparative analysis.

Monitoring of groups that receive a supplement will happen weekly. Compliance will be defined based on the number of supplement packets that are consumed per week. To assure that the mothers are compliant in using the supplement, they will be asked to return the empty supplement packages at the end of the week. Optimal compliance will be defined as consumption of at least 50% of the weekly samples. Supplements will be provided in plastic or paper bags with a clearly labelled household ID (identity number). Compliance and dose response to compliance will be reviewed in existing studies to determine minimum compliance required.