Costing Methods for a Cluster‐Randomized Cost‐Effectiveness Trial Comparing the Performance of Four Supplementary Foods in Treating Sierra Leonean Children with Moderate Acute Malnutrition (MAM)

Abstract

It is widely recognized that cost‐effectiveness (CE) estimates generated by economic evaluations should enter into decision‐making for health and nutrition policy and programming. However, many existing evaluations of nutrition interventions focus on effectiveness, assessing comparative nutrition and health outcomes without considering their associated costs. Moreover, despite recognizing the importance of measuring costs in generating CE estimates, many researchers find costing to be a challenging task with limited guidance and few examples of cost data collection methods in the international nutrition literature. This communication aims to help guide researchers on costing methods for research for cost‐effective policy‐making and programming.As part of the Food Aid Quality Review Phase III (FAQR III) Project, a USAID/FFP‐funded project implemented by Tufts University, a comprehensive costing protocol was designed for a cluster‐randomized field trial in Sierra Leone evaluating the comparative cost‐effectiveness of treating children under‐5 with MAM with one of four iso‐caloric supplementary foods: Super Cereal Plus with amylase, Corn‐Soy Blend Plus given with fortified vegetable oil (FVO), an innovative formulation called Corn‐Soy‐Whey Blend given with FVO, and Ready‐to‐Use‐Supplementary Food. The primary effectiveness outcome is recovery from MAM defined as reaching MUAC ≥ 12.5 cm within the 12‐week treatment period.To determine CE, we will assess costs using three “lenses”. The Program Costs lens includes non‐research costs incurred as part of this trial, including fixed, start‐up and ongoing costs of food products, transportation, storage, programming, beneficiaries, personnel, equipment, materials and facilities. Monetary costs and opportunity costs incurred within these components will be estimated using data collected from key informant interviews, direct observation, time tracking forms as well as accounting documents and inventory records. The Adjusted Program Costs lens will be the estimated costs for an NGO seeking to replicate this program in a similar setting. Interviews with local NGO partners and market surveys will be conducted to estimate the costs that an NGO would face. The third lens will conduct Policy Experiments to estimate costs that may be incurred when moving from the current limited setting to larger‐scale and longer interventions.Within each lens, we will calculate summary cost estimates per study arm including total program cost, cost per child treated, and cost per child‐month treated. We will link these cost estimates to the primary effectiveness measure, recovery from MAM, to derive the CE estimate of cost per recovered child and compare CE among the four study arms. Both absolute CE measures (e.g. cost per child recovered for each arm) and relative CE measures (e.g. incremental cost‐effectiveness ratio comparing each arm to a reference arm) will be calculated.Support or Funding InformationThis research is made possible by the generous support of the American people through the U.S. Agency for International Development (USAID) Office of Food for Peace of the Bureau for Democracy, Conflict and Humanitarian Assistance, and the World Food Programme under the terms of the Contract AID‐OAA‐C‐16‐00020, managed by Tufts University. The contents are the responsibility of Tufts University and its partners in the Food Aid Quality Review (FAQR) and do not necessarily reflect the views of USAID, the United States Government, or the World Food Programme.