Ruben C Arslan, Khandis Blake, Laura J Botzet, Paul-Christian Bürkner, Lisa DeBruine, Tom Fiers, Nick Grebe, Amanda Hahn, Ben C Jones, Urszula M Marcinkowska, Sunni L Mumford, Lars Penke, James R Roney, Enrique F Schisterman and Julia Stern

Not within spitting distance: salivary immunoassays of estradiol have subpar validity for cycle phase


Salivary steroid immunoassays are widely used in psychoneuroendocrinology to investigate the psychological effects of menstrual cycle phase. Though manufacturers advertise their assays as suitable, they have not been rigorously validated for this purpose. We collated data from eight studies across more than 1,200 women and more than 9,500 time points. All studies measured estradiol and progesterone and had at least one independent indicator of cycle phase (day in cycle relative to the luteinising hormone surge or a menstrual onset). Seven studies collected saliva; one study collected serum. In serum, all non-steroid cycle phase measures strongly predicted steroids in the expected manner. By contrast, salivary immunoassays of estradiol were only weakly predictable from cycle phase and showed an upward bias compared to expectations from serum. For salivary immunoassays of progesterone, predictability from cycle phase was more mixed, but two widely used assays performed poorly. Imputing average serum steroid levels from cycle phase may yield more valid values than several widely used salivary immunoassays. Tandem mass spectrometry may provide a valid alternative to widely used immunoassays and could be combined with imputation.

The preprint is available here: