Related Experiment Video
Updated: Aug 15, 2026
Extraction and Analysis of Cortisol from Human and Monkey Hair
Published on: January 24, 2014
A systematic comparison of hair sampling methods in endocrine and endocannabinoid hair analysis: Is one sample enough
Katharina Huthsteiner1, Johannes B Finke1, Tim Klucken1
1Department of Psychology, University of Siegen, Siegen, Germany.
Background:
Hair analysis is widely used in psychoneuroendocrine (PNE) research to assess long-term hormone concentrations. However, it remains unclear how hair should be sampled to achieve the highest data quality. In research practice, hair is typically sampled either as one larger strand (single sample method) or by accumulating hair material from several smaller subsamples taken from nearby positions (accumulation method). Here, we provide a first systematic evaluation of these two methods for hair endocrine and endocannabinoid analysis using an extensive within-subject design.
Methods:
This preregistered study collected a total of twelve hair samples from 53 participants across two measurement occasions, three months apart. On each occasion, three samples were collected using the single sample method and three samples were collected using the accumulation method (three smaller subsamples). All samples were obtained using a systematic, horizontally and vertically balanced scheme within a small area around the posterior vertex. Samples were analyzed for cortisol, cortisone, progesterone, dehydroepiandrosterone, and endocannabinoids (AEA, 1/2-AG, OEA, SEA, PEA) via liquid chromatography-tandem mass spectrometry. Sampling methods were compared regarding within-person variability, mean analyte concentrations, test-retest reliability, and associations with external criteria, including body mass index (BMI), waist-to-hip ratio (WHR), and perceived stress.
Results:
Compared to the single samples, the accumulation method was associated with significantly lower within-person variability for most analytes and slightly stronger positive associations of hair glucocorticoids with BMI. Mean concentrations and test-retest reliability were largely comparable between methods. Exploratory analyses of single samples revealed vertical position-related concentration differences for certain analytes.
Conclusions:
Our findings suggest that the accumulation of smaller subsamples may help to detect more consistent hair endocrine and endocannabinoid measures. The observed position-related differences within a small scalp area also further highlight the importance of careful sampling protocols and methodological standardization in PNE hair research.
