Malawi has one of the highest rates of death among infants and the under fives (69 and 110 per 1000 live births respectively in 2009) despite having made progress towards meeting the Millennium Development Goal of reducing child mortality. Pneumonia is the leading cause of death and one of the commonest causes of morbidity with around 298 per 1000 children under the age of 5 diagnosed with pneumonia every year and a case fatality rate between 2.7 and 13.2 per 1000. Exposure to smoke produced when biomass fuels (animal or plant material) are burned in open fires is a major avoidable risk factor for pneumonia. In Malawi, where at least 95% of households depend on biomass as their main source of fuel, biomass smoke exposure is likely to be responsible for a substantial burden of this disease.

Effective strategies for reducing smoke exposure exist (e.g. ventilation, improved stoves, cleaner fuels, behaviour modification) but are out of reach for the majority due to a wide range of largely poverty-related factors. The Global Alliance for Clean Cookstoves was launched in 2010 to tackle this energy poverty issue through public private partnerships. A central aim of the alliance is for 100 million homes to adopt clean and efficient stoves and fuels by 2020. However, there is very limited evidence to assess the potential benefits of such an approach.

We have conducted preparatory and pilot work in sub Saharan Africa to determine which of the currently available advanced cookstoves would be most suitable for use in a trial in Malawi in terms of improvements in combustion efficiency, reduced emissions and ability to cook local dishes. We have gone to considerable lengths to involve local communities in the development of this proposal.

We are now in a position to be able to study an efficient and locally acceptable advanced cookstove that substantially reduces smoke emissions in a trial to address three principal research questions:

1) Can an advanced cookstove intervention that substantially reduces biomass smoke exposure relative to an open fire prevent pneumonia in children under 5 years old in Malawi?
2) How affordable and cost effective is the intervention from household, healthcare system and societal perspectives?
3) What can be learned from trial participants and non-participants about adoption of the intervention that could inform effective implementation of the trial findings in the future?

High quality clinical trial evidence about the health and economic impacts seen when households adopt advanced cookstove technologies is needed to inform policy and decision makers across commercial, health, development and community sectors at local, regional and international levels. The results of this trial will be relevant to local policy makers in Malawi who will have new efficacy, economic and qualitative data to guide decisions about funding advanced cookstove programmes for improving child health; to regional commercial, non-governmental and governmental organisations in sub Saharan Africa manufacturing and distributing advanced cookstove solutions; and to international (e.g. World Health Organisation (WHO)) decision and policy makers by contributing new evidence about the health and economic impacts of an advanced cookstove intervention of broadly generalisable relevance to areas of the world where biomass fuel use is common. We have established local (e.g. community leaders), regional (e.g. commercial and non-governmental organisations and Malawi Ministry of Health) and international (e.g. WHO and Global Alliance for Clean Cookstoves) links that will allow us to disseminate the findings of the trial effectively at all levels to a wide range of stakeholders, policy and decision makers.

Technical Abstract:
Trial type: Prevention.
Design: Village-level cluster randomised controlled trial with two arms of equal size.
Interventions: Experimental - The Philips fan-assisted stove with user training (replacing open fires). Control - Continuation of traditional cooking methods (open fire).
Randomisation: By matched pairs of villages using computer-based system stratified by district to trial arms in a 1:1 ratio.
Target population: Children up to 4.5 years old in Malawi allowing for a minimum of 6 months data collection before a child's 5th birthday. The trial will be broadly inclusive. HIV infection will not be an exclusion criterion.
Duration of treatment period and follow-up: 24 months from randomisation.
Primary outcome measure: Incidence of pneumonia in children under 5 years of age diagnosed by blinded physicians or medical officers using the WHO Integrated Management of Childhood Illness (IMCI) pneumonia assessment protocol.
Economic, social, qualitative measures: Analyses will be carried out to explore factors that may be relevant for scale up of the intervention in the future.
Sample size and potential power of the trial: Randomisation will be at the village level with an expected average number of children per village of 85 (77 after allowing for 10% loss to follow up). The between cluster CV has been set at the conservatively high level of 0.10. The latest data from Malawi suggest that around 9% of control group children will develop pneumonia severe enough to require treatment at a healthcare facility every year. We are using a highly conservative estimate of 5% (allowing for impact of pneumococcal vaccine). A sample of 59 villages per group will give 80% power to detect a 20% reduction in risk in the intervention group from 5% to 4% (a=0.05).
Consumer involvement: Villagers, local stove producers and community leaders in Blantyre and Balaka districts have been involved with the development of this trial proposal through community engagement meetings.

Potential Impact:
Research participants: The research is expected to provide direct benefits to trial participants: a quarterly allocation of maize or pigeon peas, access to a mobile phone, access to antibiotics for pneumonia if indicated but unavailable at the local health facilities and two advanced cookstoves either at the beginning or end of the trial. Reduced fuel consumption is a potential additional ongoing benefit. Participation in the trial is at village-level and may therefore have further positive effects on welfare and social cohesion.
Local staff: We plan to employ a number of local people through this grant with many other local people involved indirectly. The training and experience gained should lead to improved work opportunities in the future. In addition there will be immediate financial benefits for individuals receiving a salary. Such financial benefits typically extend widely beyond these individuals to families and home villages.
Women and children living in poverty in developing countries: Should we see a reduction in childhood pneumonia from an effective biomass smoke exposure reduction intervention in this trial, this finding will be relevant across populations of people who cook using open fires living in the developing world well beyond the immediate trial setting. The benefits of adopting clean cookstove technologies could extend beyond health effects to economic, quality of life and well-being effects.
Local, national and international policy and decision makers: High quality clinical trial evidence about the health and economic impacts seen when households adopt advanced cookstove technologies is needed to inform policy and decision makers across commercial, health, development and community sectors at local, national and international levels. The results of this trial will be relevant to local policy makers in Malawi who will have new efficacy, economic and qualitative data to guide decisions about funding advanced cookstove programmes for improving child health; to regional commercial, non-governmental (NGO) and governmental organisations in sub Saharan Africa manufacturing and distributing cookstove solutions with uncertain health benefits; and to international (e.g. DFID, WHO and Global Alliance for Clean Cookstoves) decision and policy makers by contributing new evidence about the health and economic impacts of an advanced cookstove intervention of broadly generalisable relevance to areas of the world where biomass fuel use is common.
Public Sector: We will provide regular training to local healthcare workers in the use of the IMCI pneumonia assessment protocol. This training will help maintain knowledge and skills and will hopefully have positive impacts on the management of the ill child.
Business: The new scientific knowledge including exposure-response data from this research could be exploited by businesses locally and internationally developing biomass smoke exposure reduction interventions. This could contribute to wealth creation and economic prosperity in individual developing countries.
Third Sector: Concern Universal and other NGOs implementing cookstove programmes are expected to benefit from new knowledge about exposure-response effects of biomass smoke exposure reduction strategies that could be used to guide future activities.
Other: LSTM, MLW and COM have public engagement programmes that contribute to increasing public awareness and understanding of science, economic and societal issues which we will contribute to with the results of this trial.
Environment and climate change: The widespread adoption of clean burning efficient cookstoves that consume less fuel and put less pressure on forests could contribute to environmental sustainability and protection. A cookstove intervention that has health as well as environmental benefits could be particularly valuable in the carbon trading market that could help provide funding for sustainable cookstove programmes.