Sophie Sarrassat
Nicolas Meda
Hermann Badolo
Moctar Ouedraogo
Henri Some
Robert Bambara
Joanna Murray
Pieter Remes
Matthiew Lavoie
Simon Cousens
Roy Head
Publication Date
March 6, 2018

Centre for Maternal Adolescent Reproductive and Child Health (MARCH), London School of Hygiene and Tropical Medicine - LSTHM (Sarrassat, Cousens); Centre Muraz (Meda, Badolo); Africsanté (Ouedraogo, Somé); Direction Generale des Etudes et des Statistiques Sectorielles (DGESS), Ministere de la Santé (Bambara); Development Media International - DMI (Murray, Remes, Lavoie, Head)

"We found no evidence of an effect of a mass media campaign on child mortality....Nevertheless, this study provides some of the best evidence available that a mass media campaign alone can increase health facility utilisation for maternal and child health in a low-income, rural setting."

From March 2012 to January 2015, Development Media International (DMI) implemented a standalone radio campaign to address key family behaviours for improving under-5 child survival in rural Burkina Faso. (Further information about this campaign and its outputs, as well as midline data, can be found at Related Summaries, below.) A randomised controlled trial was conducted to assess the effect of the radio campaign on all-cause post-neonatal under-5 child mortality and behaviours at endline - i.e., after 32 months of campaigning.

As reported here, a review of evaluations of mass media interventions for child-survival-related behaviours done between 1960 and 2013 in low- and middle-income countries concluded that media-centric campaigns can positively affect a wide range of child health behaviours. However, much of the evidence for an effect comes from non-randomised designs, and the limited number of randomised studies that have been reported have often failed to demonstrate an effect. Hence, the present study.

Behaviour change interventions encompass a wide range of approaches, including interpersonal-based, community-based, media, and social marketing approaches. Compared with other approaches, mass media campaigns have the potential to reach a large audience at relatively low cost. So, for DMI's Burkinabé trial (a description of the theory of change and the Saturation+ methodology used to design and implement the campaign is provided at the links below), short spots, of 1 minute in duration, were broadcast approximately 10 times per day, and 2-hour, interactive long-format programmes were broadcast 5 days per week. All materials were produced in the predominant local languages spoken in each intervention cluster. Women of reproductive age and caregivers of children younger than 5 years were the main focus of the campaign, which covered 17 behaviours along the continuum of care. The spots were based on message briefs that DMI drew up for each target behaviour. The long-format programmes were followed by phone-ins to allow listeners to comment on the issues raised. Behaviours covered by spots changed weekly, while the long-format programme covered 2 behaviours a day and changed daily. During the trial period, no other radio campaigns related to child survival and of comparable intensity were broadcast in any of the clusters (distinct geographical areas in rural Burkina Faso with at least 40,000 inhabitants) included in the trial.

In the repeated cross-sectional, cluster randomised trial, conducted by an independent team from the London School of Hygiene & Tropical Medicine (LSHTM) and Centre Muraz in Burkina Faso, 14 clusters were selected by DMI based on their high radio listenership (>60% of women listening to the radio in the past week) and minimum distances between radio stations to exclude population-level contamination. Clusters were randomly allocated to receive the intervention (the radio campaign) (n=7) or control group (n=7) (no radio campaign). The average number of villages included per cluster was 34 in the control group and 29 in the intervention group. Household surveys were performed at baseline (from December 2011 to February 2012), midline (in November 2013, and after 20 months of campaigning), and endline (from November 2014 to March 2015, after 32 months of campaigning). Women were told that these surveys were about their children's health, without any mention of the radio campaign.

The primary outcome was all-cause post-neonatal under-5 child mortality, the secondary outcome was all-cause under-5 child mortality, and intermediate outcomes included the coverage of the campaign (as measured by the proportion of mothers who reported listening to the campaign) and the 17 family behaviours targeted by the campaign, as measured by the proportion of mothers who reported a given behaviour during interviews and the number of attendances at primary health facilities. The trial was designed to detect a 20% reduction in the primary outcome with a power of 80%. Routine data from health facilities were also analysed for evidence of changes in use, and these data had high statistical power. The indicators measured were new antenatal care attendances, facility deliveries, and under-5 consultations.

Selected results:

  • 2,269 (82%) of 2,784 women in the intervention group reported recognising the campaign's radio spots at endline.
  • Post-neonatal under-5 child mortality decreased from 93.3 to 58.5 per 1,000 livebirths in the control group and from 125.1 to 85.1 per 1,000 livebirths in the intervention group. Thus, there was no evidence of an intervention effect (risk ratio 1.00, 95% confidence interval (CI) 0.82-1.22; p>0.999).
  • In the first year of the intervention, under-5 consultations increased from 68,681 to 83,022 in the control group and from 79,852 to 111,758 in the intervention group. The intervention effect using interrupted time-series analysis was 35% (95% CI 20-51; p<0.0001). There was a mean 23.7% increase in all-cause consultations for children over 3 years of the intervention.
  • New antenatal care attendances decreased from 13,129 to 12,997 in the control group and increased from 19,658 to 20,202 in the intervention group in the first year (intervention effect 6%, 95% CI 2-10; p=0.004). There was a mean 7.7% increase in antenatal care attendance over 3 years of the intervention.
  • Deliveries in health facilities decreased from 10,598 to 10,533 in the control group and increased from 12,155 to 12,902 in the intervention group in the first year (intervention effect 7%, 95% CI 2-11; p=0.004). There was a mean 7.3% increase in facility deliveries over 3 years of the intervention.

In short, the study found no evidence of an effect of DMI's mass media campaign on child mortality but strong evidence of an effect on treatment-seeking behaviours.

The behavioural survey found no evidence of an effect of the campaign on self-reported habitual behaviours, such as child feeding practices, handwashing, and child stool disposal practices, but the behavioural survey was severely underpowered. For example, the behavioural survey had a sample size of 525 children with pneumonia, whereas the routine data had an annual sample of 44,591 cases. For this reason, the authors relied more on the routine data, which gave strongly positive, statistically significant results for every indicator measured. It is also true, however, that the campaign's broadcasts were heavily weighted to care seeking rather than home-based behaviours, and it might be more difficult to achieve sustained changes in habitual behaviours that need to be performed daily with little obvious immediate benefit.

The study also found that the intervention was associated with an increase in care-seeking in facilities (among those who lived less than 2.5 km away from a facility, but not those who lived further away); why, then did it not detect any evidence of a reduction in mortality? The most probable reason is that the study was not sufficiently powered: It was designed (in 2010) to detect a mortality reduction of 20% with 80% statistical power, and 54% power to detect a 15% reduction. During the trial, there were steep falls in mortality (from 98 to 55 per 1,000 in control clusters), major differences between control and intervention clusters, and large heterogeneity between clusters (with mortality rates ranging from 56 to 156 per 1,000). All of these factors reduced statistical power even further, leaving almost no power to detect a mortality reduction of 10% or less. It is possible to speculate that if the quality of care received at the facility was low, this could limit any mortality reduction through increased care seeking, implying that interventions that generate demand may need to be coupled with supply-side efforts to in order to achieve mortality reductions.

In two subsequent papers, DMI, LSHTM, and Centre Muraz will provide further, diagnosis-specific evidence from the routine data. These results provide additional support for the effect of the campaign on behaviours and enable the use of the Lives Saved Tool (LiST) to calculate the likely reduction in mortality resulting from these behaviour changes. From these results, the costs per life saved can be estimated, using a number of economic measures.

According to the authors: "More exciting findings from the trial to be published later this year [2018]. So stay tuned!"


Lancet Global Health, Volume 6, No. 3, 3330-i341 - sourced from: Making Waves: Can Radio Reduce Child Mortality, by Tim Colbourn and Audrey Prost, Lancet Global Health, Volume 6, No. 3, e238-e239; GiveWell, accessed on March 7 2018; and emails from Roy Head and Sophie Sarrassat to The Communication Initiative on March 8 2018 and March 9 2018. Image credit: DMI