Global burden of disease from household air pollution: how and why are the estimates changing?
The evidence is robust and compelling: exposure to household air pollution (HAP) is responsible for a staggering number of preventable illnesses and deaths each year. According to the most recent estimates published in The Lancet in 2017, approximately 2.6 million deaths can be attributed to HAP each year. This figure has evolved significantly over the past decade. The first global burden of disease estimates in the early 2000s suggested that approximately 2 million premature deaths occur each year as a result of exposure to indoor air pollution (IAP). Since then, updated estimates published between 2012 and 2016 attributed between 2.9 to 4.3 million deaths to direct exposure to household air pollution. Starting in 2016, the Institute for Health Metrics and Evaluation (IHME) has committed to updating estimates on an annual basis. This briefing note provides a brief description of the various factors that drive changes in estimates over time.
For the most up-to-date country specific information on burden of disease and trends, visit www.healthdata.org
Has there been a drastic decrease in the number of people exposed to household air pollution?
No. Around 3 billion people continue to cook with traditional stoves and fuels. Taking population trends into account, this translates into around half of the world’s population in 2000 and approximately 40% of the world population today. While the relative proportion of people exposed to this risk has decreased, the absolute number of people at risk has not changed.
Have there been major changes in the underlying burden of disease that affect risk estimates?
Yes. Changes in specific burdens of disease can have a positive or negative effect on risk factor-specific burdens. The implementation of new prevention, diagnostic, and treatment strategies can have major impacts on cause-specific burdens of disease, which can then impact risk factor-specific burdens. For example, with major declines in childhood pneumonia observed over time, there is a smaller pool of pneumonia that can be prevented by scaling clean cooking. On the other hand, with increasing prevalence of noncommunicable diseases, including chronic lung and cardiovascular disease, the public health relevance of scaling clean cooking to promote the health of elderly populations will only continue to increase.
How has the risk assessment methodology changed over time?
Changes in the risk assessment methodology can result in increases or decreases in burden of disease estimates. Each time IHME updates their methods and/or estimates, they also use the same methodology to recalculate previous years’ data to allow for a direct comparison of changes in burden over time.
- From the original ‘framing’ of the risk factor as ‘indoor smoke’ in initial estimates, the risk factor has been broadened to ‘household air pollution’. This is an important distinction, because not everyone cooks outdoors and cooking with solid fuels can be a major source of ambient (outdoor) air pollution.
- Over time, new health research evidence is incorporated into the models used and the overall burden of disease assessments.
- The earliest estimates were based on binary classifications of exposure (e.g., ‘exposed’ or ‘unexposed’ to indoor smoke). Since then, integrate exposure-response (IER) methodology and the availability of new exposure-response data have enabled assessments to evaluate the impact of achieving quantitative measurements of exposure, which is achievable through cooking exclusively with gas or electricity (as opposed to cooking with traditional stoves and fuels).
- Estimates incorporate evidence from IER methodology, which ensures consistency in estimating risk across major sources of exposure to combustion-source pollution. These sources include ambient air pollution, secondhand smoke, household air pollution, and active smoking. The IER methodology also allows for the indirect quantification of cardiovascular effects of household air pollution. The shapes of the IER curves also change over time as new evidence is incorporated.
- The health evidence knowledge base has grown more robust and now includes more comprehensive systematic reviews and evidence from a randomized control trial for acute lower respiratory infections, a major cause of child illness and death. As this evidence base grows, the number of health impacts assessed also increases. For example, more recent estimates now include cataracts and lung cancer from biomass.
- In 2010 and 2012, complementary estimates focused on quantifying the additional ambient air pollution-related burden of disease caused by household air pollution. This type of pollution accounts for approximately 15% of global emissions (roughly 4 μg/m3 of PM2.5), and the proportion is much larger in places with high household solid fuel use (approximately 25-30% in India, for example). More recent estimates have not attempted this at a global scale due to major gaps and uncertainties in country- or region-specific data on major sources of air pollution.
For more information, please contact email@example.com.
Referenced Data with Links:
IHME website: http://www.healthdata.org/
Institute for Health Metrics and Evaluation (IHME). GBD Compare Data Visualization. Seattle, WA: IHME, University of Washington, 2017. Available from http://vizhub.healthdata.org/gbd-compare.
World Health Organization. 2016. Ambient Air Pollution: A Global Assessment of Exposure and Burden of Disease. Geneva:World Health Organization. Available: http://apps. who.int/iris/bitstream/10665/250141/1/9789241511353-eng. pdf?ua=1
World Bank and Institute for Health Metrics and Evaluation. 2016. The Cost of Air Pollution: Strengthening the Economic Case for Action. Washington, DC: World Bank. Available: http:// documents.worldbank.org/curated/en/781521473177013155/ pdf/108141-REVISED-Cost-of-PollutionWebCORRECTEDfile.pdf
World Health Organization. WHO Global Urban Ambient Air Pollution Database (Update 2016). Available: www.who.int/phe/ health_topics/outdoorair/databases/cities/en/