Evaluating inter-study variability in phthalate and trace element analyses within the Children's Health Exposure...

Area of Science:

• Environmental Health Sciences
• Analytical Chemistry
• Biostatistics

Background:

• The Children's Health Exposure Analysis Resource (CHEAR) program enables environmental exposure assessment in biospecimens.
• CHEAR's laboratory network analyzes a wide range of environmental chemicals.
• Assessing inter-study variability is crucial for combining datasets across studies and laboratories.

Purpose of the Study:

• Establish a process for evaluating inter-study variability in analytical methods.
• Develop a standardized approach for quality control across CHEAR studies.

Main Methods:

• Created common quality control (QC) pools (two concentration levels) for urine samples.
• Inserted QC pools into sample batches at a rate of three per 100 samples.
• Assessed QC pool results for seven phthalates across five CHEAR studies and three lab hubs using multivariate control charts.
• Simulated outliers in blood QC samples to test conditions leading to out-of-control runs.

Main Results:

• Within-study analysis identified out-of-control runs for two of five studies assessing phthalates.
• Combining QC results across lab hubs brought these two studies into control but pushed two others out-of-control.
• Simulations showed 3-6 analytes with outlier values could cause out-of-control runs in 65-83% of cases.

Conclusions:

• Demonstrated a method for establishing acceptable variability bounds for analytical methods.
• Utilized QC materials across studies with multivariate control charts for robust data evaluation.
• Provides a framework for ensuring data comparability in large-scale environmental health research.