Conditional cash transfer and/or lipid-based nutrient supplement targeting the first 1000 d of life increased attendance at preventive care services but did not improve linear growth in young children in rural Mali: results of a cluster-randomized controlled trial

Abstract

ABSTRACT Background In 2014, the World Food Programme added to an ongoing health and nutrition program named “Santé Nutritionnelle à Assise Communautaire dans la région de Kayes” (SNACK), the distribution of cash to mothers and/or lipid-based nutrient supplement (LNS) to children aged 6–23 mo, conditional upon attendance at community health centers (CHCs) during the first 1000 d of life. Objective We evaluated the additional impact of the distribution of cash and/or LNS on mean height-for-age z scores (HAZ; primary outcome), stunting (HAZ < −2), and on intermediate outcomes along the program impact pathways. Methods In a cluster-randomized controlled trial using a 2 × 2 factorial design, 76 CHCs were randomly assigned to deliver either SNACK, SNACK + Cash, SNACK + LNS, or SNACK + Cash + LNS. Cross-sectional surveys among 12- to 42-mo-old children and their mothers were conducted at baseline (2013, n = 5046) and at endline (2016, n = 5098). Results Factorial analysis showed no interaction between cash and LNS treatments for HAZ, but found an antagonistic interaction for stunting (OR: 1.55; 95% CI: 1.05, 2.31; P = 0.03). There were no impacts of the cash, LNS, or cash + LNS treatments, compared with the SNACK alone, on either HAZ or stunting (treatment × time interaction). There were significant impacts of the LNS and cash + LNS treatments on attendance at ≥1 growth monitoring (GM) session (OR: 3.95; 95% CI: 1.69, 9.24; OR: 3.90; 95% CI: 1.73, 8.81, respectively) and half the expected sessions (OR: 4.72; 95% CI: 1.47, 15.17; OR: 5.25; 95% CI: 1.82, 15.11, respectively), mothers’ knowledge on importance of GM (OR: 1.98; 95% CI: 1.16, 3.39; OR: 3.12; 95% CI: 1.60, 6.09, respectively), and, only for the LNS group, appropriate timing for complementary feeding (OR: 1.62; 95% CI: 1.09, 2.41). Conclusions Implementation constraints and suboptimal participation in program activities may explain the lack of impact on child linear growth in this rural region of Mali. This trial was registered at www.isrctn.com as ISRCTN08435964.