O47-3 Using published data from us workplaces to predict historical air and blood lead concentrations for activities related to lead-based paints and cutting and joining metals. (1st September 2016)

• Record Type: Journal Article
• Authors: Locke, Sarah; Deziel, Nicole; Koh, Dong-Hee; Graubard, Barry; Purdue, Mark; Friesen, Melissa
• Is Part Of: Occupational and environmental medicine. Volume 73(2016)Supplement 1
• Page Start: A89
• Page End: A90
• Publication Date: 2016-09-01
• Subjects: Medicine, Industrial -- Periodicals; Environmental health -- Periodicals; 616.980305
• DOI: 10.1136/oemed-2016-103951.240
• Languages: English
• ISSNs: 1351-0711

Abstract

Abstract : Objectives: Historical exposure data reported in the literature are increasingly being used to estimate intensity in population-based studies. To develop lead intensity estimates for a U.S. case-control study, we used meta-regression models to identify predictors of personal air and blood lead concentrations for US workers performing activities related to lead-based paint and cutting or joining metal with heat using published data. Methods: From 69 published papers covering the study years 1962–2005, we extracted personal air and blood lead geometric means (GM), geometric standard deviations (GSD), number of measurements per statistic, and other ancillary exposure variables. Mixed-effects meta-regression models were developed separately for 221 air and 113 blood statistics, with the respective log-transformed GM as the dependent variable. Random intercept was incorporated that weighted each statistic by the inverse of its variance. Variables examined included year, industry, job, sampling duration, lead-based paint removal activities, worst case scenarios, and respirator use. Industry interactions with job and year were also tested. Results: Job, industry, and year were the main predictors of exposure. Temporal trends declined more in the model based on blood versus air concentrations (6.2 vs. 4.6% per year); however, confidence intervals overlapped. Exposure contrasts in the predicted GMs across the 9 jobs and 5 industries were higher in personal air (238- and 8-fold, respectively) vs. blood lead models (3- and 4-fold, respectively). Welders' blood lead GMs were 1.7 time higher in worst-case vs. non-worst case scenarios. Exposure differences from other ancillary variables were too sparse to incorporate, were insufficiently variable, or were not statistically significant. Conclusions: Time, job, and industry differences in lead exposure were quantified across many studies. The blood lead model's attenuated job exposure contrast likely reflected its integration of exposure over weeks. The steeper temporal trends for blood likely reflected the protective effect of personal protective equipment.