NIOSHTIC-2 Publications Search

Statistical problems in occupational safety and health.

Mathew T; Krishnamoorthy K

Atlanta, GA: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, R01-OH-003628, 2005 Feb; :1-9

20026059

The technology for monitoring occupational exposure depends crucially on the statistical analysis of the exposure data, for the purpose of developing appropriate strategies to monitor and improve occupational safety and health: in particular, for assessing health risks, for identifying cost-effective intervention strategies. for developing exposure-response relationships, for evaluating low-cost, easy to use, non-invasive test kits, and a host of other activities that will enhance occupational health. Furthermore, the collection of exposure data involves considerable human and financial resources. Consequently, it is very crucial that the data be used effectively. Needless to say, the use of valid statistical methods for exposure data analysis is critical to this endeavor. The funded project was aimed to develop such statistical methods. The following have been accomplished based on the funded project. Satisfactory procedures have been obtained for analyzing exposure data using the lognormal distribution; in particular, satisfactory confidence intervals and test procedures have been developed. These findings are important because of the relevance and widespread use of the lognormal distribution for exposure data analysis. Furthermore, the newly developed procedures are quite accurate in meeting the required statistical properties in terms of the coverage probability of the confidence interval and type I error probability of the test. For the problem of testing the equivalency of a sampling device to an OSHA standard, a statistically rigorous procedure has been developed. This is a problem that is routinely encountered while developing cheaper or more efficient exposure assessment strategies. A thorough investigation of the computation of tolerance limits has been carried out for a random effects model. This problem is of significance since an upper tolerance limit provides a limit below which most of the exposure measurements are expected to lie. Furthermore, a random effect model is relevant for exposure assessment due to the fact that such a model provides a way to account for the dependency among the exposure group. For such models, appropriate statistical procedures have also been developed for checking if the proportion of workers for whom the mean exposure exceeds the occupational exposure limit is above a threshold. We have provided a solution to the hypothesis testing problem involving the proportion of workers whose mean exposure exceeds the occupational exposure limit. Our solutions are again based on a one-way random effects model for the log transformed shift-long personal exposure measurements, where the random effect in the model represents an effect due to the worker. Numerical results show that our procedures exhibit satisfactory performance regardless of the sample size. A similar procedure is then employed for testing hypotheses concerning the overall mean exposure. Any statistical procedure for analyzing exposure data will be useful only if the procedure is easy to compute and implement. Furthermore, it is also important that the procedure be applicable to small sample sizes, since exposure monitoring could be time consuming and expensive, and large amounts of data are typically unavailable. The funded work was motivated by all of these considerations, and the results we obtained meet these requirements.

Statistical-analysis; Epidemiology; Exposure-assessment; Risk-analysis; Models; Mathematical-models

Thomas Mathew, Department of Mathematics and Statistics, University of Maryland, 1000 Hilltop Circle, Baltimore, MD 21250

20050201

Final Grant Report

mathew@math.umbc.edu

400453

Grant

2005

PB2005-103187

A03

Grant-Number-R01-OH-003628

OEP

Statistical problems in occupational safety and health

MD; LA

University of Maryland, Baltimore