Randomized controlled trial of an agricultural livelihood intervention to improve maternal and newborn health and nutrition in Kenya

PROJECT SUMMARY Food insecurity is a critical driver of maternal and infant health and nutrition, including poor birth outcomes, suboptimal breastfeeding, perinatal depression and stress, and poor child growth and development. Pregnant women living with HIV are particularly vulnerable to food insecurity and face an excess risk of poor birth and infant outcomes. In sub-Saharan Africa, where both food insecurity and HIV are highly prevalent and a third of children under five are stunted, interventions to reduce food insecurity and malnutrition that are relevant for both women with and without HIV are needed. In rural settings in this region, small-scale farming is the primary source of livelihood, yet unpredictable rainfall, severe climate events, and limited irrigation hamper crop yields. Agricultural livelihood interventions are a promising approach to raising income, bolstering food security, and ultimately improving maternal and infant health and nutrition. Yet studies of agricultural interventions initiated in pregnancy are lacking despite the fact that in utero exposures crucially predict pregnancy and infant outcomes. The objectives of this proposal are to determine the effectiveness of an agricultural livelihood intervention on improving maternal and infant health when initiated in early pregnancy, and to understand factors that influence implementation of an agricultural intervention in the perinatal period, including the need for farming support when pregnancy impacts women’s capacity to work in the fields. We propose a hybrid effectiveness- implementation trial among 410 pregnant women enrolled at ≤20 weeks gestation, half living with HIV in western Kenya. Women will be randomized 1:1 the intervention or routine care. Those randomized to the intervention will receive guidance on obtaining farming support. Our central hypothesis is that by empowering pregnant women with skills and tools for sustainable farming and perinatal nutrition, the intervention will lead to improved maternal and infant health compared to control participants. We propose an HIV status neutral approach to optimize the generalizability and potential reach of this intervention. Also, because HIV stigma and poor health present additional barriers to empowerment and healthy behaviors, this intervention, which may reduce these barriers, has the potential to alleviate infant health disparities associated with maternal HIV. We propose the following aims: Aim 1: Determine the impact of the intervention on maternal, pregnancy and infant health outcomes, among women living with and without HIV and their newborns. Aim 2: Determine the impact of the intervention on empowerment, socioeconomic, and behavioral factors that may influence maternal and infant health. Aim 3: Identify attitudes, norms, processes, and resources that influence implementation outcomes and effectiveness of the intervention initiated in early pregnancy. This research will significantly advance scientific understanding of whether, how, and under what circumstances an agricultural intervention for pregnant women affects health.