A constrained maximum likelihood approach to evaluate the impact of dose metric on cancer risk assessment: application to ß-chloroprene.

ß-Chloroprene (2-chloro-1,3-butadiene, CD) is used in the manufacture of polychloroprene rubber. Chronic inhalation studies have demonstrated that CD is carcinogenic in B6C3F1 mice and Fischer 344 rats. However, epidemiological studies do not provide compelling evidence for an increased risk of mortality from total cancers of the lung. Differences between the responses observed in animals and humans may be related to differences in toxicokinetics, the metabolism and detoxification of potentially active metabolites, as well as species differences in sensitivity. The purpose of this study was to develop and apply a novel method that combines the results from available physiologically based kinetic (PBK) models for chloroprene with a statistical maximum likelihood approach to test commonality of low-dose risk across species. This method allows for the combined evaluation of human and animal cancer study results to evaluate the difference between predicted risks using both external and internal dose metrics. The method applied to mouse and human CD data supports the hypothesis that a PBK-based metric reconciles the differences in mouse and human low-dose risk estimates and further suggests that, after PBK metric exposure adjustment, humans are equally or less sensitive than mice to low levels of CD exposure.

Lung cancer mortality in UK nickel-cadmium battery workers, 1947-2000.

AIMS: To investigate mortality from lung cancer in nickel-cadmium battery workers in relation to cumulative exposure to cadmium hydroxide. METHODS: The mortality of a cohort of 926 male workers from a factory engaged in the manufacture of nickel-cadmium batteries in the West Midlands of England was investigated for the period 1947-2000. All subjects were first employed at the plant in the period 1947-75 and employed for a minimum period of 12 months. Work histories were available for the period 1947-86; the factory closed down in 1992. Two analytical approaches were used, indirect standardisation and Poisson regression. RESULTS: Based on serial mortality rates for the general population of England and Wales, significantly increased mortality was shown for cancers of the pharynx (observed (Obs) 4, expected (Exp) 0.7, standardised mortality ratio (SMR) 559, p<0.05), non-malignant diseases of the respiratory system (Obs 61, Exp 43.0, SMR 142, p<0.05), and non-malignant diseases of the genitourinary system (Obs 10, Exp 4.1, SMR 243, p<0.05). Non-significantly increased SMRs were shown for lung cancer (Obs 45, Exp 40.7, SMR 111) and cancer of the prostate (Obs 9, Exp 7.5, SMR 116). Estimated cumulative cadmium exposures were not related to risks of lung cancer or risks of chronic obstructive pulmonary diseases, even when exposure histories were lagged first by 10, then by 20 years. CONCLUSIONS: The study findings do not support the hypotheses that cadmium compounds are human lung carcinogens.

Method-induced misclassification for a respirable dust sampled using ISO/ACGIH/CEN criteria.

The single cut measurement of exposure to respirable dust is the accepted method of exposure classification in occupational hygiene. We previously showed that actual pulmonary tissue dose may be substantially different from the dose expected, or the indicated dose, based on measurements using current single cut methodologies. We now examine exposure misclassification of workers based on any single cut respirable dust measurement using the internationally accepted ISO/ACGIH/CEN single cut respirable dust measurement criteria. Hypothetical aerosols with 12 generalized size distributions typical of the method of aerosol generation (condensation, aged condensation, mechanical low energy, mechanical high energy and mixtures thereof) were assumed. Using previously reported models for sampler penetration and pulmonary deposition, Monte Carlo simulations of actual mass dose to pulmonary tissues in comparison to the dose estimate from an ideal respirable aerosol sampler were carried out. Measurement-based indicated doses were used to classify exposures into five exposure categories and these classifications were compared with the 'true' classifications from the dose-based exposure estimates. Misclassification rates were generally severe and were greatest for aerosols with mass median aerodynamic diameter (MMAD) <1 microm (approximately 100%) and MMAD 5-15 microm (65-95%). Misclassification rates were moderate (<20%) only for extremely coarse aerosols of MMAD>15 microm. Misclassification rates for oral and nasal breathing at 750 and 1500 ml tidal volume and 15 breaths/min were similar for each aerosol examined.

A field comparison of the IOM inhalable aerosol sampler and a modified 37-mm cassette.

This research focused on comparing a modified 37-mm (Mod37) sampling cassette with an IOM inhalable dust sampler. Paired IOM and Mod37 breathing-zone air samples were collected for workers engaged in corrosion control maintenance operations on several types of aircraft at several U.S. Air Force bases in the United States. Sampled operations included hand and power sanding, blow-down and wipe-down to remove dust, and spray finishing. The cassettes' interior surfaces were swabbed and the swabs combined with the filters for chromium analysis by NIOSH Method 7300. This approach utilized total chromium as a sensitive surrogate indicator of total aspirated mass. The influences of work location, work type, sample duration, and sampler type on measured concentration were evaluated using analysis of variance techniques. Only work type (process) was found to be a statistically significant predictor of measured concentration. The relationship between IOM- and Mod37-measured values for paired samples was evaluated by work type using linear regression techniques. Linear regressions showed that the modified 37-mm cassette over-samples aerosol by 35 percent compared to the IOM when a wide range of aerosol concentrations and compositions for divergent work tasks in multiple field locations are sampled. Interpretation of these results in light of previous results involving filter-only Mod37 analyses suggests that while the Mod37 has a higher aspiration efficiency than the IOM, substantial Mod37 wall losses result in underestimation of exposure when only the 37-mm filter is analyzed rather than filters plus wall swabs.

The variability of delivered dose of aerosols with the same respirable concentration but different size distributions.

The influences of aerosol size distribution and breath tidal volume on respirable dose estimates were examined for mouth breathing using the ACGIH/ISO/CEN criterion for respirable-equivalent aerosols. Actual tissue doses predicted from a set of pulmonary empirical deposition equations, the Heyder-Rudolf equations, were compared with deposition assumed to occur under the penetration-based respirable dust sampling criterion. Deposition estimate errors ranged from approximately 1/10- to 10-fold, with aerosol mass median aerodynamic equivalent diameter and geometric standard deviation as well as tidal volume each showing a substantial influence under appropriate conditions. These findings demonstrate that reliance on respirable aerosol sampling data obtained with devices performing on a penetration-based sampling criterion may lead to erroneous dose-response relationships in exposure standard development as well as exposure misclassification errors during epidemiological studies. A more reliable dose estimate would be obtained using devices with collection efficiency performance closely matching the alveolar deposition prediction curves of Heyder and Rudolf. We believe that if it is not currently required, the development of a deposition-based aerosol sampling methodology will soon be required for the determination and quantification of inhaled aerosol-induced adverse health effects.

A review of adverse pregnancy outcomes and formaldehyde exposure in human and animal studies.

We examine the potential for reproductive and developmental effects from formaldehyde exposure. Formaldehyde is unlikely to reach the reproductive system in humans in concentrations sufficient to cause damage since it is rapidly metabolized and detoxified upon contact with the respiratory tract. While there are effects seen in in vitro studies or after injection, there is little evidence of reproductive or developmental toxicity in animal studies under exposure levels and routes relevant to humans. Most of the epidemiology studies examined spontaneous abortion and showed some evidence of increased risk (meta-relative risk=1.4, 95% CI 0.9-2.1). We found evidence of reporting biases and publication biases among the epidemiology studies and when these biases were taken into account, we found no evidence of increased risk of spontaneous abortion among workers exposed to formaldehyde (meta-relative risk=0.7, 95% CI 0.5-1.0). The small number of studies on birth defects, low birth weight, and infertility among formaldehyde workers; the limitations in the design of these studies; and the inconsistent findings across these studies make it difficult to draw conclusions from the epidemiology data alone. However, information from experimental studies and studies of metabolism indicate reproductive impacts are unlikely at formaldehyde exposures levels observed in the epidemiology studies.

Use of global positioning system technology to track subject's location during environmental exposure sampling.

Global positioning system (GPS) data recorders were worn by subjects in the Oklahoma Urban Air Toxics Study (OUATS) for automatic logging of their location as they went about their normal daily activities. The location information obtained by the GPS units had an uncertainty of about 10-20 m, which was sufficiently precise to track subjects' movements on trips outside the immediate vicinity of their homes. Due to instrument problems, primarily related to reduced battery life, the units operated for only about 30% of the total monitoring time attempted in 25 trials. The GPS data were compared to time-activity diaries kept by the subjects. In almost all cases, the GPS data confirmed all travel events reported in the subjects' diaries. Additionally, in five out of five trials in which the logging period covered most or all of the subjects' daytime activities, at least one travel event that was not recorded in the diary was detected by GPS. Notwithstanding the limitations of present technology, GPS was found to be a promising means for tracking of research subjects in community-based exposure assessment studies.

A review and meta-analysis of formaldehyde exposure and pancreatic cancer.

BACKGROUND: Most reviews on the carcinogenicity of formaldehyde have focused on cancers of the respiratory tract. Two recent studies have suggested that exposure to formaldehyde may increase the risk for pancreatic cancer. METHODS: We examine 14 epidemiology studies of workers exposed to formaldehyde where pancreatic cancer rates were reported and use meta-analytic techniques to summarize the findings. We also rank formaldehyde exposures for the industries in these studies. RESULTS: We found a small increase of pancreatic cancer risk in the studies overall (meta Relative Risk [mRR] 1.1, 95%CI 1.0-1.3); however, this increased risk was limited to embalmers (mRR 1.3, 95%CI 1.0-1.6) and pathologists and anatomists (mRR 1.3, 95%CI 1.0-1.7). There was no increased risk among industrial workers (mRR 0.9, 95%CI 0.8-1.1) who on average had the highest formaldehyde exposures. CONCLUSIONS: A small increased risk of pancreatic cancer from formaldehyde exposure cannot be ruled out from the studies examined. However, the null findings among industrial workers and the lack of biological plausibility would argue against formaldehyde as a cause. The increased risk of pancreatic cancer among embalmers, pathologists, and anatomists may be due to a diagnostic bias or to occupational exposures other than formaldehyde in these professions.

The reliability of multiple regression and an alternative method for extracting task-specific exposure estimates from time-weighted average data.

The reliability of multiple regression analysis as a method for determining task-specific exposures from multi-task time-weighted average data was evaluated in comparison with the alternative P-screen method. The performances of the two methods were tested using simulated sample data that were calculated as averages over six tasks, where task-specific concentrations drawn randomly from lognormal distributions were weighted by randomly generated task time-weights. Data sets consisted of 20 or 100 simulated samples. The simulated data sets conformed to requirements inherent in the P-screen method that at least one task be absent from each sample and each task be absent from at least one sample. In thousands of Monte Carlo trials under various conditions, the two methods were found to perform equally well when dichotomous task measures (occurrence/ nonoccurrence) were used. Combining the two methods did not improve reliability appreciably, suggesting that the methods are effectively equivalent when dichotomous task measures are used. When task durations were used as the regressors or time-weights, multiple regression was found to be more reliable than P-screen. It is well recognized that incidental or fundamental collinearities between regressors may undermine multiple regression analyses. The P-screen-related restrictions on the task structure of data sets reduces the potential for problems arising from such collinearities. However, the use of multivariate analysis of multiple-task samples will always be an imperfect substitute for single-task sampling.

Theoretical investigation of the interrelationships between stationary and personal sampling in exposure estimation.

In exposure estimation, personal sampling is the method of choice as it is a nearby representative of the contaminant concentration in the breathing zone. Due to the versatility of the stationary sampling in obtaining much higher sensitivity, in its adaptability to telemetering observations, it may also be an attractive sampling method for many circumstances. However, the two sampling methods differ in many theoretically important ways that go beyond the obvious differences. The theoretical investigation of the stationary and personal sampling methods vis-à-vis sampling for exposure estimation shows that the area sampling can be used to represent personal sampling under restricted conditions. Under the restricted conditions, an area of concentration within specified bounds may be determined in relation to a reasonably well-defined source. The extension of the theory to multiple or ill-defined sources pose potential complications that may be intractable through a theoretical analysis. These limitations and restrictions are inherent to the underlying premises of the two methods; therefore they are not amenable to easy correction. Even though these restrictions may suggest only a limited role for area sampling in exposure assessment, the theory shown also suggests areas of further applied and theoretical research to extend the proper use of area sampling in exposure assessment.

Multicomponent error model for mass measurement based size fractionating aerosol samplers.

A mass based size fractionating aerosol sampling device such as an impactor has a number of experimental measurement errors that can affect the size distribution determination. These errors are not necessarily additive, such as weighing errors, multiplicative such as airflow errors, or a power function such as bounce. In general, the cumulative errors are a combination of different relational scales and they are likely to have different functional forms across the full range of measurements. A complete theory of errors must consider a diverse set of functional relationships between mass, flow, size distribution, and other non-linear parameters such as entry losses and bounce to estimate the error bounds for a measured size distribution and aerosol concentration. In addition, aerosol exposure measurements are single sample events. The theoretical multi-compartment error model is an extension of the Rocke and Lorenzato model of measurement errors in analytical chemistry and it includes generalized parameters for all empirically meaningful transformations. Although the general theory is complicated, heuristic reductions can be made to reduce the estimation process to a manageable size. The numerical examples of error analysis of a hypothetical impactor show that the measured distribution related error bound estimation process is not difficult to perform.

Sample size-based indication of normality in lognormally distributed populations.

Occupational and environmental hygiene sampling strategies are usually dictated by factors that limit sample sizes to relatively small numbers. Often, parameters estimated from small sample sizes are then used to make further estimates of the occurrence of extreme events, which are governed by the underlying exposure distribution. We investigated the limitations superimposed by the number of samples in distinguishing an asymmetric (Lognormal) distribution through the rejection of a hypothesized symmetric (Normal) distribution. Sets of 5 to 250 synthetic samples from underlying Lognormal distributions with unit median were generated for 24 separate geometric standard deviations (GSDs), ranging from 1.25 to 7.00. Each simulated combination was repeated in blocks of 200 and each block was repeated tenfold. The synthetic samples were then tested for goodness of fit for Normality by using the Shapiro and Wilk's W Test. Results indicated that the number of samples required to distinguish between Normal and Lognormal distributions was inversely related to GSD. When GSD = 1.25, 169 samples were required for 90 percent distinction at alpha = 0.05. The criteria for success for GSD of 2.00 and 4.00 were 25 and 15 samples, respectively. These results led to the conclusion that the general inability to distinguish an underlying distribution may impose serious difficulties in the estimation of extreme events associated with occupational and environmental hygiene-related sampling.

An investigation of secondary exposure misclassification effects of lifelong occupational history in exposure estimation.

The effects of exposure misclassification on the interpretation of results of occupational epidemiological studies has been widely investigated and reported. Usually, only the direct effects of misclassification have been considered or simple estimates of misclassification rates have been assigned to various types of exposure estimation processes. Lifelong job profile data obtained from a previously published case-control study provided complete or nearly complete job histories of 511 decedents. An analysis of these work histories and the comparison of exposures related to longest-held job to estimated total lifetime exposures suggest that single job-based exposure estimates may lead to significant exposure misclassification rates. In addition, the appearance of shorter duration jobs in a study population occurring predominantly early in the work history may exacerbate problems associated with exposure misclassification. While few specific suggestions emerge from this analysis, the inclusion of extensive recording of the work history of study subjects emerges as a reasonable basis for the investigation and potential reduction of secondary misclassification of exposures in occupational epidemiological studies.

Exposure estimation in four major epidemiologic studies in the acrylonitrile industry.

The reconstruction of worker exposures has been the mainstay of modem industrial epidemiologic studies. In most cases, the researchers are faced with the difficult dilemma created by the scarcity of the exposure measurement data vis-à-vis the demands for refined quantification. The 4 industrial epidemiologic studies of workers exposed to acrylonitrile share many similarities to illustrate strengths and weaknesses of the current exposure reconstruction methodologies. The analysis of the reported exposure reconstruction methods and comparative analysis of some of the results suggest that there is a certain degree of conformity in the results of the exposure reconstruction processes in these 4 studies. At the same time, the same analysis invokes some questions with respect to the interpretation of the exposure reconstruction results.

A case-control study of lung cancer mortality in four rural Arizona smelter towns.

To investigate factors related to lung cancer mortality in four Arizona copper-smelter towns, the authors identified 142 lung cancer cases and 2 matched controls per case from decedent residents during 1979-1990. The authors obtained detailed information on lifetime residential, occupational, and smoking histories via structured telephone interviews with knowledgeable informants. The authors linked estimated historical environmental exposures to smelter emissions (based on atmospheric diffusion modeling of measured sulfur dioxide concentrations) with residential histories to derive individual profiles of residential exposure. The results of this study provided little evidence of a positive association between lung cancer and residential exposure to smelter emissions. Conditional logistic regression analysis revealed a statistically significant positive association between lung cancer and reported employment in copper mines and/or smelters, although specific factors associated with the apparently increased risk among these workers could not be identified in this community-based study.

Airborne fiber concentrations during splitting open and boxing bags of asbestos.

Measured airborne concentrations of asbestos are often unavailable to assess different epidemiologic estimates of past personal exposure levels or concentrations near specific operations involving handling asbestos. The purpose of this study was to assess the potential use of a laboratory study in estimating exposure to asbestos in operations that ceased for many years. The asbestos transfer operations were simulated by splitting and boxing 4.5-kg paper bags of chrysotile asbestos in an enclosed chamber ventilated at 28.2 air changes per hour (ACH). Two recirculation fans achieved chamber air mixing. The airborne concentration of asbestos fibers was determined by sampling air through membrane filters at a rate of 10 l/min and by analysing fibers by phase contrast optical microscopy to determine 15-min average airborne concentrations. Samplers were located at four equidistant locations from the operation. Opening a maximum of seven bags in 15 min was associated with an asbestos-in-air concentration of less than 2 x 10(-6) fibers/m3 (f/m3) in splitting open and boxing and less than 0.85 x 10(-6) f/m3 in boxing of pre-split bags. The measured airborne asbestos concentration exhibited a linear trend with the number of bags opened per 15 min. The empirical results were utilised to model fiber concentrations for various ventilation rates. It was concluded that the distribution of the operation within the workday and the ventilation rate are the two most important variables in the determination of the estimated time-weighted average concentration. Splitting open and boxing 32 bags of asbestos over an 8-h period at a rate of 4 bags/h over an 8-h period results in a calculated time-weighted average exposure of about 1 x 10(-6) f/m3 in a work space with 20 ACH and approximately 7 x 10(-6) f/m3 in a work space with 2 ACH. Splitting open and boxing at a rate of 12 bags/h for 2 h and 45 min yielded calculated concentrations less than one-half of the above mentioned values. These results should serve as a useful benchmark for epidemiologists concerned with the magnitude of exposure in historically dusty industrial tasks.

A case-control study of lung cancer mortality in six Gila Basin, Arizona smelter towns.

To investigate factors related to lung cancer mortality in six Arizona copper smelter towns, we identified 185 lung cancer cases and two matched controls per case from decedent residents during 1979-1990. Detailed information on lifetime residential, occupational, and smoking history was obtained by structured telephone interviews with knowledgeable informants. Interviews were completed for 82% of 183 eligible cases and 88% of the targeted number (366) of controls. Estimated historical environmental exposures to smelter emissions, based on atmospheric diffusion modeling of measured SO2 concentrations, were linked with residential histories to derive individual profiles of residential exposure. Occupational histories were characterized by potential exposure to smelter emissions, asbestos, and ionizing radiation. Conditional logistic regression was used to compare study factors in cases and controls with adjustment for potential confounding factors: gender, Hispanic ethnicity, and smoking. In overall and gender-specific analyses, no statistically significant associations were observed between lung cancer risk and any of the measures of residential exposure to smelter emissions considered (town of residence at time of death, highest level of exposure, and duration or cumulative exposure above background levels), or any of the estimated occupational exposures (definite or potential asbestos, potential ionizing radiation, definite or potential smelter). Among male residents of some, but not all, towns, there was some evidence of a positive association between lung cancer risk and reported copper smelter-related employment (reported as definite), with the highest risk observed for Miami, Arizona. This study provided little evidence of a positive association between lung cancer mortality and residential exposure to smelter emissions. Specific factors associated with the apparent heterogeneity in lung cancer risk across study towns cannot be identified in this community-based study.

An updated meta-analysis of formaldehyde exposure and upper respiratory tract cancers.

In this study, we summarize 47 epidemiologic studies related to formaldehyde exposure and use meta-analytic techniques to assess findings for cancers of the lung, nose/nasal sinuses, and nasopharynx. Our analyses indicate that workers with formaldehyde exposure have essentially null findings for lung cancer and a slight deficit of sinonasal cancer. Nasopharyngeal cancer rates were elevated moderately in a minority of studies. Most studies, however, did not find any nasopharyngeal cancers, and many failed to report their findings. After correcting for underreporting, we found a meta relative risk of 1.0 for cohort studies. Case-control studies had a meta relative risk of 1.3. Our review of the exposure literature indicated that the nasopharyngeal cancer case-control studies represented much lower and less certain exposures than the cohort studies. We conclude that the available studies do not support a causal relation between formaldehyde exposure and nasopharyngeal cancer. This conclusion conflicts with conclusions from two previous meta-analyses, primarily because of our consideration of unreported data.

Mortality among chemical workers in a factory where formaldehyde was used.

OBJECTIVES: An independent and updated historical cohort mortality study was conducted among chemical plant workers to investigate further an association between exposures to formaldehyde and particulates and cancers of the nasopharynx and lung reported in an earlier National Cancer Institute study of the same plant. METHODS: Subjects were 7359 workers who were first employed between 1941 and 1984 in a factory in Wallingford, Connecticut where formaldehyde was used. Vital status was determined on 31 December 1984 for 96% of the cohort and death certificates were obtained for 93% of 1531 known deaths. Exposures of individual workers were estimated quantitatively for formaldehyde, product particulates, and non-product particulates, and qualitatively for pigment. Statistical analyses focused on 6039 white men in 1945-84. Cohort data that could not have been included in the National Cancer Institute study were also analysed separately. RESULTS: Mortality among long term workers (employed > or = 1 y) was generally similar to or more favourable than that of the general population, and there was little evidence of a relation between either rates of lung cancer or standardised mortality ratios (SMRs) and several measures of exposure to formaldehyde, particulates, and pigment. For several causes including lung cancer, death rates among short term workers (employed < 1 y) were significantly increased. Short term workers did not seem to differ from long term workers for the exposures considered. Among all white men, a significant SMR of 550 (local comparison) for nasopharyngeal cancer (NPC) was based on the same four index cases identified in the earlier study of this plant. Only one case of nasopharyngeal cancer had any appreciable exposure to formaldehyde. No new cases of nasopharyngeal cancers were found among the cohort data that could not have been included in the National Cancer Institute study--that is, extended observation time and additional study members. CONCLUSIONS: Among workers employed for at least one year, this study provides little evidence that the risk of lung cancer is associated with exposure to formaldehyde alone or in combination with particulates or pigment. The significant increases in both the rates and SMRs for lung cancer seem to be primarily a phenomenon of short term workers, but the possibility remains that unmeasured occupational or non-occupational factors may have played a part.