Evaluating the Effectiveness of a Nutrition and Agricultural Linkages Project in Western Kenya: Design of the Mama SASHA Cohort Study of Vitamin A (COVA).

Objectives: The 2013 Lancet Maternal and Child Nutrition series identified rigorous evaluations of nutrition-sensitive agricultural interventions as a research priority. The Mama SASHA study in Western Kenya links delivery of vitamin A (VA) rich orange-flesh sweet potato (OFSP) vines to antenatal care to improve VA and nutritional status of pregnant and lactating women and their children. Methods: In addition to cross-sectional surveys in intervention and control communities at baseline and endline, the evaluation strategy includes a nested longitudinal study that follows women and their infants from pregnancy through 9 months postpartum. VA status of mothers and their infants is assessed during four visits using infection-adjusted plasma retinol binding protein and breast milk retinol (postpartum). Maternal and child iron and anemia status, anthropometry, dietary intakes, agricultural practices, health services uptake, household food security and program uptake are also measured Results: 505 eligible pregnant women, attending ANC at 4 control and 4 intervention facilities, were consented and enrolled. At enrollment overall prevalence of infection adjusted vitamin A deficiency was 21.8%. Women in control and intervention communities did not differ with respect to VA, iron, anemia or anthropometric status; household food security or dietary diversity scores; demographic characteristics; awareness of vitamin A; or consumption of vitamin A rich foods in the past 7 days. Only 10 women had consumed OFSP in the previous 7 days; all in intervention communities. Conclusions: The longitudinal study will contribute to rigorous impact evaluation of the OFSP intervention on maternal and child VA status and allow assessment of program impact pathways.

Prevalence of Vitamin a Deficiency among Infants Participating in the Mama SASHA Proof of Concept Project in Western Kenya.

Objectives: Vitamin A (VA) deficiency (VAD) is a significant burden among children under-5- years-old and pregnant and lactating women in sub-Saharan Africa. We assessed the levels of VA and prevalence of VAD among children age 6-23 mo in 2 counties in western Kenya. Methods: Dry-Blood-Spot (DBS) samples were obtained from 1838 infants in a community-based cross-sectional survey. Laboratory analysis of retinol-binding protein (RBP) and C-reactive protein (CRP) was carried out using a rapid EIA to estimate VA and subclinical inflammation statuses, respectively. A validation of DBS as a matrix using 60 matching serum-DBS samples was carried out. Values were adjusted for influence of inflammation using CRP (CRP, >5 mg/L) and population prevalence of VAD (RBP<0.825 μmol/L, biologically equivalent to 0.70 μmol/L retinol) estimated. Results: Mean (geometric±SD) concentration of RBP was adequate (1.56±0.79 μmol/L) with inflammation-adjusted mean (±SE) prevalence of VAD being high (18.3±1.1%). The level of CRP was within normal range (1.06±4.95 mg/L) whilst 18.4±0.9% had subclinical inflammation (CRP>5 mg/L). VAD was not associated with child sex (Chi-squared, Χ2=0.42, P=0.51), child nutritional status (wasting (P=0.68) and stunting (P=0.91), reported child intake of VA capsule within the past 1 year (P=0.84), maternal VA nutritional knowledge (Χ2=0.10, P=0.80), or reported maternal intake of VA capsule within 2 months of delivery (P=0.27). Older children had a 10% increased risk of VAD. Conclusions: Prevalence of VAD in this sample of infants was high irrespective of intake of VA capsule or maternal VA nutrition knowledge. A sustainable food-based intervention in this area of western Kenya to combat VAD especially in pregnant women and infants is warranted.

The Effects of Selling Micronutrient Powders on Morbidity Incidence among Preschool Children in Western Kenya.

Objectives: Although micronutrient powders (MNPs) are considered the preferred approach for childhood anemia control, concerns about iron-related morbidity from clinical trials challenge programmatic scale-up. We aimed to measure the association of community-based sales of MNPs with incidence of diarrhea, fever, and cough in children 6-35 months. Methods: The study was a community-based, cluster-randomized trial in 60 villages in western Kenya. MNPs and other health products (e.g., insecticide-treated bednets, soap, water disinfectant) were marketed in 30 intervention villages from July 2007-March 2008, followed by expansion of the intervention to both arms. Biweekly household visits monitored self-reported MNP use and morbidity. Using morbidity as the outcomes and MNP as the exposure variable, relative risks were estimated via generalized mixed models with random intercepts for repeated measures within a child and for multiple children from a household, while accommodating potential confounders as fixed effects. Results: Of the 1079 children enrolled, 881 (81.6%) had data on MNP use; average intake in intervention villages was 0.9 sachets/week. MNP intake was captured as continuous mean sachets and categories of use, but no significant effects were observed on incidence of diarrhea, fever or cough. Conclusions: Increased MNP use in western Kenya through community-based marketing did not result in increased incidence of infectious morbidity among young children. Integrated distribution of MNPs with other health interventions should be explored further in settings with high child malnutrition and infection burden.