Share this article on:

Exposure Intensity Revisited

Acquavella, John; Alexer, Bruce; Mandel, Jack; Burns, Carol; Gustin, Christophe

doi: 10.1097/01.ede.0000220897.92561.b2
Letters to the Editor: Letters

Monsanto Company (retired), St. Louis, Missouri,

School of Public Health, University of Minnnesota, Minneapolis, Minnesota(Alexer)

Rollins School of Public Health, Emory University, Atlanta, Georgia(Mandel)

The Dow Chemical Company, Midland, Michigan(Burns)

Monsanto Europe S.D., Brussels, Belgium(Gustin)

Back to Top | Article Outline

To the Editor:

We thank Dr. Mage1 for his comments. We agree that systemic dose is a better indicator of biologically relevant pesticide exposure intensity than 24-hour urine concentration. Indeed, the former is more frequently used in the industrial hygiene and regulatory literatures, as one of us pointed out previously,2 whereas epidemiologists more frequently refer to the latter for reconstructing exposure histories. However, as we pointed out in our report,3 our results based on 24-hour urine concentrations and systemic doses were very similar. We chose the former to simplify the presentation and pointed out the very few instances in which there were moderate differences, none of which were consequential for our assessment of the exposure algorithm. In our data, there were very high Pearson correlations between log systemic doses and log 24-hour urine concentrations: 0.88 for glyphosate, 0.97 for 2,4-D, and 0.88 for the chlorpyrifos metabolite TCP. Moreover, as exposure algorithms used in epidemiologic exposure reconstructions often seek to identify relative intensity between individuals, 24-hour or single void urine collections are appropriate for determining relative exposure intensity for single applications.

We calculated systemic dose somewhat differently than Mage suggests. We did not correct our calculations for background levels or secondary applications because we wanted our dose estimates to reflect the cumulative dose from all sources of exposure.4 In all but a few instances, however, there was little contribution from these extraneous sources. We based our corrections, when participants had measurable urine concentrations at the end of our observation period, on each participant’s observed excretion kinetics. We believe this is more accurate than using a biologic half-life derived from animal tests.

We were explicit about what we considered to be the limitations of our approach for evaluating the previously published exposure algorithm and, by extension, less sophisticated approaches to retrospective exposure assessment. Although reliance on 24-hour urine concentration might add variability under some circumstances, it seems to be a negligible consideration in our recently published analyses.2

John Acquavella

Monsanto Company (retired)

St. Louis, Missouri

Bruce Alexer

School of Public Health

University of Minnnesota

Minneapolis, Minnesota

Jack Mandel

Rollins School of Public Health

Emory University

Atlanta, Georgia

Carol Burns

The Dow Chemical Company

Midland, Michigan

Christophe Gustin

Monsanto Europe S.D.

Brussels, Belgium

Back to Top | Article Outline


1. Mage D. Exposure intensity revisited [letter]. Epidemiology. 2006;17: 483.
2. Acquavella JF, Doe J, Tomenson J, Chester G, Cowell J, Bloemen L. Epidemiologic studies of occupational pesticide exposure and cancer: regulatory risk assessments and biologic plausibility. Ann Epidemiol. 2003;13:1–7.
3. Acquavella JF, Alexander BH, Mandel J, Burns C, Gustin C. Exposure misclassification in epidemiologic studies of agricultural pesticides: insights from biomonitoring studies of farmers. Epidemiology. 2006;17:69–74.
4. Acquavella JF, Alexander BH, Mandel JS, et al. Glyphosate biomonitoring for farmer-applicators and their families: results from the Farm Family Exposure Study. Environ Health Perspect. 2004;112:321–326.
© 2006 Lippincott Williams & Wilkins, Inc.