Theses Doctoral

As the smoke clears: assessing the air pollution and health benefits of a nationwide transition to clean cooking fuels in Ecuador

Gould, Carlos Francisco

Air pollution is the world’s greatest environmental health risk factor and reducing exposure remains an ongoing challenge around the world. Among the world’s poor, marginalized, and rural populations, household air pollution from the inefficient burning of biomass fuels like firewood, charcoal, dung, and agricultural residues for daily household energy needs is a leading cause of morbidity and mortality, especially for children under the age of five years. However, household air pollution exposure is a modifiable risk factor and clean-burning cooking fuels like gas and electricity promise substantial health benefits for the 2.8 billion people reliant on biomass fuels. But, clean fuels remain prohibitively expensive or inaccessible for those most reliant on biomass fuels. It is in this context that I examine Ecuador – where substantial cooking gas subsidies have facilitated a nationwide transition to household clean fuel use over four decades – as a long-term case study to understand the potential for widespread clean fuel uptake to reduce air pollution exposure and improve children’s health. Chapter 1 provides background information that contextualizes the work presented in this dissertation.

In Chapter 2, I discuss the development of Ecuador’s clean fuel policies. Originally established as a part of broad social support reforms in the 1970s, direct-to-consumer subsidies that lowered the cost of liquefied petroleum gas (LPG) – a popular clean-burning cooking fuel used widely around the world – have driven a transition from 80% of households cooking primarily with firewood in the 1970s to now 90% relying primarily on LPG. However, widespread clean fuel use has come at a cost; each year, the Government of Ecuador spends approximately 1% of the national growth domestic product subsidizing LPG (300-700 million USD). To reduce this financial burden, the government has offered incentives to households to install and use induction electric stoves, which can be powered by the nation’s growing hydroelectric capacity, thereby reducing the cost of LPG subsidies and greenhouse gas emissions. To supplement national data, I administered household energy surveys in a northern province, finding that while all households regularly used LPG, 80% still used firewood for cooking. In these remote regions along the Colombian border, limited access to LPG cylinder refills remains a significant barrier to exclusive LPG use, even after decades of building a robust distribution system.

In Chapter 3, I describe results from tailored household surveys – covering energy end uses, costs and access to fuels, and fuel use preferences – administered in 808 households across coastal and Andean Ecuador. Nearly all participants reported using LPG for more than a decade and having frequent, convenient access to cheap LPG cylinder refills. Nonetheless, half of rural households and one-fifth of peri-urban households relied on firewood for cooking and to meet specific household energy needs, like space heating or heating water for bathing. Induction stoves were rare and many induction stove owners reported zero use because the required equipment had never been installed by electricity companies, their stove had broken, or they feared high electricity costs. Here, I show that nationally-representative surveys reporting only “primary cooking fuel” use may underestimate solid fuel use as a secondary option, particularly in rural areas where LPG fuel availability issues play an important role in cooking fuel decision making. These findings additionally indicate that persistent biomass use may curtail the benefits from even the most aggressive clean fuel policies and suggest that additional targeted interventions may be needed to more fully displace biomass. Furthermore, they highlight the need for more nuanced nationally- and subnationally-representative surveys to better understand the extent to which biomass fuels are used secondary to LPG throughout all regions of Ecuador.

In Chapter 4, I present results from a sensor monitoring study where we measured personal PM2.5 exposure and stove use for 48-hour monitoring periods among 157 households in peri- urban and rural Ecuador. Firewood-using participants had higher distributions of 48-hour and 10-minute PM2.5 exposure as compared with primary LPG and induction stove users. Accounting for within-subject clustering, contemporaneous firewood stove use was associated with 101 μg/m3 higher 10-min PM2.5 exposure (95% confidence interval: 94–108 μg/m3). Cooking events with clean fuels were not associated with contemporaneous increases in PM2.5 exposure. These findings confirm our expectations that in a region with low ambient air pollution, long-term cooking gas subsidies can lead to relatively low personal air pollution exposures. And yet, persistent secondary firewood use led to higher average and peak exposures, further motivating the displacement of firewood use to reduce health risks from air pollution. I also outline the methodological challenges faced in combining time-resolved sensor data on participant location, stove use monitors, and PM2.5 concentration and offered advice for future studies.

In Chapter 5, I assess whether Ecuador’s increased clean cooking fuel use has resulted in improvements in under-5 lower respiratory infection (LRI) mortality. Globally, LRIs are the leading cause of death for children under-5 and household air pollution exposure is a leading risk factor. I employ public use data on cooking fuel use and cause-coded mortalities from 1990 to 2019 to estimate the association between clean cooking fuel use and the rate of under-5 LRI mortalities at the canton (county) level in Ecuador. Using generalized additive mixed models with fixed effects for canton and study period, I observed a significant, non-linear negative association providing evidence that when more than 60% of households in a canton cook primarily with a clean fuel, increased clean fuel use is associated with reductions in under-5 LRI mortality. In total, I estimated that increased clean cooking fuel use is associated with 7,343 under-5 LRI mortalities averted since 1990.

In Chapter 6, I conclude by discussing the broader implications of my work. Energy is an important thread connecting climate change, air pollution, and human health, and pathways towards cleaner energy generation will be important drivers of climate change mitigation, reduced environmental exposures, and improved population health. Investments in clean cookstoves have had mixed results over the last 40 years, with many studies revealing only limited uptake of intervention stoves and substantial continued use of polluting traditional fuels, resulting in lower-than-expected exposure reductions and health benefits. Ecuador’s transition has been remarkable in the context of its peer low- and middle-income countries in the rest of Latin America and beyond. The findings presented in this dissertation demonstrate that cooking gas is popular; when it is made cheap and available, gas significantly displaces the use of polluting fuels. Nevertheless, even under ideal cost and access circumstances, my work also reveals that policies and interventions will need to consider specific local needs – like space heating in cold climates – to further encourage a transition toward cleaner indoor air. In the context of efforts to eliminate the use of polluting fuels in the Americas, my work offers hope that ambitious clean cooking fuel policies will improve health.

Geographic Areas


  • thumnail for Gould_columbia_0054D_16757.pdf Gould_columbia_0054D_16757.pdf application/pdf 9.83 MB Download File

More About This Work

Academic Units
Environmental Health Sciences
Thesis Advisors
Jack, Darby W.
Ph.D., Columbia University
Published Here
August 18, 2021