An observational analysis of the impact of indoor residual spraying in two distinct contexts of Burkina Faso

Abstract

Background: Indoor residual spraying (IRS) is a cornerstone malaria control intervention in Burkina Faso. From 2018 to 2021, non-pyrethroid IRS was implemented annually in two regions of Burkina Faso with distinct malaria transmission patterns, concurrently with annual seasonal malaria chemoprevention (SMC), and a mass insecticide-treated net (ITN) distribution in 2019.

Methods: A retrospective quasi-experimental approach was used to evaluate the impact of the 2018, 2020, and 2021 IRS campaigns on routinely reported confirmed malaria case incidence at health facilities. The 2019 campaign was excluded due to lack of data reporting during a health sector strike. Controlled interrupted time series models were fit to detect changes in level and trend in malaria case incidence rates following each IRS campaign when compared to the baseline period 24-months before IRS. IRS districts Solenzo (Sudano-Sahelien climate), and Kampti (tropical climate) were compared with neighbouring control districts and the analyses were stratified by region. Modelled health facility catchment population estimates based on travel time to health facilities and weighted by non-malaria outpatient visits were used as an offset. The study period encompassed July 2016 through June 2022, excluding July 2018 to June 2019.

Results: District-level population and structure coverage achieved by IRS campaigns was greater than 85% in 2018, 2020, and 2021 in Solenzo and Kampti. In Solenzo a significant difference in malaria case incidence rates was detected after the 2018 campaign (IRR = 0.683; 95% CI 0.564-0.827) when compared to the control district. The effect was not detected following the 2020 or 2021 IRS campaigns. In Kampti, estimated malaria incidence rates were between 36 and 38% lower than in the control district following all three IRS campaigns compared to the baseline period.

Conclusions: Implementation of IRS in Kampti, a tropical region of Burkina Faso, appeared to have a consistent significant beneficial impact on malaria case rates. An initial positive impact in Solenzo after the first IRS campaign was not sustained in the successive evaluated IRS campaigns. This study points to a differential effect of IRS in different malaria transmission settings and in combination with ITN and SMC implementation.

Keywords: Combined malaria control strategies; Indoor residual spraying; Observational analysis.

Fig. 1

Location of IRS and control districts. Solenzo and Nouna are located in the Boucle du Mouhoun region; Kampti and Gaoua are located in the Sud-Ouest region. Black points represent health facilities that were included in the primary analysis

Fig. 2

Boxplots of population and spray coverage by facility catchment area achieved during the 2018, 2020, and 2021 IRS campaigns in Kampti and Solenzo districts. The horizontal dashed line indicates the 85% coverage target recommended by the WHO. Spray coverage is calculated as the number of structures sprayed divided by the number of structures found. Population coverage is calculated as the population protected divided by the sum of the population protected and the population not protected. Population and spray coverage data were collected by spray teams during the campaign. Population coverage data was not available from the 2018 campaign

Fig. 3

Modelled catchment population sizes and annual malaria case incidence per 1000 estimated population for 144 health facilities in Boucle du Mouhoun and the Sud-Ouest regions. Population and incidence estimates shown are for the year 2018

Fig. 4

Confirmed malaria cases per 1000 estimated population from July 2016 to June 2022. Solid lines indicate incidence at the district level, and points represent values reported by each facility. The dates of the 2018, 2020, and 2021 IRS campaigns are indicated by black dashed lines, and the grey bands indicate the period of data omitted from June 2019 to May 2020 due to a health sector strike and data unavailability

Fig. 5

Model-predicted confirmed malaria cases per 1,000 estimated population in the Boucle du Mouhoun and Sud-Ouest regions. In the time-series line graphs, model predictions are represented by solid lines and observed values are represented by points. Vertical dashed lines indicate the conclusion of IRS campaigns in June 2018, 2020, and 2021. The x-axis represents months from the start of the study. The pre-IRS baseline period is July 2016 to June 2018 for all plots. The post-IRS period for plots (A) and (B) is July 2018 to May 2019. The post-IRS period for plots (C) and (D) is July 2020 to June 2021. The post-IRS period for plots (E) and (F) is July 2021 to June 2022. Bar charts display modelled annualized case incidence. IRS districts are colored red and control districts are in grey