Abstract
Conserving irreplaceable, archived serum samples may sometimes conflict with the objective of minimizing measurement error due to laboratory effects. We sought to determine whether we could successfully combine assay results for DDT-related compounds and polychlorinated biphenyls (PCBs) in serum from the same birth cohort obtained from different laboratories over time. Using the Child Health and Development Studies (CHDS) serum archive, we compared variability for assays of a quality control pool to variability for assays of subject serum. The quality control pool was created from native archived serum samples that were pooled, then aliquoted, blinded and inserted pair-wise into assay batches along with the subject serum for 5 studies using CHDS samples conducted over a 13year period by three different laboratories. We found that the variability between laboratory and over time within laboratory was small relative to inter-individual variability for p,p'-DDT (1,1,1-trichloro-2,2-bis(p-chlorophenyl)ethane), p,p'-DDE (1,1'-dichloro-2,2'-bis(p-chlorophenyl)ethylene) and o,p'-DDT (1,1,1-trichloro-2-(p-chlorophenyl)-2-(o-chlorophenyl)-ethane). Results were also consistent for most PCB congeners which were detectable in 85% or more of samples. Our results suggest that it is possible to combine assays for DDT and PCB congeners measured at positive levels as they are accumulated for cohort subjects without risking meaningful misclassification due to variation stemming from laboratory or time period. This has significant implications for future study costs, conservation of irreplaceable archived samples and for leveraging past investments for future research. For PCB congeners with very low levels, findings caution against pooling of assays without further exploration.