Stephen J. Gilbert

Exposure-response analysis of risk of respiratory disease associated with occupational exposure to chrysotile asbestos.

OBJECTIVES: To evaluate alternative models and estimate risk of mortality from lung cancer and asbestosis after occupational exposure to chrysotile asbestos. METHODS: Data were used from a recent update of a cohort mortality study of workers in a South Carolina textile factory. Alternative exposure-response models were evaluated with Poisson regression. A model designed to evaluate evidence of a threshold response was also fitted. Lifetime risks of lung cancer and asbestosis were estimated with an actuarial approach that accounts for competing causes of death. RESULTS: A highly significant exposure-response relation was found for both lung cancer and asbestosis. The exposure-response relation for lung cancer seemed to be linear on a multiplicative scale, which is consistent with previous analyses of lung cancer and exposure to asbestos. In contrast, the exposure-response relation for asbestosis seemed to be nonlinear on a multiplicative scale in this analysis. There was no significant evidence for a threshold in models of either the lung cancer or asbestosis. The excess lifetime risk for white men exposed for 45 years at the recently revised OSHA standard of 0.1 fibre/ml was predicted to be about 5/1000 for lung cancer, and 2/1000 for asbestosis. CONCLUSIONS: This study confirms the findings from previous investigations of a strong exposure-response relation between exposure to chrysotile asbestos and mortality from lung cancer, and asbestosis. The risk estimates for lung cancer derived from this analysis are higher than those derived from other populations exposed to chrysotile asbestos. Possible reasons for this discrepancy are discussed.

A Monte Carlo Maximum Likelihood Method for Estimating Uncertainty Arising from Shared Errors in Exposures in Epidemiological Studies of Nuclear Workers

Abstract Stayner, L., Vrijheid, M., Cardis, E., Stram, D. O., Deltour, I., Gilbert, S. J. and Howe, G. A Monte Carlo Maximum Likelihood Method for Estimating Uncertainty Arising from Shared Errors in Exposures in Epidemiological Studies of Nuclear Workers. Radiat. Res. 168, 757–763 (2007). Errors in the estimation of exposures or doses are a major source of uncertainty in epidemiological studies of cancer among nuclear workers. This paper presents a Monte Carlo maximum likelihood method that can be used for estimating a confidence interval that reflects both statistical sampling error and uncertainty in the measurement of exposures. The method is illustrated by application to an analysis of all cancer (excluding leukemia) mortality in a study of nuclear workers at the Oak Ridge National Laboratory (ORNL). Monte Carlo methods were used to generate 10,000 data sets with a simulated corrected dose estimate for each member of the cohort based on the estimated distribution of errors in doses. A Cox proportional hazards model was applied to each of these simulated data sets. A partial likelihood, averaged over all of the simulations, was generated; the central risk estimate and confidence interval were estimated from this partial likelihood. The conventional unsimulated analysis of the ORNL study yielded an excess relative risk (ERR) of 5.38 per Sv (90% confidence interval 0.54–12.58). The Monte Carlo maximum likelihood method yielded a slightly lower ERR (4.82 per Sv) and wider confidence interval (0.41–13.31).

Sources of uncertainty in dose-response modeling of epidemiological data for cancer risk assessment

Abstract: Epidemiologic data is increasingly being used for dose‐response analysis in risk assessment. The Environmental Protection Agency (EPA) and other U.S. agencies have expressed a preference for using epidemiologic data rather than toxicologic data when possible. However, there are a number of important sources of uncertainty in using epidemiologic data for this purpose that need to be clearly recognized and, when possible, quantified. This paper presents a critical review of the major sources of uncertainty in the use of epidemiologic data for cancer risk assessment. These may include: (1) study design issues such as potential confounding and other biases, inadequate sample size, and followup, (2) the choice of the data set, (3) specification of the dose‐response model, (4) estimation of exposure and dose, and (5) unrecognized variability in susceptibility. Examples from risk assessments for cadmium, asbestos, and diesel exhaust are used to illustrate the potential magnitude of some of these sources of uncertainty. It is shown that the overall uncertainty from these various sources combined may often result in highly uncertain risk estimates from dose‐response modeling of epidemiologic data. For this reason, we believe it is best to present a range of possible risk estimates, which, to the extent possible, reflects the variability and uncertainty inherent in the dose‐response evaluation of epidemiologic data.

Years of Potential Life Lost Due to Occupational Fatal Injury in the United States

Fatal injury surveillance data coupled with life expectancy data may be used to assess the impact of occupational fatal injuries on years of potential life lost (YPLL). We compare three definitions of YPLL and trends over time in YPLL. Two definitions determine YPLL as expected life lost to fixed life expectancies of 65 or 85 years. The third definition uses actuarial adjustments of life expectancy given survival to a given age stratified by gender and race. Fatalities from the National Traumatic Occupational Fatality (NTOF) database are used to illustrate the three definitions of YPLL. The three YPLL measures were similar in magnitude and direction of the trend in YPLL over 1980-1992. Proper interpretation of these trends can only be made in conjunction with other measures (e.g., rates). Almost all YPLL trends are declining, implying that over time fatal injuries are shifting to older workers. The exception is the increasing trend in YPLL for the retail trade industry, injury rates have also been increas...

Response to “Comments on ‘A re-examination of risk estimates from the NIOSH Occupational Noise and Hearing Survey’ ” [J. Acoust. Soc. Am. 103, 2734 (1998)]

Concern is raised by Dobie [J. Acoust. Soc. Am. 103, 2734 (1998)] regarding a recent analysis [, J. Acoust. Soc. Am. 101, 950–963 (1997)] of the NIOSH Occupational Noise and Hearing Survey data. Specifically, issues are raised concerning (1) definition of hearing handicap, (2) the use of frequency-specific articulation index (AI) weights applied to the binaural pure-tone average of 1, 2, 3, and 4 kHz, and (3) conclusions regarding significant excess risk based on this definition. We have reviewed the development of the definitions of hearing handicap and provide additional support for the use of a hearing handicap definition based on the binaural pure-tone average of 1, 2, 3, and 4 kHz and the weighting of specific frequencies. Furthermore, our definition of noise-induced hearing handicap is similar to one of several proposed by the International Standards Organization (Reference 1999Reference 1990) and the American National Standards Institute (Reference 3Reference 44Reference 1996). Additional analyses ...

Issues in longitudinal analysis of hearing conservation data bases

Several hearing conservation programs (n=15) from a cross‐section of U.S. industries were examined to test whether there are greater rates of change in hearing levels in relation to increasing cumulative noise exposure during the period of audiometric follow‐up. The analysis also examined whether rates of hearing loss were lower among workers who wear hearing protection relative to workers who do not. These 15 audiometric databases had a total of 15 794 workers with 62 095 audiograms. Differences in a linear combination of 1–4 kHz biaural hearing levels of consecutive audiometric tests were examined using mixed effects models to account for correlated, repeated observations on individuals. These differences were modeled as a function of cumulative noise, gender, race, hearing protection use, age, learning effects, and baseline hearing level (measured after 14 h of quiet time). Analyses were conducted for the combined and individual databases. The analysis indicates there is significant heterogeneity among...