Evaluation of conditional cash transfers and mHealth audio messaging in reduction of risk factors for childhood malnutrition in internally displaced persons camps in Somalia: A 2 × 2 factorial cluster-randomised controlled trial

Carlos S. Grijalva-Eternod, Mohamed Jelle, Hani Mohamed, Katie Waller, Bishar Osman Hussein, Emmanuel Barasa, Andrea Solomon, Sajia Mehjabeen, Andrew Copas, Edward Fottrell, and Andrew J. Seal


This paper presents the results of a randomized trial to estimate the effects of conditional cash transfers (CCTs) and mobile health (mHealth) audio messages in IDP camps near Mogadishu, Somalia. Specifically, the research examined whether conditionality in cash transfer programs and mHealth audio message improved health-seeking behavior and reduced risk factors for malnutrition.  

The study covered over 1,400 households in 23 IDP camps. Households in all IDP camps received cash transfers made at an emergency humanitarian level (US$70 per household per month) for three months followed by transfers at a safety net level (US$35 per household per month) for a further six months.  

Camps were randomly selected to receive additional interventions delivered as part of the cash transfer program: (1) a cash transfer conditionality that required households to take children under five years of age to a single health screening at a local clinic where they were also issued with a home-based child health record card, which was a pre-requisite for registering for the cash transfer program; and (2) an mHealth intervention that sent audio messages about health and nutrition to the mobile phones of households twice a week for 9 months.  

Data was collected from households and individuals at baseline and monthly, including information on household demographics, water sanitation and hygiene, food security, household expenditure, mother/caregiver’s characteristics and knowledge of health and nutrition, vaccination coverage, child age and anthropometry (mid-upper arm circumference measurements and bipedal pitting oedema; this last one indicates malnutrition and was recoded in case an imprint remained in both feet after pressing them with the thumbs for 3 seconds), breastfeeding practices, child morbidity, and child mortality and verbal autopsy. Children were identified to be acutely malnourished if they had a mid-upper arm circumference (MUAC) less than 12.5 cm and/or oedema and referred to a health center for treatment.  

The baseline survey found that: 

  • Most households were female headed and 15 percent were in a polygamous arrangement. The average household size was five members. 
  • All households had access to piped water (a third paid for water access) and most households had access to pit latrines. A small proportion of households had access to handwashing facilities, and only a third had soap. 
  • On average, households had 1.9 meals per day, received food assistance from a humanitarian organization for 3 days in the week prior to the survey, and had high dietary diversity (7 of 12 food groups in 24 hours prior to the survey). 
  • Households had an average expenditure of US$87 in the 30 days prior to the survey, with food being the largest expenditure category. 
  • Mothers or caregivers had an average age of 30, 80-90 percent were illiterate, two-thirds had undertaken paid labor in year prior to the survey, and a quarter of mothers/caregivers and children slept under a mosquito net on the night prior to the survey. 
  • Two thirds of children were ill in the 4 weeks prior to the survey, and acute malnutrition affected 8 percent of children.

Empirical findings: 

  • Cash conditionality significantly improved the coverage of measles and pentavalent vaccination, but timely vaccination for all antigens did not improve. Cash conditionality improved coverage of measles vaccination from 39 percent to 78 percent and pentavalent vaccination from 44 percent to 78 percent. At the end of the study, coverage remained elevated at 82 percent and 87 percent, respectively. There were not any significant changes in timely vaccination, household food and non-food expenditure, household dietary diversity, or household reliance on coping strategies. 
  • The mHealth intervention did not improve mother’s knowledge, any vaccination outcomes, or child diet diversity, although it was associated with a higher household diet diversity. The mHealth intervention did not improve measles vaccination, pentavalent vaccination, or timely vaccination. There was no change in the incidence of child mortality, acute malnutrition, diarrhoea, exclusive breastfeeding, or measles infection. While there was no evidence that mHealth increased a mother’s knowledge score, household dietary diversity increased from a mean of 7 to 9. However, this was not reflected by a significant increase in the child diet diversity score, which increased only slightly from 3.2 to 3.6.  

The authors conclude that conditional cash transfers can achieve public health benefits in humanitarian cash transfer programs by increasing the uptake of child vaccination services. The mHealth intervention was associated with an increase in household diet diversity, despite no evidence that knowledge had improved.