The association between ambient carbon monoxide levels and daily mortality in Toronto, Canada

The role of ambient levels of carbon monoxide (CO) in the exacerbation of heart problems in individuals with both cardiac and other diseases was examined by comparing daily variations in CO levels and daily fluctuations in nonaccidental mortality in metropolitan Toronto for the 15-year period 1980-1994. After adjusting the mortality time series for day-of-the-week effects, nonparametic smoothed functions of day of study and weather variables, statistically significant positive associations were observed between daily fluctuations in mortality and ambient levels of carbon monoxide, nitrogen dioxide, sulfur dioxide, coefficient of haze, total suspended particulate matter, sulfates, and estimated PM2.5 and PM10. However, the effects of this complex mixture of air pollutants could be almost completely explained by the levels of CO and total suspended particulates (TSP). Of the 40 daily nonaccidental deaths in metropolitan Toronto, 4.7% (95% confidence interval of 3.4%-6.1%) could be attributable to CO while TSP contributed an additional 1.0% (95% confidence interval of 0.2-1.9%), based on changes in CO and TSP equivalent to their average concentrations. Statistically significant positive associations were observed between CO and mortality in all seasons, age, and disease groupings examined. Carbon monoxide should be considered as a potential public health risk to urban populations at current ambient exposure levels.

Dental amalgam and multiple sclerosis: a case-control study in Montreal, Canada

BACKGROUND

The aetiology of multiple sclerosis (MS) remains poorly understood. Dental amalgams containing mercury have recently been suggested as a possible risk factor for MS.

METHODS

In a case-control study conducted between 1991 and 1994, we interviewed a total of 143 MS patients and 128 controls, to obtain information on socio-demographic characteristics and the number of dental amalgams and the time since installation based on dentists' records.

RESULTS

Neither the number nor the duration of exposure to amalgams supported an increased risk of MS. After adjustment for age, sex, smoking, and education those who had more than 15 fillings had an odds ratio (OR) of 2.57 (95% CI: 0.78-8.54) compared to those who had none; for individuals whose first amalgam was inserted more than 15 years prior to the study, we found an OR of 1.34 (95% CI: 0.38-4.72).

CONCLUSIONS

Although a suggestive elevated risk was found for those individuals with a large number of dental amalgams, and for a long period of time, the difference between cases and controls was not statistically significant.

Associations between several sites of cancer and occupational exposure to benzene, toluene, xylene, and styrene: results of a case-control study in Montreal

BACKGROUND

Except for the leukemogenic effects of benzene, there is inadequate or sparse evidence on the carcinogenicity of the most common monocyclic aromatic hydrocarbons. The purpose of this study was to generate hypotheses on associations between exposure to benzene, toluene, xylene, and styrene and various common types of cancer.

METHODS

In the context of a population-based case-control study carried out in Montreal, 3,730 cancer patients (15 types of cancers, not including leukemia) and 533 population controls were interviewed, and their job histories were translated by a team of experts into occupational exposures, including benzene, toluene, xylene, and styrene. In the present analysis, exposure to these substances was compared between each case series and a control group pooling selected cancer patients and population controls, using logistic regression analysis.

RESULTS

Exposure levels were low for most exposed subjects, and there was a high correlation between exposure to benzene, toluene and xylene. For most sites of cancer there was no evidence of excess risk due to these substances. However, limited evidence of increased risk was found for the following associations: esophagus-toluene, colon-xylene, rectum-toluene, rectum-xylene and rectum-styrene.

CONCLUSIONS

These latter observations warrant further investigation.

A comparison of methods for estimating the benchmark dose based on overdispersed data from developmental toxicity studies

Developmental anomalies resulting from prenatal toxicity can be manifested in terms of both malformations among surviving offspring and prenatal death. Although these two endpoints have traditionally been analyzed separately in the assessment of risk, multivariate methods of risk characterization have recently been proposed. We examined this and other issues in developmental toxicity risk assessment by evaluating the accuracy and precision of estimates of the effective dose (ED05) and the benchmark dose (BMD05) using computer simulation. Our results indicated that different variance structures (Dirichlet-trinomial and generalized linear model) used to characterize overdispersion yielded comparable results when fitting joint dose response models based on generalized estimating equations. (The choice of variance structure in separate modeling was also not critical.) However, using the Rao-Scott transformation to eliminate overdispersion tended to produce estimates of the ED05 with reduced bias and mean squared error. Because joint modeling ensures that the ED05 for overall toxicity (based on both malformations and prenatal death) is always less than the ED05 for either malformations or prenatal death, joint modeling is preferred to separate modeling for risk assessment purposes.

Estimates of the number of liver carcinogens in bioassays conducted by the National Toxicology Program

Estimates were made of the numbers of liver carcinogens in 390 long-term bioassays conducted by the National Toxicology Program (NTP). These estimates were obtained from examination of the global pattern of p-values obtained from statistical tests applied to individual bioassays. Representative estimates of the number of liver carcinogens (90% confidence interval in parentheses) obtained in our analysis compared to NTP's determination are as follows: female rats--49 (23, 76), NTP = 30; male rats--88 (59, 116), NTP = 35; female mice--131 (105, 157), NTP = 81; male mice--100 (73, 126), NTP = 61; overall--166 (135, 197), NTP = 108. The estimator from which these estimates were obtained is biased low by an unknown amount. Consequently, this study provides persuasive evidence of the existence of more rodent liver carcinogens than were identified by the NTP.

Statistical analysis of lacZ mutant frequency data from MutaMouse mutagenicity assays

Transgenic mouse assays have provided an unprecedented opportunity to study mutagenesis in diverse rodent tissues. In this article data from MutaMouse mutagenicity assays based on the Escherichia coli gene lacZ were analyzed systematically using liver and bone marrow as potential target tissues. Sources of variation, including plates (within packaging reactions), packaging reactions (within animals) and animals, were evaluated for extra-binomial variation. Although hardly any evidence of overdispersion was detected at the plate level, limited evidence of extra-binomial variation was observed at the packaging reaction level. There was, however, much stronger evidence of overdispersion at the animal level. Statistical tests for increasing trend in mutant frequency with increasing dose were also performed at the animal level. A significant increasing trend following exposure to N-nitrosodibenzylamine was detected in liver but not in bone marrow. A logistical model was used to further describe the dose-response relationship observed in N-nitrosodibenzylamine-treated liver tissue.

Adjusting for temporal variation in the analysis of parallel time series of health and environmental variables

Time series of daily administrative cardio-respiratory health and environmental information have been extensively used to assess the potential public health impact of ambient air pollution. Both series are subject to strong but unrelated temporal cycles. These cycles must be removed from the time series prior to examining the role air pollution plays in exacerbating cardio-respiratory disease. In this paper, we examine a number of methods of temporal filtering that have been proposed to eliminate such temporal effects. The techniques are illustrated by linking the number of daily admissions to hospital for respiratory diseases in Toronto, Canada for the 11 year period 1981 to 1991 with daily concentrations of ambient ozone. The ozone-hospitalization relationship was found to be highly sensitive to the length of temporal cycle removed from the admission time series, and to day of the week effects, ranging from a relative risk of 0.874 if long wave cycles were not removed at all to 1.020 for models which removed at least cycles greater than or equal to one month based on the interquartile pollutant range. The specific statistical method of adjustment was not a critical factor. The association was not as sensitive to removal of cycles less than one month, except that negative autocorrelation increased for series in which cycles of one week or less were removed. We recommend three criteria in selecting the degree of smoothing in the outcome: removal of temporal cycles, minimizing autocorrelation and optimizing goodness of fit. The association between ambient ozone levels and hospital admissions for respiratory diseases was also sensitive to the season of examination, with weaker associations observed outside the summer months.

Uncertainty and variability analysis in multiplicative risk models

Currently, there is a trend away from the use of single (often conservative) estimates of risk to summarize the results of risk analyses in favor of stochastic methods which provide a more complete characterization of risk. The use of such stochastic methods leads to a distribution of possible values of risk, taking into account both uncertainty and variability in all of the factors affecting risk. In this article, we propose a general framework for the analysis of uncertainty and variability for use in the commonly encountered case of multireplicative risk models, in which risk may be expressed as a product of two or more risk factors. Our analytical methods facilitate the evaluation of overall uncertainty and variability in risk assessment, as well as the contributions of individual risk factors to both uncertainty and variability which is cumbersome using Monte Carlo methods. The use of these methods is illustrated in the analysis of potential cancer risks due to the ingestion of radon in drinking water.

Use of generalized linear mixed models in analyzing mutant frequency data from the transgenic mouse assay

The transgenic mouse assay is now widely used for the study of mutagenesis in diverse rodent tissues and to test chemicals for genotoxic potential. This kind of assay generally involves nested observations at several levels of sampling, e.g., animals, packaging reactions, and plates. Due to the common origin, the mutant frequency (MF) in tissues from the same animal are likely to be positively correlated, inducing extra variation relative to the common binomial variation. In this article, a generalized linear mixed model is used to analyze the overdispersed binomial data on mutant frequency from the transgenic mouse assay, with a random effect for each level of the sampling hierarchy. This is a comprehensive framework within which different sources of variation in the data can be evaluated in nested factorial experiments and treatment effects can be assessed simultaneously. It avoids the current practice of repeated testing for excess binomial variability at each level of the sampling hierarchy and aggregating data up the levels, but fits the data with one single model. Parameters associated with the fixed effects, particularly dose, and the variance components for the random effects (e.g., animals, packages, and plates) can be estimated and tested for significance. Data previously reported in the literature involving the lacl gene from the Big Blue mouse are used to illustrate the proposed method.

Cohort mortality study of pulp and paper mill workers in British Columbia, Canada

The authors studied a cohort of 30,157 male pulp and paper workers in British Columbia, Canada. Of these, 20,373 worked in kraft mills only, 5,249 in sulfite mills only, and 4,535 in both kraft and sulfite mills. All workers with at least 1 year of employment on January 1, 1950, or thereafter until December 31, 1992, were studied. Standardized mortality ratios (SMRs) were used to compare the mortality rates of the cohort with those of the Canadian male population. Ninety percent confidence intervals (CIs) for the SMRs were obtained. Cancer risks significantly associated with work duration and time from first employment of 15 years or more were observed: 1) total cohort: pleura (SMR = 3.61, 90% CI 1.42-7.58); kidney (SMR = 1.69, 90% CI 1.13-2.43); brain (SMR = 1.51, 90% CI 1.03-2.16); 2) workers in kraft mills only: kidney (SMR = 1.92, 90% CI 1.04-3.26); 3) workers in sulfite mills only: Hodgkin's disease (SMR = 4.79, 90% CI 1.29-12.37); 4) workers ever employed in both kraft and sulfite mills: esophagus (SMR = 1.91, 90% CI 1.00-3.33). These malignancies have been associated with the following known or suspected carcinogens to which pulp and paper workers may have been exposed: asbestos (pleura), biocides (kidney), formaldehyde (kidney, brain, Hodgkin's disease), hypochlorite (esophagus). A nested case-control study with detailed exposure assessment is under way to help determine whether excess risks for specific cancers reflect exposure among subsets of workers.

The role of particulate size and chemistry in the association between summertime ambient air pollution and hospitalization for cardiorespiratory diseases

In order to address the role that the ambient air pollution mix, comprised of gaseous pollutants and various physical and chemical measures of particulate matter, plays in exacerbating cardiorespiratory disease, daily measures of fine and coarse particulate mass, aerosol chemistry (sulfates and acidity), and gaseous pollution (ozone, nitrogen dioxide, sulfur dioxide, and carbon monoxide) were collected in Toronto, Ontario, Canada, in the summers of 1992, 1993, and 1994. These time series were then compared with concurrent data on the number of daily admissions to hospitals for either cardiac diseases (ischemic heart disease, heart failure, and dysthymias) or respiratory diseases (tracheobronchitis, chronic obstructive long disease, asthma, and pneumonia). After adjusting the admission time series for long-term temporal trends, seasonal variations, the effects of short-term epidemics, day of the week effects, and ambient temperature and dew point temperature, positive associations were observed for all ambient air pollutants for both respiratory and cardiac diseases. Ozone was least sensitive to adjustment for the gaseous and particulate pollution measures. However, the association between the health outcomes and carbon monoxide, fine and coarse mass, sulfate levels and aerosol acidity could be explained by adjustment for exposure to gaseous pollutants. Increases in ozone, nitrogen dioxide, and sulfur dioxide equivalent to their interquartile ranges corresponded to an 11% and 13% increase in daily hospitalizations for respiratory and cardiac diseases, respectively. The inclusion of any one of the particulate air pollutants in multiple regression models did not increase these percentages. Particle mass and chemistry could not be identified as an independent risk factor for the exacerbation of cardiorespiratory diseases in this study beyond that attributable to climate and gaseous air pollution. We recommend that effects of particulate matter on health be assessed in conjunction with temporally covarying gaseous air pollutants.

Statistical analysis of the lacI transgenic mouse mutagenicity assay

The transgenic mouse assay is now widely used to test chemicals for genotoxic potential. In this article, we consider statistical tests for increasing trend in mutant frequency with increasing dose, along with statistical models that may be used to describe the observed dose-response relationships. The application of these methods is illustrated using data on 2-acetylaminofluorene, di(2-ethylhexyl)phthalate, heptachlor, and sodium phenobarbital. No strong evidence of extra-binomial variation was detected at the plate level, but greater evidence was noted when the data were aggregated to the package or animal level in liver, necessitating the use of statistical methods that allow for overdispersion relative to binomial variation. Clear increase on mutant frequency induced by 2-acetylaminofluorene was detected in both liver and bladder, but no apparent trends were noted with di(2-ethylhexyl)phthalate, heptachlor, and sodium phenobarbital. The exponential model provides a good fit to the observed dose-response relationship in liver, whereas a Weibull model provides a better fit for bladder.

Association between ambient carbon monoxide levels and hospitalizations for congestive heart failure in the elderly in 10 Canadian cities

We examined the role that ambient air pollution plays in exacerbating cardiac disease by relating daily fluctuations in admissions to 134 hospitals for congestive heart failure in the elderly to daily variations in ambient concentrations of carbon monoxide, nitrogen dioxide, sulfur dioxide, ozone, and the coefficient of haze in Canada's 10 largest cities for the 11-year period 1981-1991 inclusive. We adjusted the hospitalization time series for seasonal, subseasonal, and weekly cycles and for hospital usage patterns. The logarithm of the daily high-hour ambient carbon monoxide concentration recorded on the day of admission displayed the strongest and most consistent association with hospitalization rates among the pollutants, after stratifying the time series by month of year and adjusting simultaneously for temperature, dew point, and the other ambient air pollutants. The relative risk for a change from 1 ppm to 3 ppm, the 25th and 75th percentiles of the exposure distribution, was 1.065 (95% confidence interval = 1.028-1.104). The regression coefficients of the other air pollutants were much more sensitive to simultaneous adjustment for either multiple pollutant or weather model specifications.

Association between ozone and hospitalization for respiratory diseases in 16 Canadian cities

The effects of tropospheric ozone on lung function and respiratory symptoms have been well documented at relatively high concentrations. However, previous investigations have failed to establish a clear association between tropospheric ozone and respiratory diseases severe enough to require hospitalization after controlling for climate, and with gaseous and particulate air pollution at the lower concentrations typically observed in Canada today. To determine if low levels of tropospheric ozone contribute to hospitalization for respiratory disease, air pollution data were compared to hospital admissions for 16 cities across Canada representing 12.6 million people. During the 3927-day period from April 1, 1981, to December 31, 1991, there were 720,519 admissions for which the principle diagnosis was a respiratory disease. After controlling for sulfur dioxide, nitrogen dioxide, carbon monoxide, soiling index, and dew point temperature, the daily high hour concentration of ozone recorded 1 day previous to the date of admission was positively associated with respiratory admissions in the April to December period but not in the winter months. The relative risk for a 30 ppb increase in ozone varied from 1.043 (P < 0.0001) to 1.024 (P = 0.0258) depending on the selection of covariates in the regression model and subset of cities examined. The association between ozone and respiratory hospitalizations varied among cities, with relative risks ranging from 1.000 to 1.088 after simultaneous covariate adjustment. Particulate matter and carbon monoxide were also positively associated with respiratory hospitalizations. These results suggest that ambient air pollution at the relatively low concentrations observed in this study, including tropospheric ozone, is associated with excess admissions to hospital for respiratory diseases in populations experiencing diverse climates and air pollution profiles.

An exact representation for the generating function for the Moolgavkar-Venzon-Knudson two-stage model of carcinogenesis with stochastic stem cell growth

The two-stage clonal expansion model of carcinogenesis provides a convenient biologically based framework for the quantitative description of carcinogenesis data. Under this stochastic model, a cancer cell arises following the occurrence of two critical mutations in a normal stem cell. Both normal cells and initiated cells that have sustained the first mutation undergo birth-and-death processes responsible for tissue growth. In this article, a new expression for the probability generating function (pgf) for the two-stage model of carcinogenesis is derived. This characterization is obtained by solving a partial differential equation (pde) satisfied by the pgf derived from the corresponding Kolmogorov forward equation. This pde can be reduced to the hypergeometric differential equation of Gauss, which leads to a closed-form expression for the pgf requiring only the evaluation of hypergeometric functions. This result facilitates computation of the exact hazard function for the two-stage model. Several approximations that are simpler to compute are also given. Numerical examples are provided to illustrate the accuracy of these approximations.

Uncertainty, variability, and sensitivity analysis in physiological pharmacokinetic models

Physiologically based pharmacokinetic (PBPK) models are now commonly used to predict the dose of toxic metabolites of chemical substances reaching target tissues. A typical PBPK model can involve 20 or more physiological, physiochemical, and biochemical parameters, each of which is estimated with some degree of error. In this article, methods for assessing the impact of uncertainty in the parameter values on prediction of tissue dose are proposed, along with methods for identifying those parameters to which predictions of tissue doses are most sensitive. Many of the model parameters are related to body weight, which is assumed to vary in accordance with a doubly truncated normal distribution. The application of the proposed methods is illustrated using a PBPK model for benzene.

The future of mechanistic research in risk assessment: where are we going and can we get there from here?

Quantitative estimates of human health risk are often based on mathematical models fit to experimental or epidemiological data. Recent years have witnessed a trend towards the use of mechanistic models in risk assessment applications. Such models afford a more biologically based interpretation of the data and a firmer scientific basis for extrapolation beyond the conditions under which the original data were obtained. In this article, we review some recent advances in the development of biologically based models for mutagenesis, carcinogenesis and developmental toxicity. Pharmacokinetic and receptor-binding models and their roles in mechanistic risk assessment are also discussed. The future of mechanistic research in risk assessment is contemplated, including the need for more elaborate experiments to obtain the data necessary for mechanistic modeling.

Cancer risk due to occupational exposure to polycyclic aromatic hydrocarbons

Polycyclic aromatic hydrocarbons (PAHs) demonstrate carcinogenic activity in animal models. Although some epidemiologic studies have implicated PAHs as risk factors for human cancer, the evidence reported to date has not been consistent. The purpose of this report is to describe the associations between occupational exposure to PAHs in the workplace and each of 14 types of cancer. A population-based, case-control study was carried out in Montreal to investigate associations between a large variety of environmental and occupational exposures on the one hand, and several types of cancer on the other. A detailed job history was obtained from each subject along with information on a number of potential confounders. Each job history was reviewed by a team of experts, who used this information to construct a corresponding history of occupational exposures. Among the PAH exposures considered were benzo(a)pyrene (B(a)P) and five categories of PAHs defined on the basis of the source material, namely, wood, petroleum, coal, other sources, and any source. Altogether, 3,730 cancer patients and 533 population controls were interviewed and their job exposure histories coded. For each of 14 types of cancer analyzed, three control groups were available: other cancer patients, population controls, and the pooled set of cancer and population controls. The associations between 14 cancer types and 6 PAH exposures were analyzed using logistic regression methods. For most types of cancer evaluated, there was no evidence of excess risk due to PAHs at the levels encountered in the occupations in which PAH exposure has been prevalent in the Montreal area. For a few cancer sites--the esophagus, the pancreas, and the prostate gland--there were suggestions of excess risk; these observations are noteworthy hypotheses for further investigation. For lung cancer, there appeared to be an increased risk due to PAHs among nonsmokers and light smokers, but not among heavy smokers.