Published Studies on the Health Effects of Woodsmoke.

UNE Air Quality Research Group

This research was published as a refereed paper in Clean Air, V32, No 3, August 1998.



Developing Countries

Developed Countries

City-Wide Detrimental Health Effects

Laboratory evidence



Developing Countries

Exposure to woodsmoke can have serious consequences for health. In developing countries, some stoves may emit smoke to indoor air. Several studies have shown that exposure to woodsmoke reduces lung function, especially in children, and increases coughs and other respiratory diseases. In Bogota, woodsmoke exposure may explain about half of all cases of obstructive airways disease. In Mexico City, famous for its traffic pollution, women exposed to woodsmoke had 3.9 times the risk of chronic bronchitis and 9.7 times the risk of chronic bronchitis plus chronic airway obstruction. If exposed for an average of 33 mins or more a day (200 or more hours/year), risks were 15 and 75 times higher than in women not exposed to woodsmoke. In Arfica, cooking with wood increased the risk of stillbirth by 50%. Homes with woodstoves were found to have greater concentration of mutagenic material in the air, resulting in increased risk of cancer. There results are summarised in Table 1.

Table 1. Symptoms from woodstove use, developing countries. Text in italics constitutes direct quotes from the paper.


Symptoms in households using wood heat or cooking in developing countries

Dennis et al., 1996. Conclusions: "This study showed that among elderly women of low socioeconomic status in Bogota, woodsmoke exposure is associated with the development of Obstructive Airways Disease (OAD) and may help explain around 50% of all OAD cases."
Ellgard, 1996 "The association between exposure to air pollution from cooking fuels and health aspects was studied in Maputo, Mozambique. Almost 1200 randomly selected women residing in the suburbs of Maputo were interviewed … Wood users were found to have significantly more cough symptoms than other groups. This association remained significant when controlling for a large number of environmental variables."
Gharaibeh, 1996 A significant negative impact was found with regard to environmental exposure to both passive smoking and wood and kerosene unvented cooking stoves. Reduction in lung function for wood and kerosene were FVC (L): 1.02-1.32; FEV(1) (L): 0.91-1.25; FEF(25-75) (L/S): 1.24-1.86; PEFR (L/S): 1.67-2.64. Primary school children in Jordan.
Guneser et al., 1994 "Pulmonary function levels were diminished in passive smokers and in children whose houses were heated by a wood-burning stove As a result, passive smoking, using a wood-burning stove for heating, and family history of respiratory diseases are to be considered risk factors for the respiratory system." 617 primary school children, aged 9-12 studied in Turkey.
Hamada et al., 1992 "Strikingly higher (p < 0.01) levels of Polycyclic Aromatic Hydrocarbons, and much higher (p = 0.07) levels of Suspended Particulate Matter were found in the kitchens with wood stoves. These findings appear to support the hypothesis that domestic wood burning stoves are risk factors for some upper digestive and respiratory tract cancers in Brazil."
Perez-Padilla et al., 1996 "A case-control study was performed in women older than 40 yr of age in Mexico City to evaluate the risk of cooking with traditional wood stoves for chronic bronchitis and chronic airway obstruction (CAO). Crude odds ratios for wood smoke exposure were 3.9 (95% CI, 2.0 to 7.6) for chronic bronchitis only, 9.7 (95% CI, 3.7 to 27) for CAO plus chronic bronchitis, and 1.8 (95% CI, 0.7 to 4.7) for CAO … The findings support a causal role of domestic wood smoke exposure in chronic bronchitis and chronic airflow obstruction."
Sandoval et al., 1993 Describes " the clinical, radiologic, functional, and pulmonary hemodynamic characteristics of a group of 30 nonsmoking patients with a lung disease that may be related to intense, long-standing indoor wood-smoke exposure. ... Fibrous and inflammatory focal thickening of the alveolar septa as well as diffuse parenchymal anthracotic deposits are the most prominent pathologic findings. Pulmonary arterial hypertension in wood-smoke inhalation-associated lung disease appears to be more severe than in other forms of interstitial lung disease and tobacco-related COPD." Authors' address: Mexico City.
Zhang et al., 1996 "Among the 6 fuel/stove combinations, wood stove generated the highest cancer risk and LPG generated the least risk … The cancer risk of benzene or that of styrene from use of biomass cookstoves might exceed the risk from all sources of airborne benzene or styrene in the US."
Ardayfio Schandorf E, 1993 "cooking over an open fuel wood stove resulted in an almost 50% greater chance of stillbirth among pregnant women."

Developed Countries

Many recent studies have also found significant effects in woodstove users in developed countries such as the US. Children living in such homes tend to have increased respiratory symptoms, acute lower respiratory infection, increased frequency and severity of wheeze, increased frequency of cough and waking up at night with a cough. Measured lung function has also been found to decline (See Table 2). Forest firefighters in the US had decreased lung function after fighting fires. Use of wood for indoor heat by these firefighters was also associated with decreased lung function. Van Houdt et al. (1986) found increased indoor mutagenicity in 8 out of 12 homes using woodstoves. In New York, woodstove use was found to be significantly associated (P<.05) with increased otitis (an inflammation of the middle ear marked by pain, fever, dizziness, and abnormalities of hearing.) Studies detecting adverse health effects in families using wood heating are listed in Table 2. Three studies, also listed in Table 2 show increased mutagenicity or genotoxity of the air inside, or in the vicinity of, homes using wood heating. Three case studies, from North Carolina, Pennsylvania and Canada, describe illnesses attributed specifically to fumes from woodstoves.

Very few studies have found little or no effect of exposure to woodsmoke or woodstove use. Exceptions are one study, in Bavaria, (Von Mutius et al., 1996) which found that woodstove use was associated with a reduced incidence of allergies to pollen. The authors postulated that this may have been due to wood heating being associated with farming or country areas, where children were more likely to be exposed to pollens eg from haymaking. A study in South Australia (Volkmer et al. 1995) found parental smoking was associated with decreased eczema and use of evaporative cooling was associated with increased dry cough. Woodstove use was also associated with many factors. However, when the data for Adelaide where considered by themselves, most of these apparent effects were no longer significant. Some of the associations could therefore be explained by differences in woodstove use or other factors between city and country areas. The study was not adjusted for socio-economic factors except on an area-wide basis. Some of the methodological problems which led to the apparent associations in the State-wide analysis may well remain in the data for Adelaide, explaining the unusual associations.

It is also possible that, as woodstoves become more air tight, neighbours downwind of the stove receive at least as much of the noxious pollution as the family using the stove, making such effects easier to detect on an area-wide basis, rather than by heating methods for individual families. Alfheim et al. (1984) found the effect on the mutangenicity of the air of burning wood in an airtight stove was in the vicinity of the house, rather than in the room with the stove. Table 3 shows how use of woodstoves increases health risk for entire cities or areas.

Table 2. Studies in the US and other developed countries of symptoms associated with households using wood heat.


Respiratory and other symptoms in households using wood heat

Honicky et al., 1985 Moderate and severe respiratory symptoms were significantly greater ( P<.001) in 34 children, aged 1-7 years in houses with woodstoves than in 34 children houses without. Conclusion: "Present findings suggest that indoor heating with wood-burning stoves may be a significant etiologic factor in the occurrence of symptoms of respiratory illness in young children." Michigan, US.
Butterfield, et al., 1989 Significant correlation (P<.01), between woodstove use and frequency of wheeze, severity of wheeze, frequency of cough and waking up at night with cough , based on 59 subjects aged 1 to 5.5 years.
Lipsett et al., 1991 Presence of woodstove or fireplace in the home was associated with shortness of breath in females and both shortness of breath and moderate or severe cough in males (p<0.01 for all cases). 182 asthmatics living in Denver, Colorado.
Betchley et al., 1997 Forest firefighters had significant declines in lung function (FEV(1)). Average declines, pre-shift to midshift of 0.089 L, 0.190 L, and 0.439 L/sec in TVC, FEV(1) and FEF (25-75). The use of wood for indoor heat also was associated with the declines in FEV(1).
Morris et al., 1990 58 Navajo children under 2 years with diagnosed pneumonia or bronchiolitis were compared with matched control children. Use of a wood burning stove was associated with a 4 times higher risk of lower respiratory tract infection (P<.001).
Robin et al., 1996 Matched pair analysis revealed an increased risk of Acute Lower Respiratory Infection (ALRI) for children living in households that cooked with any wood (odds ratio 5.0; 95% confidence interval 0.6 to 42.8. Cooking with wood-burning stoves was associated with higher indoor air concentrations of respirable particles and with an increased risk of ALRI in Navajo children. Studied 45 children under 2 years.
Tuthill, 1984. Risk of respiratory symptoms increased by 10%, but this was not statistically significant. Study of children aged 5-11, 258 with woodstoves, 141 without. Exposure to formaldehyde from any source, including wood burning, significantly increased risk.
Daigler et al., 1991 A comparison of patients in New York with physician-diagnosed otitis media (n = 125, 74% response), and controls (n = 237, 72% response) showed exposure to a woodburning stove was significantly associated (P<.05 with increased otitis (an inflammation of the middle ear marked by pain, fever, dizziness, and abnormalities of hearing.)
Hogg, 1997 The author comments on the case report by Dr. David T. Janigan and colleagues of classic bronchiolitis obliterans in a man who used a wood-burning stove to dispose of construction materials in Candada.
Dean et al., 1992 Case of methemoglobinemia, sudden onset of cyanosis, irritability, metabolic acidosis, and a lethal methemoglobin level of 71.4% in a 10 week old infant. Family history revealed a wood-burning stove which emitted pine tar fumes as the potential environmental methemoglobin-producing source. The infant's cradle was situated five feet from the stove. The baby was treated and recovered.
Ramage et al., 1996 Case study of 61-yr-old woman suffering shortness of breath on exertion and interstitial lung disease. Bronchoalveolar lavage revealed numerous carbonaceous particulates and fibers, as well as cellular and immunoglobulin abnormalities. Inflammation and fibrosis were found surrounding them on open biopsy. The particle source was traced to a malfunctioning wood-burning heater in the patient's home.
Von Mutius et al., 1996 Children in Bavarian homes heated with wood had less hayfever and were less sensitive to pollen, cold air or other irritatants. Postulated that wood heating may be more common in farming or country areas where children are more likely to be exposed, and hence desensitised, to pollens eg from haymaking.
Volkmer et al., 1995 "The use of a wood fire/heater compared to other forms of heating was significantly associated with a reduced prevalence rate for dry cough (OR 0.84) and ever having wheezed (OR 0.82)." Findings contradict most other studies. Data not adjusted for confounders such as socio-economic status, except by area code.
van Houdt et al., 1986 "The use of wood stoves caused an increase of indoor mutagenicity in 8 out of 12 homes."
Boone et al., 1989 "Woodsmoke prove to be a major source of indirect genotoxins in homes. The increase is probably due to higher concentrations of polycyclic aromatic hydrocarbons in the wood smoke aerosol …" USA.
Alfheim et al, 1984 "Whereas wood heating in an "airtight" stove was found to cause only minor changes in the concentration of PAH and no measurable increase of mutagenic activity of the indoor air, both these parameters increased considerably when wood was burned in an open fireplace, yielding PAH concentrations comparable to those of ambient urban air. Woodburning in the closed stove did, however, result in increased concentrations of mutagenic compounds and PAH on particles sampled in the vicinity of the house."


City-Wide Detrimental Health Effects When a Proportion of the Population uses Wood Stoves

In areas with airtight stoves, the smoke is vented to the outside air. Neighbours living downwind probably suffer greater exposure. The whole neighbourhood has been found to suffer as a result. For example, in Seattle, where 90% of particulate air pollution was found to originate from woodsmoke, hospital admissions for asthma were found to increase on days with high pollution. Similar effects have been observed in Santa Clara County, where residential wood burning is also a principal source of particulate air pollution. Separate studies in Klamath Falls, Oregon and in Montana found declines in measured lung function in children when ambient levels of woodsmoke were high. Declines in lung function with woodsmoke exposure were also found found for asthmatic children in Seattle. In Sydney, the majority of particle pollution in winter arises from a less than 13% of of households using wood heaters (EPA NSW, 1996). Particle pollution on a daily basis has found to be associated with increased death rates (Morgan, 1996). In Boise, Idaho, US, air containing smoke emitted from woodheaters was tested for mutagenicity using the Ames test on salmonella and tumor initiation assays in mice. Woodsmoke was estimated to be 12 times more carcinogenic than an equal concentration of cigarette smoke (Lewtas et al., 1991).

Table 3. Studies relating outdoor concentrations of woodsmoke to adverse health effects in the whole population.


Adverse effects of Outdoor Woodsmoke

Schwartz, 1993 Significant association between visits to 8 hospital emergency departments in Seattle for asthma and PM10 pollution. In 1993, wood burning was found to be the dominant source of PM10 pollution in Seattle in all seasons of the year, ranging from 60% in summer to 90% in winter.
Koenig et al., 1993 Significant association in Seattle (where the majority of particulate air pollution originates from woodsmoke) between outdoor fine particle pollution and decreased lung function (measured by spirometry) in asthmatic children aged 8-11.
Heumann et al., 1991 Children with the highest exposure to wood smoke had a significant decrease in lung function, measured by FEV1 and FVC. 410 children aged 8-11 in Klamath Falls, Oregon.
Johnson, 1990 Particle pollution from woodsmoke in the air was associated with significant decreases in lung function in children aged 8-11. 495 subjects in Montana.
Browning, et al., 1990 No statistically significant differences, but a pattern of increased symptoms and chronic illness in children aged 1-5 in the area with high wood smoke.
Lipsett et al., 1997 Conclusion from abstract: "These results demonstrate an association between ambient wintertime PM10 and exacerbations of asthma in an area where one of the principal sources of PM10 is Residential Wood Combustion." Santa Clara County, California.
Betchley, et al., 1997 Forest firefighters had significant declines in lung function (FEV(1)). Average declines, pre-shift to midshift of 0.089 L, 0.190 L, and 0.439 L/sec in TVC, FEV(1) and FEF (25-75). ). The use of wood for indoor heat also was associated with the declines in FEV(1).
Morgan, 1996

EPA NSW, 1996 for % from wood

In Sydney, Australia, death rates are generally higher in winter and increase on days of, or days following high particle pollution. Particle concentrations are generally higher in winter. Carbon dating of air samples at Rozelle, near the CBD in July and August 1993, found that two-thirds of particles originated from wood, not coal, oil or diesel. At Winmalee in the Blue Mountains, the proportion was 80%. Samples were taken when the air was essentially free of smoke from bushfires or hazard reduction burns. Solid fuel use in Sydney increased from 7% of households in 1988 to 13% in 1995. The majority of winter particle pollution in Sydney is thus caused by a small minority of households using wood heaters.
Lewtas et al., 1991 Mutagenicity testing of air containing smoke emitted from woodheaters in Boise, Idaho, US, using the Ames test on salmonella and tumor initiation assays in mice found that woodsmoke was 12 times more carcinogenic than an equal concentration of cigarette smoke.
Larson & Koenig, 1994. "We conclude that the preponderance of the data suggest a causal relationship between elevated wood smoke levels and adverse respiratory health outcomes in young children."


Laboratory evidence

Mounting laboratory evidence is now available to explain many of these adverse effects. Lal et al. (1993) describe in detail some of the nasty effects seen in lungs of rats exposed to woodsmoke. Other researchers have found deleterious effects after only 1 hour's exposure to 800ug/m3 of woodsmoke. Following exposure, bacteria in rats' lungs were found to be more virulent Stone, 1995). In Australia, woodsmoke concentrations can approach this level and remain high for serveral hours. For example, in Armidale, NSW, measured hourly woodsmoke concentrations have peaked at 467 ug/m3 and remained over 200 ug/m3 for several hours (for more details see Monitoring Armidale's Air Quality.) Another experiment (Stone, 1995), subjected mice either to woodsmoke, oil furnace fumes, or clean air for 6 hours. Mice were then challenged by a respiratory bug. 21% of the mice exposed to wood smoke were dead two weeks later, compared with only 5% mice exposed to fumes from the oil furnace or to clean air. The article comments: "Part of the problem is that wood smoke is a witch's brew of carcinogens, including aldehydes and polycyclic aromatic hydrocarbons, carbon monoxide, and organic particles less than 10 microns in diameter, called PM10. PM10s have been implicated in increased morbidity and mortality on days of heavy air pollution."

Very recent research has found that only the smaller particles less than 2.5 microns are linked with mortality and morbidity. Schwartz et al. (1996) analysed daily mortality in six Eastern US Cities over 8 years. Particles between 2.5 and 10 microns had no association with daily mortality. The strongest association was with PM2.5s, particles less than 2.5 microns in diameter, which arise mainly from combustion. Woodsmoke is almost entirely PM2.5s. (Larson and Koenig, 1994). Autopsies have shown that particles less than 2.5 microns in diameter are retained in human lungs, but not larger particles (Churg et al., 1997). Godleski et al. (1996) exposed rats with bronchitis for 6 hours per day to 272ug/m3 PM2.5 (ie daily average of 68 ug/m3 - similar or lower to PM2.5 concentrations measured in a residential area of Armidale on 32% of nights in June-August 1995 and 1996). 37% of rats exposed to particles died, compared to none exposed to filtered air.


Physiological measurements of effects of exposure to woodsmoke

Stone, 1995 Mice were exposed for 6 hours to wood smoke, emissions from an oil furnace or no pollution (control) and then an aerosol of the bacterium Streptococcus zooepidemicus, which causes severe respiratory infections. After 2 weeks, 5% of the mice in the control group exposed to air and bacteria had died, along with a similar percentage of the mice breathing the oil fumes. But 21% of the wood-smoked mice were felled.
Stone, 1995 Rats were exposed to no pollution or 800 ug/m3 wood smoke for 1 hour, then to golden staph bacteria. The bacteria were more virulent in animals which breathed the woodsmoke. This was attributed to a suppression in activity of the rats' macrophages, immune cells that roam the body, looking to engulf and destroy foreign particles.
Kou et al. , 1997 "These results suggest that an increase in OH. burden following smoke inhalation is actively involved in evoking the acute irritant effects of wood smoke on breathing in rats."
Rao et al., 1995 Metabolites of woodsmoke condensate accumulate in cultured rat eye lenses, compromising ability to accumulate rubidium-86 (mimic of K) and choline. Says may explain implication of smoke in cataract.
Lal et al., 1993 Rats exposed to woodsmoke suffered "bronchiolitis, hyperplasia and hypertrophy of bronchiolar epithelial lining cells, some necrosed lining cells desquamated into lumens, congestion of parenchymatous blood vessels, oedema, hyperplasia of lymphoid follicles, peribronchiolar and perivascular infiltration of polymorphonuclear cells, and mild emphysema" Conditioned worsened with accumulated exposure ."The results indicate progressive pathomorphological pulmonary lesions with subsequent exposure to wood smoke in controlled conditions."
Churg et al., 1997. Autopsies were carried out of lung tissue from 10 never-smoking long-term residents of Vancouver. Retained particles in human lung parenchyma were counted, sized, and identified by analytical electron microscopy. 96% of particles had aerodynamic diameter less than 2.5microns.
Godleski et al., 1996 Rats with bronchitis were exposed for 6 hours per day to 272ug/m3 PM2.5 (ie daily average of 68 ug/m3 - similar or lower to PM2.5 concentrations in parts of Armidale). 37% of rats exposed to particles died, compared to none exposed to filtered air.



More than 40 studies relating woodsmoke to health are referenced here. They cover all aspects, from use of wood for cooking or heating in developing countries, to use of modern heaters in developed countries, to detrimental health effects found in whole towns or cities when woodsmoke is allowed to build up. The epidemiological evidence is supported by tests on laboratory animals, which show exposure to woodsmoke reduces the ability of the lungs to fight infection. After exposure to a pathogen, 21% of mice exposed to woodsmoke became sick and died within 2 weeks, compared with 5% exposed to oil furnace fumes.

It would be totally dishonest and misrepresentative of the wealth of scientific evidence presented here to selectively quote only the two studies cited here that do not show adverse effects of woodsmoke.

Back to Air Quality Home Page