Impact of odour-baited mosquito traps for malaria control: design and evaluation of a trial using solar-powered mosquito trapping systems in western Kenya

T. Homan

Research output: Thesisinternal PhD, WU

Abstract

The parasites belonging to the genus Plasmodium are the cause of the second deadliest infectious disease in the world, malaria. Sub Saharan Africa harbours more than 90% of malaria attributable mortality and morbidity, and most deaths occur in children under 18 years old. Malaria is transmitted to humans by a bite of a Plasmodium infected arthropod vector from the genus Anopheles. Halfway the 20th century malaria was successfully eliminated from most developed countries, nonetheless in the third world effective control remains a laborious challenge. Intensive efforts undertaken to control and eventually eradicate malaria during the past decade have led to substantial reductions in morbidity and mortality. Conversely, scientists became increasingly aware that with the current preventative and curative tools against malaria successful eradication seems unlikely. Not only do current tools not suffice to attain that goal, their efficacy to control malaria as it is, maybe severely threatened. Proper treatment and diagnosis are becoming increasingly less effective because of the adaptive nature of the parasite. Parasites get resistance against drugs and carriers are more often found to have subclinical infections. Likewise prevention of malaria, by vector control, becomes less effective. Malaria vectors become resistant to insecticides and transmission patterns are shifting away from where preventive measures are functional: outside and during the day. It this gap where the SolarMal project experimented with a novel malaria vector control tool, complimentary to existing malaria control methods: odour-baited mosquito traps that mimic human beings to lure and kill mosquitoes to eventually reduce malaria. The ultimate aim of this thesis was to seek proof of principle of the effect of mass trapping of malaria vectors on malaria and mosquito densities by rolling out over 4000 odour-baited mosquito traps at household level on Rusinga Island, Kenya.

Chapter 2 is a study protocol of the SolarMal project and provides a general understanding of how the objectives of the project are translated into a research design. The study comprises of  a medical, an entomological and a sociological discipline. A multidisciplinary strategy is presented in which the intervention is explained. Experimental designs of all disciplines are introduced including time frames, participant eligibility, and randomisation. Furthermore, a general overview of the data collected and how it is evaluated and analysed using health and demographic surveillance and monitoring is provided.

In chapter 3 a novel data collection and management platform is presented. The health and demographic surveillance as well as other disciplines in the project are an example of one of the first fully digital data collection systems in a low and middle income country. The development of digital questionnaires and the conducting of these by means of Open Data Kit software enabled the project to efficiently collect data. All residential structures were documented by GPS, and data of individuals attached. Converting the geo-located data to a geodatabase and displayed with Google Earth mobile made navigating from house to house an easy task. By daily uploading of data to the server at the project campus, scientists have access to a near real time database. Once uploaded to the server, data is transferred to the OpenHDS database in which the demography of the study population is updated accordingly. Data quality was further increased by a tool that looked for inconsistencies.

In chapter 4 we explore what experimental design would fit the SolarMal project best. A stepped wedge cluster-randomized trial [SWCRT] design was chosen to make sure that the whole area would cross over from the control to the intervention arm over a period of two years. As elimination was the goal, universal coverage was required. Subsequently, strategies for randomization and crossover of clusters that could measure a possible intervention effect best were simulated with a generic model of disease transmission. Considering sufficient numbers and sizes of clusters a hierarchical SWCRT would best measure a possible effect of OBTs on Rusinga Island. Special care was given to quantifying spill over effects into the control arm. Finally, two new measures of intervention effectiveness are proposed.

Chapter 5 reports on the outcomes of the health and demographic surveillance system on Rusinga Island. Running an HDSS is a thorough but complex method to monitor intervention effects in an area where health surveillance is minimal. As part of the overarching HDSS institution, INDEPTH, data collection methods and reporting are harmonious with many other HDSSs around the world. Demographic parameters are calculated and the HDSS practices are described.

Chapter 6 uses the baseline cross sectional prevalence surveys to elucidate how the epidemiology of malaria on Rusinga Island. Firstly, the malaria distribution and hot spots are identified. Consequently, a standard epidemiological model and a geographically weighted regression are compared, and used to identify risk factors for malaria. The latter model, taking into account non-stationarity, performs better and is able to produce geographically varying risk factors. The strength of the relationship of risk factors for malaria are heterogeneous over the whole island, and for instance social economic status and occupation are strong predictors of malaria in some areas but less in other areas. Considering these risk factor distributions can aid in guiding the implementation of malaria intervention methods.

Chapter 7 presents the main outcomes of the SolarMal project. The impact of OBTs on the prevalence of malaria is pronounced in the contemporaneous comparison between the intervened and the intervened arm. Comparison of baseline data with the intervened clusters does not yield significant effects. A strong decline in cases of clinical malaria was observed starting already in the baseline period, and therefore we cannot attribute this decline to the intervention. Effects on the most prominent malaria vector were large, whereas other vectors did not suffer under the intervention.

Chapter 8 is a general discussion of the work provided. The most important implications of the thesis are discussed underscoring the societal and scientific relevance, and putting the research in a wider perspective. Unaddressed issues are raised and recommendations for further research are provided.

Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
  • Wageningen University
Supervisors/Advisors
  • Takken, Willem, Promotor
  • Smith, T.A., Promotor, External person
Award date18 Apr 2016
Place of PublicationWageningen
Publisher
Print ISBNs9789462577435
DOIs
Publication statusPublished - 2016

Keywords

  • 016-3935
  • culicidae
  • mosquito-borne diseases
  • vector control
  • malaria
  • bait traps
  • odours
  • solar energy
  • randomized controlled trials
  • kenya

Fingerprint

Dive into the research topics of 'Impact of odour-baited mosquito traps for malaria control: design and evaluation of a trial using solar-powered mosquito trapping systems in western Kenya'. Together they form a unique fingerprint.

Cite this