Premium
Inter‐laboratory variation in corticosterone measurement: Implications for comparative ecological and evolutionary studies
Author(s) -
Fanson Kerry V.,
Németh Zoltán,
Ramenofsky Marilyn,
Wingfield John C.,
Buchanan Katherine L.
Publication year - 2017
Publication title -
methods in ecology and evolution
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.425
H-Index - 105
ISSN - 2041-210X
DOI - 10.1111/2041-210x.12821
Subject(s) - corticosterone , variation (astronomy) , biology , ecology , hormone , endocrinology , physics , astrophysics
Interspecific comparisons of endocrine data are useful for drawing broad conclusions concerning the role of ecological variables in the evolution of physiological pathways. However, comparisons of endocrine data generated by different research groups are problematic, due to inter‐laboratory variation in measured hormone values. To date, we know of no study which has quantified the extent of inter‐laboratory variation in the measurement of hormone levels, outside of biomedical studies. To evaluate the extent to which laboratories differ in their measurement of hormones, we prepared seven samples of avian plasma with known concentrations of corticosterone and sent them for blind analyses to 19 laboratories and asked them to report the methods used and the values obtained. Both absolute hormone concentrations and the ratios between samples were equally variable, up to an order of magnitude different for some concentrations. Laboratory identity accounted for more than 80% of the variation in reported corticosterone, but we could not identify any methodological factors that consistently contributed to this inter‐laboratory variation. In addition, laboratory measurement error was significantly correlated with the latitude of the primary study species for each laboratory, suggesting that inter‐laboratory variation has the potential to drive trends in corticosterone datasets. Inter‐laboratory variation in corticosterone measurement may have serious implications for quantitative comparisons of endocrine values across laboratories, although comparisons of qualitative patterns may be more robust because rank order of the samples was relatively consistent across laboratories. Ignoring laboratory effect and the non‐independence of data may lead to an inflated rate of Type I error and spurious correlations.