Reanalysis of the Harvard Six Cities Study, part I: validation and replication

Because the results of the Harvard Six Cities Study played a critical role in the establishment of the current U.S. ambient air quality objective for fine particles (PM(2.5)), the U.S. Environmental Protection Agency, industry, and nongovernmental organizations called for an independent reanalysis of this study to validate the original findings reported by Dockery and colleagues in the New England Journal of Medicine (vol. 329, pp. 1753-1759) in 1993. Validation of the original findings was accomplished by a detailed statistical audit and replication of original results. With the exception of occupational exposure to dust (14 discrepancies of 249 questionnaires located for evaluation) and fumes (15/249), date of death (2/250), and cause of death (2/250), the audit identified no discrepancies between the original questionnaires and death certificates in the audit sample and the analytic file used by the original investigators. The data quality audit identified a computer programming problem that had resulted in early censorship in 5 of the 6 cities, which resulted in the loss of approximately 1% of the reported person-years of follow-up; the reanalysis team updated the Six Cities cohort to include the missing person-years of observation, resulting in the addition of 928 person-years of observation and 14 deaths. The reanalysis team was able to reproduce virtually all of the original numerical results, including the 26% increase in all-cause mortality in the most polluted city (Stubenville, OH) as compared to the least polluted city (Portage, WI). The audit and validation of the Harvard Six Cities Study conducted by the reanalysis team generally confirmed the quality of the data and the numerical results reported by the original investigators. The discrepancies noted during the audit were not of epidemiologic importance, and did not substantively alter the original risk estimates associated with particulate air pollution, nor the main conclusions reached by the original investigators.

Reanalysis of the Harvard Six Cities Study, part II: sensitivity analysis

Following the validation and replication of the Harvard Six Cities Study (Krewski et al., this issue), we conducted a wide range of sensitivity analyses to explore the observed associations between long-term exposure to fine particle or sulfate air pollution and mortality. We examined the impact of alternative risk models on estimates of risk, taking into account covariates not included in the original analyses. These risk models provided a basis for identifying covariates that may confound or modify the association between fine particle or sulfate air pollution and mortality, and for identifying sensitive population subgroups. The possibility of confounding due to occupational exposures was also investigated. Residence histories were coded for the study subjects and were used to examine temporal patterns of exposure and risk. Our sensitivity analyses showed the mortality risk estimates for fine particle and sulfate air pollution to be highly robust against alternative risk models of the Cox proportional hazards family, including models with additional covariates from the original questionnaires not included in the original published analyses. There was limited evidence of departures from the proportional hazards assumption. Flexible exposure-response models provided some evidence of departures from linearity at both low and high sulfate concentrations. Incorporating information on changes over time in cigarette smoking and body mass index had little effect on the association between fine particles and mortality. There was limited evidence of variation in risk with attained age, gender, smoking status, occupational exposure to dust and fumes, marital status, heart or lung diseases, or lung function. However, air pollution risk did appear to decreasing with increasing educational attainment. Extensive adjustment for occupation using aggregate indices of occupational "dirtiness" and occupational exposure to known lung carcinogens had little impact on the mortality risks associated with particulate air pollution. Our evaluation of population mobility indicated that relatively few subjects moved from their original city of residence. Attempts to identify critical exposure time windows were limited by the lack of marked interindividual variation in temporal exposure patterns throughout the study period. Overall, this extensive sensitivity analysis both supported the conclusions reached by the original investigators and demonstrated the robustness of these conclusions to alternative analytic approaches.

Specific fatty acid intake and the risk of pancreatic cancer in Canada

The possible association of specific fatty acid (FA) intake and pancreatic cancer risk was investigated in a population-based case–control study of 462 histologically confirmed cases and 4721 frequency-matched controls in eight Canadian provinces between 1994 and 1997. Dietary intake was assessed by means of a self-administered food frequency questionnaire. Unconditional logistic regression was used to assess associations between dietary FAs and pancreatic cancer risk. After adjustment for age, province, body mass index, smoking, educational attainment, fat and total energy intake, statistically significant inverse associations were observed between pancreatic cancer risk and palmitate (odds ratios (ORs)=0.73; 95% confidence intervals (CIs) 0.56–0.96; P-trend=0.02), stearate (OR=0.70; 95% CI 0.51–0.94; P-trend=0.04), oleate (OR=0.75; 95% CI 0.55–1.02; P-trend=0.04), saturated FAs (OR=0.67; 95% CI 0.50–0.91; P-trend=0.01), and monounsaturated FAs (OR=0.72; 95% CI 0.53–0.98; P-trend=0.02), when comparing the highest quartile of intake to the lowest. Significant interactions were detected between body mass index and both saturated and monounsaturated FAs, with a markedly reduced risk associated with intake of stearate (OR=0.36; 95% CI 0.18–0.70; P-trend=0.001), oleate (OR=0.36; 95% CI 0.19–0.72; P-trend=0.002), saturated FAs (OR=0.35; 95% CI 0.18–0.67; P-trend=0.002), and monounsaturated FAs (OR=0.32; 95% CI 0.16–0.63; P-trend<0.0001) among subjects who are obese. The results suggest that substituting polyunsaturated FAs with saturated or monounsaturated FAs may reduce pancreatic cancer risk, independently of total energy intake, particularly among obese subjects.

Epidemiology of pancreatic cancer: an overview

INTRODUCTION

The incidence of pancreatic cancer worldwide appears to correlate with increasing age, and it is slightly more common among men and Jewish people. There is evidence that the incidence rate is higher among blacks than among whites.

METHODS

The published literature was reviewed for preparation of an overview on epidemiology of pancreatic cancer.

RESULTS

A possible role of diabetes in the etiology of pancreatic cancer has been suggested by different epidemiological studies. Several investigations indicate that a history of pancreatitis may increase the risk of pancreas cancer, and it appears that people with a history of pernicious anemia or partial gastrectomy for ulcer as well as cholecystectomy may be at higher risk. Individuals with familial adenomatous polyposis (FAP) also have a high risk of developing this cancer. Pancreatic cancer is seen in some breast cancer families with BRCA1 and BRCA2 mutations. Epidemiological studies have confirmed that relatives of individuals with pancreatic cancer have an increased risk of this malignancy. Affected family members of the familial atypical multiple-mole melanoma (FAMMM) as well as those with a positive family history of ataxia-telangiectasia (AT) have much higher risk of developing pancreatic cancer, compared with the general population. A positive association has been reported between pancreatic cancer risk and dietary intake such as fat and oil, meat, and dairy products, as well as with high intake of energy, fried foods, carbohydrates, cholesterol, and salt. The risk is found to decrease with increased consumption of fresh fruits and vegetables, fiber, natural foods, and Vitamin C. Cigarette smoking has shown the strongest positive association with risk of pancreatic cancer.

CONCLUSION

Some diseases and medical conditions such as diabetes, chronic pancreatitis, AP, family aggregation of pancreatic cancer, FAMMM, AT, as well as nutrition and lifestyle factors, like smoking may play important role in the etiology of pancreatic cancer.

Association between particulate- and gas-phase components of urban air pollution and daily mortality in eight Canadian cities

Although some consensus has emerged among the scientific and regulatory communities that the urban ambient atmospheric mix of combustion related pollutants is a determinant of population health, the relative toxicity of the chemical and physical components of this complex mixture remains unclear. Daily mortality rates and concurrent data on size-fractionated particulate mass and gaseous pollutants were obtained in eight of Canada's largest cities from 1986 to 1996 inclusive in order to examine the relative toxicity of the components of the mixture of ambient air pollutants to which Canadians are exposed. Positive and statistically significant associations were observed between daily variations in both gas- and particulate-phase pollution and daily fluctuations in mortality rates. The association between air pollution and mortality could not be explained by temporal variation in either mortality rates or weather factors. Fine particulate mass (less than 2.5 microns in average aerometric diameter) was a stronger predictor of mortality than coarse mass (between 2.5 and 10 microns). Size-fractionated particulate mass explained 28% of the total health effect of the mixture, with the remaining effects accounted for by the gases. Forty-seven elemental concentrations were obtained for the fine and coarse fraction using nondestructive x-ray fluorescence techniques. Sulfate concentrations were obtained by ion chromatography. Sulfate ion, iron, nickel, and zinc from the fine fraction were most strongly associated with mortality. The total effect of these four components was greater than that for fine mass alone, suggesting that the characteristics of the complex chemical mixture in the fine fraction may be a better predictor of mortality than mass alone. However, the variation in the effects of the constituents of the fine fraction between cities was greater than the variation in the mass effect, implying that there are additional toxic components of fine particulate matter not examined in this study whose concentrations and effects vary between locations. One of these components, carbon, represents half the mass of fine particulate matter. We recommend that measurements of elemental and organic carbon be undertaken in Canadian urban environments to examine their potential effects on human health.

Adjusting for measurement error in the Cox proportional hazards regression model

BACKGROUND

The Harvard Six Cities Study (Dockery et al.) was the first large-scale cohort study to demonstrate an association between long-term exposure to fine particulate matter less than 2.5 microns in aerodynamic diameter (PM2.5) and mortality in urban centres in the United States. Because of the pivotal role of this study in the establishment of the first U.S. national ambient air quality objective for PM2.5 in 1997 (Greenbaum et al.), the results of this study were subjected to an independent detailed re-analysis to test the robustness of the findings to alternative analytic methods (Krewski et al.), including an assessment of the effect of exposure measurement error on estimates of risk based on the Cox proportional hazards model. It is well-known that random measurement error leads to downward bias in estimates of risk, and overstatement of the precision of such estimates.

METHODS

Data from the Harvard Six Cities Study were used to evaluate the potential impact of measurement error on estimates of risk. After introducing a known amount of measurement error into the original data, estimates of risk were calculated using two methods for adjusting for measurement error: regression calibration (RCAL) and simulation extrapolation (SIMEX). With RCAL, the observed value of PM2.5 is replaced by its expected value with respect to the measurement error distribution. SIMEX adjusts for measurement error by adding progressively larger errors to the data and then extrapolating back to the case of no measurement error. Computer simulation was used to evaluate the accuracy and precision of both RCAL and SIMEX, and to assess the robustness of RCAL to mis-specification of the measurement error distribution.

RESULTS AND CONCLUSIONS

When the measurement error distribution was correctly specified, RCAL greatly reduced the downward bias in risk estimates induced by random measurement error, even when the degree of measurement error was relatively large. SIMEX, on the other hand, failed to adequately adjust for the effects of random measurement error in the Cox model, even in the presence of a moderate degree of measurement error. Although RCAL is thus preferable to SIMEX, RCAL was not robust against mis-specification of the measurement error distribution, seriously overestimating (underestimating) risk when the measurement error was overstated (understated).

The use of biologically based cancer risk models in radiation epidemiology

Biologically based risk projection models for radiation carcinogenesis seek to describe the fundamental biological processes involved in neoplastic transformation of somatic cells into malignant cancer cells. A validated biologically based model, whose parameters have a direct biological interpretation, can also be used to extrapolate cancer risks to different exposure conditions with some confidence. In this article biologically based models for radiation carcinogenesis, including the two-stage clonal expansion (TSCE) model and its extensions, are reviewed. The biological and mathematical bases for such models are described, and the implications of key model parameters for cancer risk assessment examined. Specific applications of versions of the TSCE model to important epidemiological datasets are discussed, including the Colorado uranium miners' cohort; a cohort of Chinese tin miners; the lifespan cohort of atomic bomb survivors in Hiroshima and Nagasaki; and a cohort of over 200,000 workers included in the National Dose Registry (NDR) of Canada.

Effect of short-term exposure to gaseous pollution on asthma hospitalisation in children: a bi-directional case-crossover analysis

STUDY OBJECTIVE

Assess associations between short-term exposure to gaseous pollutants and asthma hospitalisation among boys and girls 6 to12 years of age.

DESIGN

A bi-directional case-crossover analysis was used. Conditional logistic regression models were fitted to the data for boys and girls separately. Exposures averaged over periods ranging from one to seven days were used to assess the effects of gaseous pollutants on asthma hospitalisation. Estimated relative risks for asthma hospitalisation were calculated for an incremental exposure corresponding to the interquartile range in pollutant levels, adjusted for daily weather conditions and concomitant exposure to particulate matter.

SETTING

Toronto, Ontario, Canada.

PARTICIPANTS

A total of 7319 asthma hospitalisations for children 6 to 12 years of age (4629 for boys and 2690 for girls) in Toronto between 1981 and 1993.

MAIN RESULTS

A significant acute effect of carbon monoxide on asthma hospitalisation was found in boys, and sulphur dioxide showed significant effects of prolonged exposure in girls. Nitrogen dioxide was positively associated with asthma admissions in both sexes. The lag time for certain gaseous pollutant effects seemed to be shorter in boys (around two to three days for carbon monoxide and nitrogen dioxide), as compared with girls (about six to seven days for sulphur dioxide and nitrogen dioxide). The effects of gaseous pollutants on asthma hospitalisation remained after adjustment of particulate matter. The data showed no association between ozone and asthma hospitalisation in children.

CONCLUSIONS

The study showed positive relations between gaseous pollutants (carbon monoxide, sulphur dioxide, and nitrogen dioxide) at comparatively low levels and asthma hospitalisation in children, using bi-directional case-crossover analyses. Though, the effects of certain specific gaseous pollutants were found to vary in boys and girls.

Sex-related interactive effect of smoking and household pets on asthma incidence

The authors examined the interactive effect of smoking and pets at home on the incidence of asthma and the difference between sexes. The longitudinal data from the first two cycles of the National Population Health Survey, conducted in Canada, were used. A total of 12,636 subjects who reported no asthma at baseline were included in the analysis. The 2-yr cumulative incidence of asthma was higher in females than in males. Female sex and household pets demonstrated a significant interaction in the development of asthma. After adjustment for age, immigration and history of allergy, the odds ratio for smoking in relation to the asthma incidence was 2.50 (95% confidence interval: 1.24-5.05) for females who had pets at home and close to unity for those who had no pets. The incidence of asthma was not associated with smoking status and household pets in males. These results indicate that smoking, having pets at home and other environmental factors can partly explain asthma morbidity among female Canadians.

Interaction between two carcinogens in the two-stage clonal expansion model of carcinogenesis

BACKGROUND

Exposure to two or more carcinogens may result in interactive effects in which the joint effect may be greater or less than that expected to arise as the sum of the effects of the two agents alone. In this article, we investigate the joint effects of exposure to two carcinogens within the context of the two-stage clonal expansion model of carcinogenesis.

METHODS

Different measures of interaction are considered based on the notions of response and dose additivity, and an index of synergy S due to Thomas (1982) used to broadly characterise the effects of joint exposure.

RESULTS

Interactive effects based on the index S were found to be qualitatively similar, regardless of whether cancer risk was defined in terms of age-specific relative risk, or the cumulative probability of cancer occurrence at the same age. For joint exposure to two initiators or to two completers (affecting the first or second mutation rate in the two-mutation model, respectively), S assumed values near zero, reflecting an additive relative-risk relationship. For joint exposure to two promoters (which increase the rate of proliferation of initiated cells that have sustained the first mutation), the relative-risk relationship was found to range from supramultiplicative (S > 1) in younger age groups, to subadditive (S < 0) in older ages. Other combinations of carcinogens involving promotion also displayed a broad range of interaction effects.

CONCLUSIONS

These results differ markedly from those reported previously by Kodell et al. (1991) for an approximate form of the two-stage model, which predicts much higher values of the index of synergy S than the exact form of the model when promotion is involved.

The spatial association between community air pollution and mortality: a new method of analyzing correlated geographic cohort data

We present a new statistical model for linking spatial variation in ambient air pollution to mortality. The model incorporates risk factors measured at the individual level, such as smoking, and at the spatial level, such as air pollution. We demonstrate that the spatial autocorrelation in community mortality rates, an indication of not fully characterizing potentially confounding risk factors to the air pollution-mortality association, can be accounted for through the inclusion of location in the model assessing the effects of air pollution on mortality. Our methods are illustrated with an analysis of the American Cancer Society cohort to determine whether all cause mortality is associated with concentrations of sulfate particles. The relative risk associated with a 4.2 microg/m(3) interquartile range of sulfate distribution for all causes of death was 1.051 (95% confidence interval 1.036-1.066) based on the Cox proportional hazards survival model, assuming subjects were statistically independent. Inclusion of community-based random effects yielded a relative risk of 1.055 (1.033, 1.077), which represented a doubling in the residual variance compared to that estimated by the Cox model. Residuals from the random-effects model displayed strong evidence of spatial autocorrelation (p = 0.0052). Further inclusion of a location surface reduced the sulfate relative risk and the evidence for autocorrelation as the complexity of the location surface increased, with a range in relative risks of 1.055-1.035. We conclude that these data display both extravariation and spatial autocorrelation, characteristics not captured by the Cox survival model. Failure to account for extravariation and spatial autocorrelation can lead to an understatement of the uncertainty of the air pollution association with mortality.

The spatial association between community air pollution and mortality: a new method of analyzing correlated geographic cohort data

We present a new statistical model for linking spatial variation in ambient air pollution to mortality. The model incorporates risk factors measured at the individual level, such as smoking, and at the spatial level, such as air pollution. We demonstrate that the spatial autocorrelation in community mortality rates, an indication of not fully characterizing potentially confounding risk factors to the air pollution-mortality association, can be accounted for through the inclusion of location in the model assessing the effects of air pollution on mortality. Our methods are illustrated with an analysis of the American Cancer Society cohort to determine whether all cause mortality is associated with concentrations of sulfate particles. The relative risk associated with a 4.2 microg/m(3) interquartile range of sulfate distribution for all causes of death was 1.051 (95% confidence interval 1.036-1.066) based on the Cox proportional hazards survival model, assuming subjects were statistically independent. Inclusion of community-based random effects yielded a relative risk of 1.055 (1.033, 1.077), which represented a doubling in the residual variance compared to that estimated by the Cox model. Residuals from the random-effects model displayed strong evidence of spatial autocorrelation (p = 0.0052). Further inclusion of a location surface reduced the sulfate relative risk and the evidence for autocorrelation as the complexity of the location surface increased, with a range in relative risks of 1.055-1.035. We conclude that these data display both extravariation and spatial autocorrelation, characteristics not captured by the Cox survival model. Failure to account for extravariation and spatial autocorrelation can lead to an understatement of the uncertainty of the air pollution association with mortality.

Cohort cancer incidence among pulp and paper mill workers in British Columbia

OBJECTIVES

A study was conducted to investigate cancer risks in a cohort of pulp and paper workers.

METHODS

All male workers with > or =1 years of employment in 14 pulp and paper mills in 1950-1992 were studied. Standardized incidence ratios (SIR) were used to compare the cancer incidence of the cohort with that of the Canadian male population. Record linkage with the National Cancer Registry was performed using the generalized iterative record linkage method.

RESULTS

Altogether 1756 cancer cases were observed in the entire cohort. For > or =15 years of work, the entire cohort had significantly increased SIR values for pleural and prostate cancer and skin melanoma; there was also a significantly increased risk for skin melanoma among workers in the kraft process only, rectal cancer among workers in the sulfite process only, and stomach and prostate cancer and all leukemias combined among workers in both the kraft and sulfite processes. A separate analysis comparing workers in pulping and papermaking with those in the pulping process only did not reveal any difference in cancer risk and hence did not modify the results. The SIR values for skin melanoma were not significantly increased in a comparison using the British Columbia male population. Nine of 10 pleural cancers were mesotheliomas, which likely reflect past asbestos exposure.

CONCLUSIONS

The results suggest that long-term work in the pulp and paper industry is associated with excess risks of prostate and stomach cancers and all leukemias for work in both kraft and sulfite processes and of rectal cancer for work in the sulfite process only.

A case-control study of toenail selenium and cancer of the breast, colon, and prostate

To study the possible role of dietary and supplementary selenium intake in the etiology of cancer, we carried out a case-control study of breast, colon, and prostate cancer in Montreal between 1989 and 1993. In this study, we were able to interview a total of 1,048 incidence cases of colon (402), breast (414) and prostate (232) cancer subjects and 688 population-based controls matched for age and gender. Of these, a total of 501 cancer cases and 202 controls produced toenail samples for their selenium concentrations, which were determined by neutron activation analysis. We found no association between toenail selenium and breast cancer (odds ratio [OR], 0.72; 95% confidence interval [CI], 0.4-1.31) or prostate cancer (OR, 1.14; 95% CI, 0.46-2.83), though we did observe a statistically significant inverse association between toenail selenium level and the risk of colon cancer for both genders combined (OR, 0.42; 95% CI, 0.19-0.93; P = .009) and for female subjects (P = .050). We also found that nonsmoker case and control subjects had higher selenium in their toenail samples. This could be due either to the nature of tobacco, which reduces selenium absorption, or to smokers' consumption of certain foods containing less selenium. Further epidemiologic studies are required to clarify the role of selenium in the etiology of certain cancers.

Asthma and the risk of hospitalization in Canada : the role of socioeconomic and demographic factors

OBJECTIVE

Asthma is an important determinant of hospitalization. The study aims to examine the modifying effects of demographic and socioeconomic factors on the relationship between asthma and the overall number of hospitalizations.

METHODS

We examined the data on 17,601 Canadians who were > or = 12 years of age to explore the combined effects of asthma and other factors on hospitalization within the context of a publicly funded health-care system. Asthma was determined by an affirmative response to the question: "Do you have asthma diagnosed by a health professional?" The subjects also were asked whether they had been an overnight patient in a hospital during the past 12 months.

RESULTS

Asthma as a risk factor explained 3.7% of all hospitalizations of men and 2.4% of all hospitalizations of women. Overall, hospitalization was positively associated with female gender, old age, and low household income. The odds ratio for asthma as a risk factor for overall hospitalization (ie, hospitalization for any reason and all causes, not only for asthma) was greater for younger men than for older men, for less-educated women than for well-educated women, and for men with middle or high incomes than for men with low incomes.

CONCLUSIONS

These results suggest that demographic and socioeconomic factors play a role in the relationship between asthma and the overall number of hospitalizations, with certain population subgroups being at greater risk of hospitalization in relation to asthma.

Association between ozone and hospitalization for acute respiratory diseases in children less than 2 years of age

To clarify the health effects of ozone exposure in young children, the authors studied the association between air pollution and hospital admissions for acute respiratory problems in children less than 2 years of age during the 15-year period from 1980 to 1994 in Toronto, Canada. The daily time series of admissions was adjusted for the influences of day of the week, season, and weather. A 35% (95% confidence interval: 19%, 52%) increase in the daily hospitalization rate for respiratory problems was associated with a 5-day moving average of the daily 1-hour maximum ozone concentration of 45 parts per billion, the May-August average value. The ozone effect persisted after adjustment for other ambient air pollutants or weather variables. Ozone was not associated with hospital admissions during the September-April period. Ambient ozone levels in the summertime should be considered a risk factor for respiratory problems in children less than 2 years of age.

First analysis of cancer incidence and occupational radiation exposure based on the National Dose Registry of Canada

A cohort study was conducted to investigate the relation between cancer incidence and occupational exposure to ionizing radiation. Records containing dose information from 1951 to 1988 for 191,333 persons were extracted from the National Dose Registry of Canada. The records were linked to the Canadian Cancer Data Base, with incidence data from 1969 to 1988. Standardized incidence ratios were calculated using Canadian cancer incidence rates stratified by age, sex, and calendar year. Excess relative risks were obtained from internally based dose-response analyses. The following significant results were found for males and females combined: a deficit in the standardized incidence ratio for all cancers combined; elevated standardized incidence ratios for thyroid cancer and melanoma; and elevated excess relative risks for rectum, leukemia, lung, all cancers combined, all except lung, and all except leukemia. For males, cancers of the colon, pancreas, and testis also showed significantly elevated excess relative risks. The specific cancer types listed above have been implicated in previous studies on occupational exposure to ionizing radiation, except for testis, colon, and melanoma, while the findings on thyroid cancer from previous studies are inconclusive. The thyroid standardized incidence ratios in this study are highly significant, but further investigation is needed to assess the possibility of association with occupational radiation exposure.

Leisure-time energy expenditure in asthmatics and non-asthmatics

An association between obesity and asthma has been documented previously, but the nature of this relationship remains unknown. This study aimed to determine if asthma is associated with a sedentary lifestyle which may explain this association. The energy expenditure (EE) on leisure activities was examined in 16,813 subjects, of at least 12 years of age, who participated in the Canadian National Population Health Survey 1994-95. Energy expenditure was calculated by multiplying the duration of leisure-time physical activity by its estimated metabolic energy cost. Asthma was considered to be present if an affirmative response was given to the question, 'Do you have asthma diagnosed by a health professional?' The average EE (+/- standard error) in males was 2.47 (+/- 0.11) kcal kg(-1) day(-1) for asthmatics and 1.98 (+/- 0.03) kcal kg(-1) day(-1) for non-asthmatics. The corresponding average EEs in females were 1.77 (+/- 0.08) and 1.54 (+/- 0.02) kcal kg(-1) day(-1) for asthmatics and non-asthmatics, respectively. The mean EE values decreased with increasing age. A significant interaction between asthma and age was noted with respect to EE; asthmatics tended to have higher mean EE values than non-asthmatics among younger subjects, and lower mean EE values in older subjects. This effect was more pronounced in females than in males. It was concluded that asthmatics were not consistently inactive compared with non-asthmatics. Leisure-time physical activity cannot explain the positive association between obesity and asthma.