Impact of Long-lasting Insecticidal Nets (LLINs) Treated With Pyrethroid Plus Pyriproxyfen vs LLINs Treated With Pyrethroid Plus Piperonyl Butoxide on Malaria Incidence in Uganda: a Cluster-randomised Trial

This rigorous, cluster-randomised trial will evaluate the impact of long-lasting insecticidal nets (LLINs) distributed in Uganda through the 2020-21 national universal coverage campaign. A cluster has been defined as the catchment area of a Malaria Reference Centre (MRC). A total of 64 clusters have been included in the study, covering 32 high malaria burden districts in Uganda where IRS is not being implemented. Clusters have been randomised in a 1:1 ratio in blocks of two by district to receive one of two types of LLINs: (1) pyriproxyfen (PPF) LLINs (Royal Guard) [n=32] and (2) piperonyl butoxide (PBO) LLINs (PermaNet 3.0) [n=32].

The intervention, including delivery of the LLINs and social and behaviour change communication (SBCC), will be led by the Ugandan National Malaria Control Division (NMCD) and other stakeholders. Currently, LLINs are scheduled to be delivered in the study areas from October 2020 to January 2021. The evaluation will include health facility surveillance at the MRCs to generate continuous estimates of malaria incidence for each MRC catchment area, cross-sectional community surveys at baseline (if additional resources are available), and at 12- and 24-months after LLIN distribution to gather information on net survivorship and use, and parasite prevalence in children 2-10 years of age, entomology surveys, and assessment of net durability and efficacy. The primary outcome of the trial will be malaria incidence as estimated using the health facility surveillance.

For each cluster the study will use a 'fried egg' approach for delivering the intervention ('egg white') and measuring our outcomes ('egg yolk'). The 'white' of the egg will include one sub-county per cluster, where the MRC is located. PPF LLINs and PBO LLINs will be distributed to the designated sub-county, as allocated in the randomisation. The 'yolk' of the egg will be the catchment area directly surrounding each MRC, where care-seeking at the MRC is expected to be high (i.e. if someone within the catchment area develops malaria, they are likely to seek care at the MRC). To determine the population of the MRC catchment areas, and to generate a sampling frame for the community surveys, the investigators will do the following: (1) define the catchment area of each MRC before the onset of the trial using data on village of residence from patients attending the MRCs, (2) map and enumerate all households within the MRC catchment areas before the 12-month community survey, and (3) conduct a census survey within each MRC catchment area to generate an accurate estimate of the study population in which study outcomes will be measured concurrently with the 12-month community survey.