The role of improved housing in the African malaria recession

Malaria is a mosquito-borne disease, mainly confined to the tropics. Today, malaria control is at a pivotal juncture. The disease remains a major source of human illness and death, killing around 600,000 people annually. Yet the past fifteen years have seen great advances in controlling malaria, with a third reduction in the annual incidence of cases. These reductions have been mainly achieved by distributing chemically-treated bed nets for people to sleep beneath and by spraying houses with chemicals that kill mosquitoes. However, mosquitoes have become resistant to the chemicals used, meaning that additional tools are urgently needed to avoid losing the progress made.

Since malaria is a 'disease of poverty', strongly influenced by its environment, there is heightened interest in using broader, non-chemical approaches for its control, by working with sectors outside health such as agriculture, water and sanitation and urban planning. This is especially pertinent as we approach the launch of the Sustainable Development Goals for international development later in 2015. The research proposed here will specifically investigate how one such approach - improving housing - can help to reduce malaria in sub-Saharan Africa.

Good housing, traditionally a key pillar of public health, remains underexploited in malaria control. Yet most bites by malaria mosquitoes in sub-Saharan Africa are received indoors at night, so the home can be a risky place. Historically, improved living conditions and house screening contributed to malaria elimination in Europe and the USA. Today, studies have shown that well-built housing can help to protect against malaria in many African settings. However, house design has been largely neglected as a potential malaria control tool, due to reliance on other interventions.

Today, unprecedented population expansion and socioeconomic development in sub-Saharan Africa presents an unrivalled opportunity to build healthy homes. The continent's population is projected to double between 2010 and 2040 to nearly two billion and may surpass three billion by 2070, with 144 million new houses needed by 2030 in rural areas alone. Housing quality is also transforming across much of the continent, as incomes increase. This economic and cultural revolution represents an exceptional opportunity for malaria control.

The aim of this fellowship is to investigate the potential contribution of better housing to malaria control in Africa. This will be achieved by addressing three questions:
i) How is malaria associated with housing quality in different parts of sub-Saharan Africa?
ii) To what extent has housing improved in Africa, and how does this change relate to trends in malaria endemicity, 2000-2015?
iii) What is the potential impact of using better housing as an intervention against malaria?

To answer these questions, state-of-the-art modelling approaches will be used (1) to assess the association between house design and malaria in multiple African countries, (2) to quantify how housing improved in Africa between 2000 and 2015, (3) to understand how trends in malaria between 2000 and 2015 were related to changes in housing and (4) to estimate the total number of malaria cases in Africa that may be averted by ongoing housing improvements. This study will exploit the world's largest collection of malaria surveys, the Malaria Atlas Project, together with nearly 200 demographic and health surveys from over 40 countries in sub-Saharan Africa.

In summary, this study will evaluate the potential of building better homes to impact on malaria in Africa. The outcomes of this research will provide critical information to guide research and policy decisions relating to sustainable malaria control, at a pivotal time.

BACKGROUND: Since most malaria transmission occurs indoors in sub-Saharan Africa (SSA), housing quality is an important determinant of malaria risk, through its effect on house entry by mosquitoes. Unprecedented population expansion and rapid socioeconomic development in SSA presents an unrivalled opportunity to advance malaria control and elimination by building healthy homes.

AIM: To investigate the role of improved housing in the African malaria recession.

METHODS: Two principal data sources will be used: (i) the Demographic and Health Survey (DHS), Malaria Indicator Survey (MIS) and Multiple Indicator Cluster Survey (MICS) programs and (ii) the Repository of Open Access Data of the Malaria Atlas Project (ROAD-MAP II). First, the association between house quality and prevalence of malaria in children will be examined (a) in Upper River Region, The Gambia, using a 2010 parasitological survey of c3800 children and remotely-sensed imagery to classify house type and (b) in 22 countries in SSA, by applying multilevel modelling to MIS and DHS surveys. Second, changes in housing quality at the district level across SSA, 2000-2015, will be estimated by fitting housing data from georeferenced DHS/MIS/MICS surveys within a multilevel model using a Bayesian framework. Third, the relationship between housing improvements and changes in malaria endemicity in SSA, 2000-2015, will be modelled at three spatial scales using data from Objective 2 and ROAD-MAP-II. Fourth, the total number of malaria cases averted in SSA in 2015 due to housing improvements during 2000-2015 will be estimated using a spatiotemporal model and counterfactual framework.

SUMMARY: The application of geospatial modelling to large existing datasets will generate novel evidence on the relationship between housing and malaria and the impact of housing improvements on malaria endemicity, yielding critical and timely policy recommendations for sustainable malaria control.

The proposed research will examine the potential for better housing to control malaria, helping to address the urgent need for vector control tools beyond long-lasting insecticidal nets (LLINs) and indoor residual spraying (IRS). Immediate beneficiaries will be academics, policy-makers, malaria control programmes and non-governmental organisations. The ultimate beneficiaries will be the two billion people living at risk of malaria worldwide.

Academics within public health will benefit through the generation of new data on changes in housing quality in Africa, the assembly of new evidence on housing and malaria and the application of tailored statistical approaches to large survey datasets and Malaria Atlas Project data (see Academic Beneficiaries). The research will be of particular benefit to epidemiologists conducting field studies of housing as a malaria intervention, by generating macro-level data to (a) gauge the potential impact of housing in different settings and (b) guide the design of future field studies. Other beneficiaries will include public health entomologists seeking supplementary tools for malaria vector control; malaria elimination specialists seeking long-term, sustainable interventions; malariologists interested in understanding of the causes of the African malaria recession; and spatial epidemiologists and health geographers studying the effects of the environment on health. The use of large demographic and health survey datasets will benefit methodologists who work with similar data to study risk factors for health outcomes. Exploitation of the Malaria Atlas Project will make constructive use of this valuable resource. The research will exploit the UK Medical Research Council's strategic investments in health research in The Gambia (see letter). Academic beneficiaries outside the health sphere will include remote-sensing experts and urban geographers, who will benefit from the application of remotely-sensed data and the analysis of housing variables within large survey datasets; architects working on housing and health; and cross-disciplinary research teams from social science, public health and development economics, specialising the relationship between development and health.

Policy-makers and national malaria control programmes (NMCPs) will benefit in three ways. First, given that LLINs and IRS are increasingly threatened by insecticide resistance, there is an urgent need to evaluate supplementary methods of vector control, such as housing. Second, the proposed research will generate new evidence on multisectoral malaria control, with which to guide programme design and to support advocacy. Third, if housing shows potential as a sustainable intervention, the long-term costs of malaria control and elimination may be reduced. Tangible, short-term benefits for policy-makers and NMCPs will be: (a) estimations of the potential impact of housing improvements, (b) identification of the malaria transmission settings where housing is likely to be more and less effective and (c) generation of data and maps for advocating for multisectoral (housing-health) malaria control. In turn, the development and housing sectors will benefit from important new data documenting changes in housing quality at the district level in Africa since 2000 and from data to promote advocacy for better housing (the 2013 UNDP Multisectoral Action framework for Malaria states that 'encouraging better housing is a social objective, not simply a malaria control action'). Should housing improvements be more systematically implemented as part of the broader development and health agenda, the ultimate beneficiaries will be the populations of malaria-endemic countries, whose quality of life will improve through better health and access to decent housing.