Data Comparability Between Biomonitoring Studies for PCDD/Fs—Issues for the Use of the National Health and Nutrition Examination Survey (NHANES) Data : Epidemiology

Secondary Logo

Journal Logo

Abstracts: ISEE 22nd Annual Conference, Seoul, Korea, 28 August–1 September 2010: Biomarkers and Biomonitoring

Data Comparability Between Biomonitoring Studies for PCDD/Fs—Issues for the Use of the National Health and Nutrition Examination Survey (NHANES) Data

Patterson, Donald G. Jr1,2; O'Sullivan, Gwen1; Sandau, Court D.1

Author Information
doi: 10.1097/01.ede.0000391761.94075.15
  • Free

O-29A5-4

Background/Aims:

The goal of this study was to examine data from other case studies that used National Health and Nutrition Examination Survey (NHANES) data to compare individuals or populations for blood levels of polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans (PCDD/Fs). Our aim was to evaluate whether the data collected from a number of studies could be correctly compared to NHANES data.

Methods:

The methods used in the collection and analysis of samples for both NHANES and other case studies were examined to determine if methodologies were similar enough to conduct direct comparisons of blood PCDD/F data.

Results:

Numerous considerations and issues were discovered when examining other studies and their comparison to NHANES data. These included detection limits from their results being higher than those generated by Centers for Disease Control and Prevention (CDC). These artificially amplify the calculated toxic equivalents (TEQs) for individuals. NHANES uses enzymatic lipid determinations to calculate blood lipids and many studies still use gravimetric lipid determinations. This further amplifies perceived TEQs for individuals being compared to NHANES. These errors combined with other data quality issues are exacerbating exposure scenarios and potentially causing a misclassification of individuals or study cohorts.

Conclusion:

Many studies that are comparing their data to NHANES data are doing so incorrectly. They are either not conducting the appropriate statistical treatment of the NHANES data or they have results from laboratories that are not capable of producing the quality of data required to compare with the NHANES dataset and therefore, misrepresenting the exposures that they are reporting in their studies.

© 2011 Lippincott Williams & Wilkins, Inc.