Published ahead of print on June 23, 2006, doi:10.1164/rccm.200503-479OC
© 2006 American Thoracic Society doi: 10.1164/rccm.200503-479OC
The Effect of Biomass Burning on Respiratory Symptoms and Lung Function in Rural Mexican WomenNational Institute of Respiratory Diseases, Mexico City, Mexico; and School of Occupational and Environmental Hygiene and Respiratory Division, Department of Medicine, University of British Columbia, Vancouver, British Columbia, Canada Correspondence and requests for reprints should be addressed to Justino Regalado, M.D., Instituto Nacional de Enfermedades Respiratorias, Tlalpan 4502, Mexico DF 14080, Mexico. E-mail: jregalad{at}compuserve.com.mx
Background: The use of biomass as a cooking fuel is commonplace in developing countries and has been associated with chronic bronchitis and obstructive airway disease. Methods: A cross-sectional survey was done in the village of Solis, close to Mexico City. Lifelong nonsmoking women 38 yr of age or older (n = 841) completed a questionnaire on respiratory symptoms and illnesses and on cooking fuel use, and performed spirometry in their homes. Particulate matter concentration was measured with a nephelometer in the kitchen for 1 h, while the subject was cooking. Results: The peak indoor concentration of particulate matter (PM10, particles with a diameter of 10 µm or less) often exceeded 2 mg/m3. Compared with those cooking with gas, current use of a stove burning biomass fuel was associated with increased reporting of phlegm (27 vs. 9%) and reduced FEV1/FVC (79.9 vs. 82.8%). Levels of FEV1 were 81 ml lower and cough was more common (odds ratio, 1.7; 95% confidence interval, 1.0–2.8) in women from homes with higher PM10 concentrations. All women found with moderate airflow obstruction (Global Initiative for Chronic Obstructive Lung Disease stage II and above) were cooking with biomass stoves. Conclusion: Women cooking with biomass fuels have increased respiratory symptoms and a slight average reduction in lung function compared with those cooking with gas.
Key Words: biomass • indoor air pollution • Mexico • women • wood The use of wood and other forms of biomass as a cooking fuel is commonplace in developing countries (1–3). In rural areas of Mexico, biomass is used as the primary cooking fuel in 69% of households (4). Wood smoke contains hundreds of chemical compounds (5). Some of the components present in wood smoke that are of concern for health include particles, polycyclic aromatic hydrocarbons, and carbon monoxide (6). Exposure to biomass smoke has an important global impact on mortality and morbidity (1, 7). Reported respiratory effects of chronic exposure to wood smoke and other forms of biomass in adults from developing countries include an increased prevalence of chronic bronchitis (8–10), respiratory failure, and cor pulmonale (10, 11). We previously reported an increased risk of chronic bronchitis and obstructive airway disease associated with cooking with wood in a case-control study of Mexican women (12); similar findings have been described by other investigators in case-control studies (13, 14) and in community studies (15–19). Acknowledged limitations of the observational evidence to date on the adverse effects of exposure to biomass smoke include inadequate measurement of exposure and inadequate control for potential confounders. To attempt to address these limitations, and to determine the health impact of biomass burning in a community setting, we examined the effect of burning biomass on respiratory symptoms and on level of lung function among lifelong, nonsmoking women in the Mexican village of Solis. An objective measure of exposure to biomass burning was used. Some of the results have been previously reported in the form of an abstract (20).
The study was conducted in the rural village of Solis, composed of 13 small communities, located 200 km northwest of Mexico City and at an elevation of 2,600 m above sea level. Women in Solis have cooked with wood, crop residues, and corn cobs as fuel, but now some of them also use natural gas either alone or supplemented with biomass fuel. The study was approved by the Ethics Committee of the National Institute of Respiratory Diseases (Mexico City, Mexico). The survey was discussed with the community leaders and announced to the population. All women in Solis 38 yr of age or older, identified in a house-by-house census, were asked to participate and give written consent. The women completed a respiratory questionnaire (21) with additional questions about the type of current and past cooking fuels used. From the questionnaire and home visit, we identified the current fuel used for cooking, and those women previously exposed to biomass smoke for at least 6 mo. We expressed the questionnaire-based cumulative exposure to biomass smoke in hour-years (the product of number of years of exposure, and average hours per day of exposure). Subjects performed spirometry with a calibrated Pony turbine spirometer (Cosmed USA, Inc., Chicago, IL) in accordance with American Thoracic Society recommendations (22). At least three acceptable and two reproducible (FVC and FEV1 within 5% and 100 ml) forced expiratory maneuvers were required for analysis. FEV1 and FVC were expressed in liters and also as a percentage of the predicted value by internally derived equations, obtained by linear regression models (adjusted by age and height, R2 of 0.55 for FEV1 and 0.50 for FVC) in the 372 subjects lacking respiratory symptoms and diseases diagnosed by physicians. Chronic obstructive pulmonary disease (COPD) was defined as an FEV1/FVC less than 70% (23). Indoor particle concentrations were measured in the kitchen for 1 h when the cooking stove was burning the fuel most commonly used. We used a portable integrating nephelometer (M903; Radiance Research, Seattle, WA) as described previously (24, 25). At the end of each day of work, the nephelometer zero value was verified by injecting dry filtered room air, and it was reset with dry filtered Freon 12 if necessary. Inhalable particle concentration (PM10, particles with a diameter of 10 µm or less) was estimated in a simple linear regression model, in which PM10 was regressed on the extinction coefficient obtained from nephelometry (24, 25).
Data Analysis
Data were collected from September 1994 to June 1995. A total of 871 women agreed to participate in the study (83% of those eligible), but the report was done with data from 841 (97%) who never smoked. The population had a 5.1% prevalence of asthma diagnosed by a physician, 23% were obese, 13% had an FEV1/FVC < 70%, and 37% reported having cough or phlegm most of the day. At the time of the survey most women were using at least some biomass fuel for cooking. Only 8% of the subjects used a gas stove exclusively; 10% reported the use of a wood stove only, 73% used wood and gas stoves, and 8% reported the use of a wood stove with a chimney, without a gas stove. Table 1 shows demographic data by current biomass smoke exposure. Women exposed to biomass smoke were slightly older, less educated, and reported less income than those currently using only natural gas. Women cooking with biomass stoves at the time of the survey had used biomass stoves for more days, years, and hour-years than women cooking with gas stoves, as would be expected. However, it is apparent that most women using gas stoves had used previously biomass stoves. The concentration of particulate matter measured while cooking was considerably higher among women cooking with biomass stoves (Figure 1).
Both groups were exposed similarly to second-hand tobacco smoke (28.8% of those cooking with gas, and 27.3% of those cooking with biomass stoves; p = not significant). Other occupational exposures to dust or smoke were not present, as expected in women living in rural areas of Mexico. No significant differences were found in FEV1 or FVC, but the FEV1/FVC ratio was lower in women cooking with wood stoves than in those cooking with gas stoves. Table 2 presents the prevalence of respiratory symptoms, diseases, and level of lung function by current cooking fuel. COPD in Global Initiative for Chronic Obstructive Lung Disease (GOLD) stage I (FEV1/FVC < 70%) and higher was slightly more common in women cooking with biomass stoves (p = not significant), but all women with GOLD stage II and above (FEV1/FVC < 70% and FEV1 < 80% predicted) belonged to the group cooking with wood stoves.
The group currently burning biomass fuel reported more frequent respiratory symptoms, but especially more phlegm, than the group cooking with gas stoves (see Table 2). Even though women cooking with wood stoves reported more frequent wheeze than the gas stove group, no significant difference was found in doctor-diagnosed asthma. Results of the logistic regression analysis of symptoms are shown in Table 3. After adjusting for passive smoking and level of income, women cooking with biomass stoves had more phlegm "most of the day" and phlegm for more than 3 mo/yr than did women cooking with gas stoves. Hour-years of exposure to biomass smoke was associated with cough or phlegm most of the day, especially if exposure lasted more than 3 mo/yr. Wheezing in the past had a marginal association with current cooking with a biomass stove, after adjustment for passive smoking and income. In these models, passive smoking was associated with wheezing in the past (odds ratio, 1.5; 95% confidence interval, 1.1–2.0).
In regression models adjusted by age, passive smoking, and income (Table 4), current exposure to biomass smoke was associated with a 2.8% decrease (3.5% of predicted) in FEV1/FVC ratio but no significant effect for FEV1 or FVC. Further adjustment by cumulative exposure (hour-years) reduced the coefficients and made them statistically nonsignificant (see Table 4).
There were no differences between the wood and gas groups in prevalence of airway obstruction defined by FEV1/FVC ratios of 70 (Table 2), 65, or 60% (data not shown). Peak PM10 concentrations were highest in homes burning biomass when no chimney was present. As expected, peak PM10 concentrations were lower in homes when only gas was used for cooking (Figure 1). To evaluate the effect of current exposure to indoor PM10, the population sample was restricted to those 410 women who had performed measurements of PM10 while using a wood stove, 12% with a chimney. Results of multiple logistic regression analysis and multiple regression analysis can be seen in Tables 3 and 4. A peak PM10 concentration higher than 2.6 mg/m3 (compared with those with less than 2.6 mg/m3) was associated with cough for more than 3 mo/yr and with COPD at GOLD stage II and above (FEV1/FVC < 70% and FEV1 < 80% of predicted), but not with phlegm. Analysis of lung function in this subgroup (adjusted by age, income, passive smoking, and hour-years) showed a decrease in FEV1 (–81 ml or –4.7% predicted) and in FVC (–122 ml, or –3.9% predicted) without any significant change in FEV1/FVC ratio for women with a peak PM10 concentration higher than 2.6 mg/m3 compared with those with a peak PM10 concentration less than 2.6 mg/m3. Models with and without adjustment for cumulative exposure to biomass (hour-years) had similar coefficients. Similar results were obtained in regression models restricted to the 808 women with measurements of PM10, including those done while using a biomass stove and the remaining while using gas stoves.
Indoor particle concentrations were high in homes using biomass fuel for cooking even with a chimney (Figure 1). Peak particle concentrations measured indoors during biomass combustion were greater than 3 mg/m3 in 11% of households. This finding is similar to previous reports of indoor particle concentrations in developing countries (3, 5, 7, 24). The major source of indoor particulate matter in this setting is the cooking stove, as the measurement was done in the kitchen while cooking, to estimate better the exposure in women. Nephelometry has been shown to be a feasible and valid method for monitoring indoor concentration in this setting (25). Women who currently cooked with biomass fuels had more phlegm production than did women who cooked with gas, and had relatively lower FEV1/FVC ratios. The average impact of exposure on lung function was mild. In the subset of women using biomass fuels during the measurement of indoor PM10 concentrations, PM10 concentration was associated with a reduction in FEV1 and FVC, and an increase in reported cough. On the other hand, all 21 women with airflow obstruction (FEV1/FVC < 70%) and FEV1 < 80% predicted (GOLD stage II and above for airflow obstruction) were cooking with wood stoves. Given the findings of our earlier case-control study (12) and those of others (13, 14), in which odds ratios of chronic bronchitis and airflow obstruction were increased severalfold in women who had been exposed to wood smoke, we had expected to find more substantial adverse effects. A difference in study population and design is one possible reason for this apparent disparity. In the case-control study, patients were referred to a referral chest hospital with illness or symptoms. In the current cross-sectional study, women were recruited regardless of previous health status from a rural community setting. These women likely comprise a relatively homogeneous population that exhibits little variation in health status. On the other hand, it is clear that a group of women exposed to biomass smoke develop severe disease with many features of COPD due to smoking (26, 27) and, in addition, have a shortened survival just as smokers with COPD (26). However, on average, those women with COPD associated with biomass smoke exposure and seen at a referral center had milder airflow obstruction than did patients with COPD due to smoking (26), consistent with our survey. We do not know the reasons why exposure to biomass smoke, even with the severity shown in this and other studies, does not produce airflow obstruction more often, but it is clear that exposure increases considerably the prevalence of respiratory symptoms, even in population-based surveys, and contributes to premature mortality as demonstrated in hospital studies. In other community studies done in developing countries, biomass exposure was also associated consistently with chronic bronchitis, but effects on level of lung function were variable or small (15–17, 28–31). Only 17 subjects out of the total sample population (2%) reported never having used wood or other forms of biomass as a cooking fuel. We could not compare properly this small-sized group of never-exposed women with those ever exposed to biomass smoke. We used, as alternatives, a cumulative exposure index (hour-years) to assess possible effects of chronic exposure in a dose–response fashion, an indicator variable of current exposure to biomass smoke and the peak concentration of PM10. Hour-years of exposure in the whole population was associated with cough alone or mixed with phlegm, even taking into account current exposure. However, the impact on lung function was mild and inconsistent, confirming either the low average impact of exposure to biomass smoke on lung function, or the reduced susceptibility of the studied women. Hour-years is based on a report by the participants, and therefore dependent on memory and subject to bias, but, on the other hand, women in rural areas tend to recall well the time when they started to cook regularly, usually when they marry. Because of the cross-sectional nature of the study, it is impossible to decipher properly the causal sequences occurring over time. The studied town was in transition from the use of biomass fuel to gas stoves, like other, but not all, small towns in Mexico. This rise in the "energy ladder," toward cleaner fuels is usually associated with changes in other indicators of socioeconomic status that may also influence the development of respiratory symptoms and disease. In summary, women exposed to biomass smoke reported more frequent cough and phlegm than did women cooking with gas, but with minimal average adverse changes in lung function. Severe airflow obstruction in our cross-sectional community survey was uncommon (2.5% of women had COPD stage II and higher; and 0.7% had COPD stage IV); all of them were cooking with biomass stoves.
The authors express their deepest thanks to the people of the Mexican village of Solis, without whose help and cooperation this study could not have been carried out.
Supported by a grant from the International Development and Research Council of Canada. Originally Published in Press as DOI: 10.1164/rccm.200503-479OC on June 23, 2006 Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
