Smoke, health and household energy

Volume 2: Researching pathways to scaling up sustainable and effective kitchen smoke alleviation

Please note that this is historical research conducted by Practical Action and partners in 2005. While the methodologies and conclusions may remain of value, it does not necessarily reflect the current situation. Please refer to our main pages on indoor air pollution for more recent research.

In May 2005, Practical Action published Smoke, health and household energy Volume 1, describing a participatory approach researched by the organisation into the design, installation, monitoring and assessment of smoke alleviation technologies or interventions.

The thirty households in each of three communities involved in the original research project were from: Kenya, Nepal and Sudan.

  • an urban community in Kisumu, Kenya
  • a displaced community close to Kassala town in Sudan
  • Gatlang, a high cold region in northern Nepal

However, achieving improved indoor air quality in thirty households in three countries does not even start to address the problem when millions are in need of new technologies. Indoor air pollution (IAP) is responsible for around 1.5 million deaths each year. This demands a more radical and long-lasting approach. Many millions of households need to get rid of the smoke from their homes if their lives and the health of their children is not to be seriously compromised. How can this be done?

download Smoke, health and household energy, Volume 2 as a PDF - 2.4Mb

Annexes to the publication

All annexes are in MS Word format.

See also Smoke, health and household energy Volume 1

Overview of activities

This second phase of research had, as its goal, Millennium Development Goal 4: Reduce Child Mortality, for which the study used proxy methods to determine the reductions in child mortality that could be expected for a given population from a known reduction in carbon monoxide (CO). The research focused around the impacts that could be achieved if ways could be found to enable a substantial percentage of the population to access interventions.
The health of the women cooks is very closely associated with the well-being of their young children, and the research also focused on women's health, using similar proxy methods as well as questionnaires and monitoring. As many initiatives on improved household energy have foundered in the past through households not electing to use interventions in the longer term, the work sought to identify 'desirable product attributes' that would encourage purchase and long-term adoption of these interventions

Thus, to make substantial inroads on ill-health, the study needed to ensure that indoor air pollution was reduced, and at the same time, research approaches to achieve this for substantial numbers of people - identifying and overcoming the barriers that prevent people being able to alleviate the kitchen smoke in their homes. Provided that the pollution reduction levels could be maintained or improved, then if large numbers of people were enabled to adopt interventions, and continue to use them, the ill health associated with indoor air pollution would be ameliorated.

To achieve the large numbers required, the research looked at whether a semi-commercial approach could be implemented, identifying and overcoming the barriers to people being able to alleviate the kitchen smoke effectively in their own homes. An additional advantage of this approach is that people tend to value more those changes for which they themselves have paid. Based on the interventions selected by households in each country in the first project, activities focused on:

  • Development of those interventions to improve their product attributes
  • Awareness-raising on the problems of indoor air pollution
  • Working with customers (beneficiaries) to identify desirable product attributes
  • Micro-credit for those with insufficient up-front capital to purchase goods and for start-up raw materials for entrepreneurs.
  • Training in manufacture, marketing and sales
  • Support to sales through health and well-being promotional messages
  • For entrepreneurs: Support with set-up; purchase of raw materials, wholesale goods, technology promotion, transport (in the early stages) and feedback from customers
  • At policy level, raising the profile of indoor air pollution internationally


The numbers of households adopting interventions was disaggregated by savings group and by month in each country, to establish whether smoke-alleviation would be viable beyond the project end.

At the same time, recognising that the transition from small-scale project to a larger commercial approach often leads to reductions in impact (caused by lack of adoption, quality issues etc.) levels of smoke were monitored, and specific key impacts were investigated using questionnaires.

Finally, based on the numbers of households adopting interventions, and on the levels of smoke alleviated, a cost/benefit analysis looked at the effect of reducing indoor air pollution in terms of improved health, time-saving and income / savings. .

Critical success factors

All countries

The most critical factor throughout this study has been the transfer of ownership of the problem of smoke alleviation from the project itself to the community, and the creation of a framework that continues to allow people either to access the interventions they desire, or to make short-term changes with the aspiration to make further improvements in the future.

Another factor which proved key in each country was the creation of micro-credit systems to help overcome the barrier of the up-front cost of the more costly interventions. Very often, it was found that people regarded items as very costly, when their durability, and other positive impacts, meant that it was financially advantageous for the household to purchase them, provided ways were found to afford them. The smoke interventions were designed to address other issues, so, for example, fuel savings could be set against repayments, time savings could be monetised by some by additional employment or time spent growing food.

For those using LPG, it is important to involve the suppliers of the LP gas. If it is demonstrated that those on low incomes can still afford to buy LPG, this is a very big market. Suppliers should be persuaded to supply gas bottles at a low price as their market is in fuel, not in bottles. There is evidence in other programmes that if the bottles are given away, they get sold a local market, but with awareness-raising on the benefits of clean fuel provided by the project, only one LPG set is known to have been sold in this way (by the husband of a cook who was ill and away from home).


The project team recognised that it was essential to start from the reality of life for most people, rather than from where the team might hope them to be. Those households adopting such interventions are reducing the levels of emissions by almost as much as those purchasing the more costly ones - albeit they are less convenient. This is a strong indicator that making people aware of the problem is very worthwhile even if they cannot afford the more costly interventions. People with very little money can reduce the pollutant levels in their homes.

Time-savings, convenience and modernity favoured bottled gas (LPG) among the more costly items - but just the ownership of an LPG stove did not necessarily mean that it was used all the time. Many households just used it for fast cooking - making tea in the morning etc whilst using polluting fuels to cook main meals. Fireless cookers (insulated boxes) were popular to reduce fuel costs.

Theatre and drama groups were very effective in Kenya, both in terms of awareness-raising, and because they raised the status of the community members.


By the end of the first project, the smoke hood that had been designed, through collaboration between community and project staff, alleviated the smoke by an insufficient margin for the team to wish to disseminate it further. However, the community liked the hood and, because they were fully engaged in the process, they were very willing to try out improved designs. The result was a hood that removed over 85% of the CO (compared to ~40% in the first project), and had more desirable attributes than the original design.

A sense of ownership of the revolving seed capital lead high levels of repayment of revolving finance - people discussed what would happen to 'their' seed capital if they did not make the repayments - so it was no longer 'NGO money' and therefore it had to be repaid. From low levels of repayment in the early stages, repayments are now very high. Default is usually only associated with those leaving the area - and if they return, it is expected that they will restart their repayments.

Engaging local government and other NGOs is vital if some form of direct subsidy is needed. The community have paid far more for the smoke hoods than they had originally indicated was feasible in the marketing survey, and no subsidy at all would have led to complete failure of the initiative, rather than the 500 hoods already installed or under installation. Smoke hoods have made a very major impact on levels of smoke. By engaging local government, the project has worked to get longer-term subsidies for smoke hoods. Lower-cost hoods and less transport costs in the lower regions are already designed and planned. These are unlikely to need subsidy, but the design is inappropriate for high cold regions.

Space heating takes up a lot of fuel, and leads to emissions of smoke over the longer periods when the fire is lit. Demonstrating how to insulate the dry-stone walls to reduce heat loss and to prevent cold air from blowing through the cracks has saved fuel and reduced emissions. The all-metal design of the hood allows the heat to radiate from the hood. Although this is more expensive, it does prevent households from lighting open fires in other parts of their house. This is important insofar as smoke going up through the chimney takes heat with it - and saving the heat through other means counterbalances the effect.

Smoke is a major issue in a country with a lot of high cold regions requiring space heating, and through the project, the Indoor Air Pollution and Health Forum was set up to provide an interface between those living in poverty and affected by indoor air pollution, and those with policy influence to effect change. Though Practical Action has handed over responsibility for smoke alleviation in this region to local government, local groups and entrepreneurs, it will continue to play an active role in smoke alleviation at a national level, incorporating other interventions that are more appropriate for other communities.


Sudan produced very mixed findings, with revolving finance leading to nearly 1500 households with LPG stoves, and a continuing buoyant demand for finance to buy LPG appliances at the project close. This is largely because LPG is cheaper and promoted by government as it reduces deforestation. Women aspire to use LPG and like using it.

Over several years, Practical Action has promoted Women Development Organisations (WDAs), to empower displaced women through engaging in productive activities. These WDAs were chosen to run the business-side of the scaling up, including microfinance (as this was already part of their role). This allowed a rapid response during the early part of the project, but also caused problems due to lack of experience in dealing with the high demand.

Issues and problems

The change process in Kenya may take a number of years for any given household, and this sort of study did not have the time to evaluate the impact of the work on many of the households who have started on the process of alleviating smoke, but had future plans for further improvements.

Repayments, though made, often take much longer than had been agreed - so for some households, the actual monitoring process only took place towards the very end of the project and not all the households could be monitored. There is an acceptance between those running the revolving finance, and those using it, that payments may not be made on time - it could take up to fifteen months. As a result, the revolving finance will not work as a business opportunity for an entrepreneur wanting to use smoke-intervention sales as their main source of income. However, the revolving finance is running sustainably at a lower level which is acceptable to both supplier and customer and this suggests that a fully commercial approach, where several women use the small profits as part of their income, is a good model within such low-income societies. The 'spin-offs' of setting up revolving finance are all positive; the groups now run their own bank accounts and are planning to use the finance more flexibly in future for income-generating activities as well as smoke-alleviating options.

The other major issue in Kenya is that there are few interventions of intermediate cost to reduce very substantial amounts of smoke. Work is ongoing on the design of a rocket-type stove suitable for household use.

Smoke hoods have been adopted less than had been expected. They are seen as less modern than LPG, yet those who use them do so all the time, not just for quick meals, so they get rid of most of the smoke all the time. Smoke hood adoption has been disappointing.

Running revolving finance through local committees has lead to very good levels of repayment. A less positive effect is that it is difficult to move the capital to support other villages as it is either 'theirs' to make other improvements, or 'NGO money' so does not need to be repaid….It is therefore important to try to get the balance right between the time that people have to wait for their turn to use this funding, and the need to make sure that it benefits as many households as possible.

Subsidy is an issue in Nepal - but without it none of the households living in poverty would have purchased smoke hoods. It is important not to dismiss subsidy out of hand. Where subsidies are needed, routes should be planned for terminating the subsidy, whilst supporting the ongoing subsidy whilst it is needed. A local entrepreneur now runs a business with four staff members, and he is aware of the need to cut back on subsidy or his business will run out of customers. He has worked closely with the team on the design of the smoke hood, and also on desirable attributes to make the item more attractive to customers.

In the third round of monitoring, it was found that many households had reverted to charcoal. An additional survey found that this was mainly due to the difficulty in getting the bottles refilled because:

  • the depot was at some distance from the households
  • women would only start to save for the refills once the bottle ran out

This problem is likely to be resolved in the near future, as sales of LPG appliances within these low-income areas is still growing fast, and the gas supplier has promised to provide local depots and a mobile shop as the profits have been very worthwhile.

A second reason for the reversion to charcoal is that many displaced families are returning home at this present time. Charcoal prices are currently fairly low in Kassala - the convenience of buying charcoal by the day is thus a big factor in its favour.

The rapid growth in demand for LPG appliances meant that the WDAs had to take on much greater responsibilities more quickly than had been expected. Despite a lot of training from Practical Action, it was shown that many of those trained were not commercially focused, having engaged in WDA activities to help 'beneficiaries' through very small businesses, rather than to run relatively large operations. A new set up is planned, with recommendations that those running the WDAs are not necessarily those who run the micro-finance, will ensure that those running the microfinance are both well-qualified and well-trained. In the meantime, support is being continued through Practical Action until the team is more confident to run it autonomously.


Smoke alleviation

This table looks at the levels of carbon monoxide, used as an indicator of pollution levels, before and after smoke interventions had been installed. For Sudan, because carbon monoxide is only used as a proxy, the maximum predicted levels for particulates are given, as a switch from wood and charcoal mixtures to almost exclusively charcoal reduced the levels of particulates (PM) - which are believed to be the main cause of the health problems. In both Kenya and Nepal, either the same fuel or LPG gas (with virtually no particulates) was used - so the CO data can be used directly to indicate percentage reductions.








% change



% change



% change

Measured mean CO (ppm)










Predicted mean PM
(μg/m3 )










Adoption and continuing use of interventions

Adoption rates for the more costly interventions are relatively slow. During the project period, around 350 households adopted interventions for which revolving finance was needed - this is around 2.5% of the total households. The natural adoption of other no-cost or low-cost technologies, such as fireless cookers, eaves spaces, and behavioural changes could not be monitored, but is believed to be high. The interventions purchased were still in use in all cases among those responding to the questionnaire in Kenya. Virtually all LPG users among those questioned used it exclusively for making tea -the only cooking done at breakfast time, although only ~25% used LPG exclusively for cooking their main meals.

During the project period, around 450 smoke hoods were installed, or were just about to be installed. Continuing growth will depend largely on the priority given to alleviating smoke given by the District Development Committee, who will run the revolving fund and have committed to continue to support the initiative and to provide subsidy in the immediate future. As there are already larger subsidies on lighting and stoves, this is a positive development. There is currently a full order book for the entrepreneur and five assistants who are manufacturing the hoods in the area. All the smoke hoods installed are in constant use.

Adoption levels in Sudan continue to be high, with a steady and ongoing installation rate of around 50 stoves per month. Problems associated with coordinating the revolving finance across all the WDAs has been addressed, and a new system and regulations, particularly the legal side regarding guarantees for repayment, has been put in place. Practical Action will remain in an advisory role during this period. In the third round, only one fifth of those monitored were using LPG as their fuel, although 50 of the 148 respondents said they used it either exclusively (31) or in conjunction with charcoal or wood (19) for cooking main meals. Setting up savings groups for fuel savings has already been started to address this issue, and local depots are promised to improve access.

Community satisfaction

Satisfaction ratings in all countries are close to 100% on all counts. Even in Sudan, where many households had reverted to charcoal, there was almost universal support for the use of LPG stoves. In Kenya, the major benefits were reported as time, fuel reduction, smoke reduction; in Nepal, the key benefits included smoke alleviation, fuel savings and health improvements: in Sudan, cooks cited time-savings, increased comfort and smoke reduction. Health was reported as improved in all three countries - though again, this was less prominent for some questions in Sudan. In Kenya and Nepal, all households were using the interventions that they had bought or built. Households were also very willing to discuss problems, which reflected the good relationship set up between the researchers and the household members.

Transferability of findings

The participatory approach has been tested in three very diverse situations, and has proved highly effective in each country. Findings from Nepal, where the process initially created a less-than-ideal smoke hood, but led to an excellent technology for which people were willing to pay far more than they had at first indicated, suggest that it is necessary to keep faith with the process - it means that people make their own decisions so adoption is not a problem.

Awareness-raising can change people's priorities and create markets provided that the goods and appliances are available. Working with community groups to dissemination messages through theatre groups, drama, design of outputs can be a powerful tool to reach very large numbers of people.

Poverty is not a static condition, but it is where people sometimes find themselves, so working with those living in poverty to identify ways in which people can help themselves immediately and build up gradually will start an ongoing process of change.

Micro-credit, run by the communities themselves, has been highly effective, and has been shown to be useful even where it is not used for income generation. This is an important finding and runs contrary to conventional wisdom. This approach could be used for other household initiatives, such as water, sanitation, shelter provision etc. Discussions around its implementation, duration of loan, maximum loan etc. have given rise to good repayments and a sense of ownership of the system. Ideally, it should be used for a range of interventions across all services, as this would give those in community a greater say in its use.

Making savings in costs and time can be very valuable and should be part of the 'mix' wherever possible.

A holistic approach to the impacts is important if goods and services are to be promoted widely. Provided that interventions do not lose sight of their prime objective, other 'desirable product attributes' can encourage people to adopt such interventions.

The impact of making changes in the kitchen goes far beyond just alleviating smoke. Particularly in societies where the woman tends to have a lower status, many of the impacts reported are social, rather than environmental, benefits; women are aware of their improved health and that of their children; their homes are cleaner, their drudgery is reduced. This research indicates that improving the quality of life in the home can lead to increased wealth, better health, more time and improved status.

Overall conclusions and findings


The approach taken has been to use the change (reduction) in IAP levels as the main indicator of expected impact on child health, by relating measured reductions in pollution to the best available and emerging evidence on the association between exposure levels and incidence of pneumonia. This indicated that if interventions were introduced into 25% of households such as those in the study, the number of cases of pneumonia could be reduced annually by between 150-300 in Kenya and Nepal, and around 100 in Sudan in populations ranging from around 63000 in Kenya, 45000 in Nepal, and 79000 in Sudan. In Sudan this impact is indicated despite the numbers of households using charcoal the time of the study.

As well as pneumonia, substantial reductions in other health problems were reported in all three countries, particularly in Kenya and Nepal. In Kenya, there was strong evidence of a reduction in the prevalence of the general cough and wheeze symptoms, with many fewer indicating chronic problems.

There was a marked and statistically significant reduction in headaches in terms of frequency and strength of headaches. Although numbers are low, there is strong evidence that where interventions were in use, the risk of burns and scalds to children has been reduced. In all three countries there was a reduction noted in the number of visits to health providers and the cost of health provision. In Kenya and Nepal this was particularly marked.

Scaling up

In all three countries, scaling up through addressing both supply and demand sides has proved successful. Locally run revolving funds continue to provide ways for people to purchase smoke interventions. Entrepreneurs have been trained, and continue to work on interventions. All these operations are operating commercially, although in Nepal, subsidy is still being provided by the District Development Committee.


The impacts of the work have been overwhelmingly positive. This is the case when specific questions were asked regarding health, time, income/savings, environment and prestige/quality of life, and also when people were asked to identify the impacts that the interventions had made.

The team are not aware of any major negative impacts, but problems (such as rain ingress in Nepal) have mainly been sorted, or are currently being addressed. Issues around savings in Sudan are being addressed.

Cost benefit analysis

Overall, these results show positive benefit to cost for all three countries, and a very high ratio for Kenya. Time saving is by far the most important component, but fuel cost savings are also substantial for Kenya and Sudan. The direct health benefits appear as a very small component, for reasons which are discussed in Chapter 11.

As health impacts were a small component, variations in estimates of exposure reduction, disease incidence rates, health costs and time lost due to illness would make little difference to the overall results. Even a substantial reduction in estimated time and fuel savings costs (up to 50 per cent) would still yield positive cost-benefit ratios for Kenya and Sudan. For Nepal, fuel cost savings were not relevant, and it was fuel collection time that was more critical.

This initial, household perspective CBA has shown positive benefit to cost ratios for households purchasing interventions that are achieving at least useful and often very substantial reductions in IAP. This should encourage efforts to assist prospective adopters with the financial arrangements needed for them to obtain these benefits.

Skills transfer

All three teams have developed very considerable expertise in monitoring and data processing. One of the Nepal team has provided expertise to the Cost Benefit Analysis (CBA), working with the University of Liverpool to provide this important policy output. Work is continuing on a Societal CBA as the project comes to an end.

Knowledge sharing

  • Efforts were made to ensure that all the key team members attended conferences throughout the research so that other organisations could benefit from field knowledge as well as findings.
  • The work has been reported through several papers focused on health, energy and development. It has fed into websites including those of WHO, HEDON and the Partnership for Clean Indoor Air.
  • Two clean air networks have been set up - one in Nepal for the whole of South Asia, and another in West Kenya to help continue to disseminate the need for clean air beyond the end of the current work.
  • Radio and TV programmes have been broadcast in all three countries and several programmes have been broadcast on the World Service.
no comments