This study aims to provide National Malaria Control Programs (NMCP), international donors and other key stakeholders with clear evidence on the impact and cost-effectiveness of using indoor residual spraying (IRS) with a non-pyrethroid insecticide in a high malaria transmission area that has universal long-lasting insecticidal net (LLIN) coverage. This is an interventional study with IRS serving as the research intervention.

The district of Mopeia, in the province of Zambezia, Mozambique will be the study site. This is a high transmission area with a malaria parasite prevalence of 54% in children. The Ministry of Health distributed LLINs in Mopeia in 2014-2015.

The NMCP through funding from President's Malaria Initiative Africa Indoor Residual Spraying Project (PMI-AIRS) was able to cover half a district with indoor residual spraying. A simplified census took place in mid-2016 to determine the number of children five years of age and under in the district and enumerate and map the households to assist in implementation.

From the 115 villages/bairros existent in Mopeia, 86 clusters were randomized in a government randomization ceremony to either receive IRS with Actellic or maintain no IRS. The IRS was implemented through a partnership between the NMCP and PMI-AIRS according to standard operational and consent procedures. From each cluster, a cohort of 18 children five years of age and under will be followed monthly to assess malaria incidence at the community level in both IRS and non-IRS villages. There will be 774 children in the IRS villages and 774 children in the no-IRS villages (total cohort will be 1548). Additionally, the routine health centre reporting system will be strengthened to assess malaria incidence in children five years of age and under by passive case detection. Three cross sectional studies in April 2017, April 2018, and April 2019 will assess changes in net use, health seeking behaviour and malaria prevalence at the community level.

Entomological data will be collected from both IRS and non-IRS areas to assess the vector dynamics and insecticide resistance pattern of the local vector populations from sprayed and unsprayed areas. Data on the costs of the implementation as well as health-related expenditures at health system and household levels will be collected prospectively throughout the study. These costs will be determined using both health system and societal perspectives.

The incidence rate in IRS and no-IRS areas will be combined with the micro-costing data to calculate the cost per case averted at community and health facility level.

These findings will be disseminated to the NMCP and international donors and stakeholders to complement the World Health Organization (WHO) guidance on combining indoor residual spraying and long-lasting insecticidal nets.

Mopeia is a district in the Zambezia Province. Mopeia borders the district of Morrumbala to the North, the district of Chinde and the province of Sofala to the South, the districts of Nicoadala and Inhassunge in the East and the Provinces of Sofala and Tete in the West. It has an area of 7671 km2. The projected population for 2016 is 162.188 individuals with 31.927 (19.7%) under five years of age (National Institute of Statistics, Mozambique). There are three administrative posts, eleven localities and 224 villages (Bairros) and approximately 34.603 households. There are 12 health facilities (PMI-AIRS Mozambique, unpublished data). There is little socio-economic data available from Mopeia.

The malaria burden is high in Zambezia with a parasite prevalence of 54% in under-fives [19]. The parasite prevalence in children 1-15 years of age in Mopeia is 47.8% (38.7%-57.1%) [20]. The mean rapid diagnostic test (RDT) positivity rate during a recent enhanced surveillance exercise at health facilities was 62.8% (range 50-72%) [21]. The same data suggest high incidence at health facilities, showing 470 cases per 1000 children during the same period (June-Nov 2014). Mopeia recently received 175.297 LLINs in 2013 and IRS with pyrethroids in 2014 [22]. Residents of Mopeia will receive new LLINs in early 2017.

Data from February 2015 in the neighbouring districts of Mocuba and Morrumbala show Pyrethroid resistance in the local Anopheles gambiae s.l. population [23]. Further north, in the district of Milange, tested Anopheles gambiae s.l. remain susceptible to pyrethroids [23]. See table 2 below for further details. Mopeia has been selected to receive IRS because of its high malaria transmission intensity, the presence of LLINs in the district, and the aforementioned regional indications of reduced pyrethroid susceptibility in the target vector population, and the existence of IRS infrastructure and capacity from previous campaigns.

The villages selected for spraying will receive IRS with Actellic according to standard operating and consent procedures [24] in addition to existing LLINs. Non IRS areas will have existing LLINs, but will not receive IRS. In late 2017 (year two), IRS will be repeated using the same village selection and insecticide. Additionally in 2017, the whole district will be subject to universal LLIN distribution.

Individuals enrolled in the cohort were recruited independent of their household acceptance or refusal of spray. IRS status will be confirmed with questionnaires during the monthly active cohort visits and during the cross-sectional surveys, including questions about wall replastering/painting and net usage (SSPs COST 001, 002 and 003). Additional information will be obtained through cone bioassays.

The standard of malaria care at community and health center will remain unchanged throughout the study and stock levels of malaria commodities will be ensured.

This study of IRS implementation in Mozambique will provide detailed information of the impact and cost-effectiveness of adding IRS with an extended release formulation of the organophosphate insecticide pirimiphos-methyl (Actellic®300CS) in a high transmission area with high LLIN coverage. This information will be disseminated to the NMCP as well as to local and international stakeholders and decision makers to inform policy recommendations and choices regarding the combination of vector control strategies.

A simplified census was conducted in June-July 2016 to obtain the total number of children five years of age and under per household and village. A randomization ceremony for the intervention took place to randomly assign the villages to receive IRS or maintain their current status. Participants in both IRS and non-IRS areas were randomly chosen and consented to participate in cohorts to be followed prospectively. The total population disaggregated in under five years of age and above five in each village in the district will be used as a denominator for the passive case detection component of the study. Each house was geopositioned during the census visit as part of the household enumeration for spraying. This information can assist in defining the cluster size, core and buffer areas. Each household will receive a unique permanent ID. This is an interventional study with IRS serving as the research intervention.

To determine the incidence by active case detection, a cohort of children five years of age and under will be followed monthly from the core zone of each cluster. During each visit the care taker will answer a short questionnaire regarding health, net usage and health-related expenditure. The temperature of each recruited child will be recorded and an RDT performed. If the child has a positive RDT, irrespective of accompanying clinical symptoms he/she will receive treatment according to the national guidelines.

At health facility level, a separate facility-based team will ensure the collection of the household location (village) of each malaria case to determine the incidence in the different study clusters by passive case detection. Joint work with the community health workers (Agentes Polivantes Elementares [APEs]) will also strengthen quality of their data and they will be asked to include information on the location (village) of each malaria case they diagnose/treat.

Costing data will be prospectively collected using standardized data collection tools to determine the cost of interventions (IRS and LLIN distribution) and of care-seeking.

Cross-sectional studies will be carried out at the peak of transmission season in 2017, 2018, and 2019. There is a potential for the monthly visits for active case detection to have an influence on household behavior and expenditure. Additional data on house expenditure will be collected during the cross-sectionals to assess societal costs of malaria care that is independent from study visits.

Entomological data including mosquito densities, sporozoite rates, resistance status and indoor/outdoor biting ratios will be sampled from IRS and non-IRS villages throughout the study following standard PMI procedures.

Inclusion Criteria:

• all consenting adults, assenting minors (12-18) and caregivers of children under 12

Exclusion Criteria:

• all infants 0-6 months of age