Figure 1 illustrates the proposed scheme of the stepped wedge cluster-randomised controlled trial for the assessment of the targeted FBPM and its associated components of SBCC and home fortification based on the original design of a 1 year data collection period with 3 steps having 2 measurements each. In this scheme, data collection is done at 2 month intervals but change over from control to intervention happens at 4 month intervals.
 

Figure 1: Schematic diagram of the proposed stepped wedge cluster-randomised controlled trial

Cluster Baseline Step 1 Step 2 Step 3
6
5
4
3
2
1
Measurements Measurement 1 Measurement 2 Measurement 3 Measurement 4 Measurement 5 Measurement 6
Time (month) 0 1 2 3 4 5 6 7 8 9 10 11 12
  Control   Intervention

 

This diagram illustrates the predictable components of the MAM prevention package/s namely the targeted FBPM, SBCC and home fortification.

The study will also investigate the impacts of the addition of the blanket FBPM or the eBSFP during periods of emergency (GAM rate > 20%). Since this intervention is triggered externally based on current GAM prevalence rates, the timing of its introduction is not predictable. We assume that on any given 1 year period, it would be likely that an emergency will be declared / diagnosed based on WFP’s rapid assessment mechanisms. Once this happens and a blanket FBPM or eBSFP is initiated in one or more of our study areas or clusters, we will consider this as a ‘sub-step’ within the stepped wedge design which would allow us to analyse how existing programming (whether at ‘baseline’ or with targeted FBPM) is effected by the blanket FBPM or eBSFP.

To give an example, let’s suppose that on the 3rd month of the study, clusters 2, 4 and 5 have been assessed as having GAM prevalence > 20% which prompted the provision of blanket FBPM or eBSFP for a period of 3 months. The emergency eventually abated and blanket FBPM and eBSFP was stopped. However, this emergency prompted state officials and WFP to initiate targeted FBPM right away.

Given this scenario, the study scheme will continue as planned but now will be able to collect information on clusters 2, 4 and 5 at a period in which blanket FBPM or eBSFP is being implemented. Also, the study will just adjust its data collection such that areas or clusters that have switched to targeted FBPM earlier than scheduled due to the emergency will then be considered intervention areas or clusters accordingly. This example is illustrated in Figure 2 below.
 

Figure 2: Schematic diagram of the proposed stepped wedge cluster-randomised controlled trial in response to an emergency

Cluster Baseline Step 1 Step 2 Step 3
6
5
4
3
2
1
Measurements Measurement 1 Measurement 2 Measurement 3 Measurement 4 Measurement 5 Measurement 6
Time (month) 0 1 2 3 4 5 6 7 8 9 10 11 12

 

  Control   Intervention   Emergency   Early switch