Heterogeneity in sources of exposure variability among groups of workers exposed to inorganic mercury

Many exposure assessment strategies rely on the occupational group as the u nit of analysis in which workers are classified on the basis of job title, location, or on other characteristics related to the workplace or the job. Although statistical methods that combine exposure data collected on worker s from different occupational groups are more efficient, the underlying ass umption that the degree of variation over time and among workers is the sam e for all groups has yet to be fully investigated. Given the utility of dif ferent modeling approaches when assessing exposures, we investigated assump tions of homogeneity of variance within and between workers using both rand om- and mixed-effects models. In our study of four groups of workers expose d to inorganic mercury (Hg) at a chloralkali plant, there was no evidence o f significant heterogeneity in the levels of variation over time or between workers for air Hg levels. For the biological monitoring data, however, ou r findings indicate that groups did not share common levels of variability and that it was not appropriate to pool the data and obtain single estimate s of the within- and between-worker variance components. Classification of job group as a random or fixed effect had no effect on the results and yiel ded the same conclusions when the models were compared. To illustrate effec ts related to the proper specification of a model, the likelihood of exceed ing certain levels (which is a function of the parameters of the underlying distribution of the natural log-transformed exposures) was evaluated using the results obtained from the different models. Although the probability t hat workers' mean exposures exceeded occupational exposure limits for air, urine and blood Hg was generally low (< 10%) for all groups except maintena nce workers, the estimated values sometimes varied depending upon the parti cular model that was applied. Given the growing use of random- and mixed-ef fects models that combine data across occupational groups, additional studi es are warranted to evaluate whether it is reasonable to assume common vari ances and covariances among measurements collected on workers from differen t groups. (C) 2001 British Occupational Hygiene Society. Published by Elsev ier Science Ltd. All rights reserved.