. 2020 Dec 29;10:22403. doi: 10.1038/s41598-020-79859-x

Table 2.

Results of bivariate and semi-partial correlation analyses.

	Cortisol	2-AG	AEA	SEA¹	OEA¹	PEA²
Quantitative workload
r_S	0.419**	− 0.122	− 0.202	− 0.186*	0.046	− 0.016
p	0.005^a	0.387^e	0.303ⁱ	0.043^l	0.618^l	0.827^l
Childhood maltreatment exposure (MACE-20 score)
r_S	− 0.023	0.290*	0.086	0.215*	0.286**	0.235**
p	0.886^b	0.039^f	0.669^j	0.019^l	0.002^l	0.002^l
Number of experienced major life events (private life) (LEC-5 score)
r_S	− 0.169	− 0.018	0.133	0.027	0.237**	0.175*
p	0.274^a	0.900^e	0.499ⁱ	0.771^l	0.009^l	0.019^l
Number of experienced potentially traumatic events on duty (RESQ-CE score)
r_S	− 0129	− 0.052	0.091	− 0.045	0.080	− 0.026
p	0.408^b	− 0.716^f	0.652^j	0.628^l	0.385^l	0.731^l
Severity of posttraumatic stress symptoms (PCL-5 score)
r_S	− 0.027	0.097	− 0.008	0.143	0.109	0.056
p	0.864^c	0.503^g	0.969^k	0.125^m	0.242^m	0.458^m
Severity of physical symptoms (PHQ-15 score)
r_S	− 0.249	− 0.087	− 0.063	0.110	0.373***	0.204**
p	0.116^d	0.551^h	0.760^k	0.241ⁿ	< 0.001ⁿ	0.007ⁿ
Severity of depressive symptoms (PHQ-9 score)
r_S	− 0.208	− 0.103	− 0.050	0.279**	0.197*	0.079
p	0.192^d	0.479^h	0.809^k	0.002ⁿ	0.035ⁿ	0.299ⁿ

*p < 0.050, **p < 0.010, ***p < 0.001, two-tailed. Bold-faced correlation are significant. Underlined p-values are significant after correction with false discovery rate. Semi-partial Spearman correlations were computed to control the associations of ¹sex and hair treatment or ²sex, age, and hair treatment with the respective hair concentration. Sample sizes varied due to differing number of detected steroids and lipids in hair, as well as due to missing values in sociodemographic and psychological data (for details see the methods section). Each correlation was analysed using all respectively available data of the whole study cohort: n = ^a44, ^b43, ^c42, ^d41, ^e52, ^f51, ^g50, ^h49, ⁱ28, ^j27, ^k26, ^l60, ^m59, ⁿ58.