Background: The global burden of acute malnutrition among children remains high, and prevalence rates are highest in humanitarian contexts such as Niger. Unconditional cash transfers are increasingly used to prevent acute malnutrition in emergencies but lack a strong evidence base. In Niger, non-governmental organisations give unconditional cash transfers to the poorest households from June to September; the ‘hunger gap’. However, rising admissions to feeding programmes from March/April suggest the intervention may be late.

Methods/design: This cluster-randomised controlled trial will compare two types of unconditional cash transfer for ‘very poor’ households in ‘vulnerable’ villages defined and identified by the implementing organisation. 3,500 children (6–59 months) and 2,500 women (15–49 years) will be recruited exhaustively from households targeted for cash and from a random sample of non-recipient households in 40 villages in Tahoua district. Clusters of villages with a common cash distribution point will be assigned to either a control group which will receive the standard intervention (n = 10), or a modified intervention group (n = 10). The standard intervention is 32,500 FCFA/month for 4 months, June to September, given cash-in-hand to female representatives of ‘very poor’ households. The modified intervention is 21,500 FCFA/month for 5 months, April, May, July, August, September, and 22,500 FCFA in June, providing the same total amount. In both arms the recipient women attend an education session, women and children are screened and referred for acute malnutrition treatment, and the households receive nutrition supplements for children 6–23 months and pregnant and lactating women. The trial will evaluate whether the modified unconditional cash transfer leads to a reduction in acute malnutrition among children 6–59 months old compared to the standard intervention. The sample size provides power to detect a 5 percentage point difference in prevalence of acute malnutrition between trial arms. Quantitative and qualitative process evaluation data will be prospectively collected and programme costs will be collected and cost-effectiveness ratios calculated.