Timing matters: combining insights from a private sector indoor residual spraying programme with mapped malaria seasonality patterns across Malawi.

HoekSpaans, Remy (2023) Timing matters: combining insights from a private sector indoor residual spraying programme with mapped malaria seasonality patterns across Malawi., Thesis (Doctoral), Liverpool School of Tropical Medicine.

Text R Spaans_Thesis_V3.pdf - Accepted Version Restricted to Repository staff only until 23 July 2025. Download (20MB)

Abstract

Introduction: Global progress towards malaria elimination has been slowly reversing over the past few years, and Malawi is no exception. Routinely collected malaria data can help to identify where and when additional interventions should be deployed. Indoor Residual Spraying (IRS) is a key intervention for Malawi, but it is not always clear under what circumstances IRS can have the most impact on malaria cases. This thesis will take lessons learned from a long-running private IRS campaign by Illovo and explore how they can be applied to malaria control activities in Malawi. Objectives are 1) to evaluate whether IRS is effective at Illovo using routine health data, and 2) if using finer-scale data would provide added detail on the optimal timing of IRS. Finally, we look at seasonal malaria patterns at a national level to 3) map malaria seasonality for informed use of
malaria interventions.

Methods: Chapter 2 evaluates the IRS programme run by Illovo over the period 2015 –2018, using monthly aggregated malaria case data collected at clinic level, as is routinely entered through the digital health information system (DHIS2). For Chapter 3, we digitized original health records from Illovo, which resulted in a dataset with daily data at the individual level, with village-level coordinates. A general additive model (GAM) was used to model the effect of IRS over time as a smooth term with an interaction-term for IRS round. This fine scale of the data allowed estimation of the duration of protection and optimal timing of IRS. Chapter 4 uses monthly DHIS2 data for 2017 – 2020, fitting a spatio-temporal INLA-R model to the proportion of malaria cases for each month. Kullback–Leibler divergence was used to characterise transmission patterns.

Results: Using monthly data, IRS was found to be highly effective in reducing case incidence rates at Illovo with an adjusted IRR of 0.38 (95% CI: 0.32 – 0.45). Using a GAM with daily data, a more detailed picture emerges; the odds of an individual testing positive for malaria were reduced for up to 124 days post-IRS compared to pre-IRS and locations where IRS was not implemented (P< 0.001), but this effect reversed for later time points. The effects of separate IRS rounds on malaria test positivity were an OR 0.67 (95%CI: 0.52 - 0.87) for 2015, OR of 0.98 (95%CI: 0.78 - 1.22) for 2016, an OR of 0.75 (95%CI: 0.60 - 0.93) for 2017, and OR of 0.32 (95%CI: 0.14 - 0.73) for the the start of the 2018 IRS round. Maps and classifications of seasonality patterns across Malawi were produced and showed distinct areas with an irregularly timed malaria season.

Discussion: Moving from monthly, clinic-level data to daily, individual-level data can provide more actionable results for the planning of malaria control. As an example, the reduced protective period of IRS within Illovo using the more granulated dataset, raises questions about the cost-effectiveness and timing of IRS in this setting. Better estimates of the protective period of IRS combined with maps of the timing of seasonality can aid the planning and roll-out of malaria control interventions in Malawi.

Item Type:	Thesis (Doctoral)
Subjects:	QX Parasitology > Insects. Other Parasites > QX 600 Insect control. Tick control WC Communicable Diseases > Tropical and Parasitic Diseases > WC 750 Malaria WC Communicable Diseases > Tropical and Parasitic Diseases > WC 765 Prevention and control
Faculty: Department:	Biological Sciences > Vector Biology Department
Depositing User:	Lynn Roberts-Maloney
Date Deposited:	23 Apr 2025 10:05
Last Modified:	23 Apr 2025 10:10
URI:	https://archive.lstmed.ac.uk/id/eprint/26597

Statistics

View details

Actions (login required)

Edit Item