Menstrual variation in the acute testosterone and cortisol response to laboratory stressors correlate with baseline testosterone fluctuations at a within- and between-person level

Figures & data

Figure 1. Baseline salivary testosterone and cortisol concentration in athletic women collapsed across testing days (1 A, 1B) and stressor type (1 C, 1 D). Data are plotted as geometric marginal means (red diamonds) with a 95% CI overlaying a box plot with individual observations (grey circles). D7 = day 7, D14 = day 14, D21 = day 21. *D14 is significantly different from D7 and D21 p < 0.01, ^#Significant difference between stressors p < 0.01.

Figure 2. Change scores in log-transformed salivary testosterone and cortisol concentration in athletic women collapsed across testing days (A, B) and stressor type (C, D). Data are plotted as marginal means (red diamonds) with a 95% CI overlaying a box plot with individual observations (grey circles). D7 = day 7, D14 = day 14, D21 = day 21. *D14 is significantly different from D7 and D21 p < 0.01, ^#Significant difference between stressors p < 0.01.

Table 1. Regression models predicting the acute testosterone response to the physical and psychological stressors.

	Model 1			Model 2			Model 3			Model 4
Predictors	Est (%)	95% CI	p Value	Est (%)	95% CI	p Value	Est (%)	95% CI	p Value	Est (%)	95% CI	p Value
(Intercept)	8.42	7.11, 9.75	<0.001	8.42	7.40, 9.45	<0.001	8.40	7.37, 9.44	<0.001	9.31	8.57, 10.1	<0.001
D14	5.96	4.75, 7.18	<0.001	5.96	4.75, 7.18	<0.001	6.13	4.78, 7.50	<0.001
D21	−0.26	−1.40, 0.89	0.655	−0.26	−1.40, 0.89	0.655	−0.28	−1.44, 0.89	0.636
Stressor [Psychological]	−2.02	−2.77, −1.27	<0.001	−2.02	−2.77, −1.27	<0.001	−2.08	−2.85, −1.30	<0.001
Testosterone.pmc				0.14	0.10, 0.18	<0.001	0.14	0.10, 0.18	<0.001	0.14	0.10, 0.19	<0.001
Cortisol.pmc				−0.05	−1.99, 1.93	0.962	−0.05	−1.99, 1.93	0.962	−0.05	−2.43, 2.39	0.969
Testosterone.pc							−0.03	−0.11, 0.06	0.551	0.19	0.09, 0.28	<0.001
Cortisol.pc							−0.02	−0.89, 0.85	0.958	0.33	−0.79, 1.45	0.568
Observations	180			180			180			180
Marginal R^2/Conditional R^2	0.391/0.698			0.593/0.699			0.591/0.700			0.372/NA^a

^aUnable to compute random effect variances for estimating conditional R². Coefficients are expressed as a percentage [100×(exponentiate (coefficient)−1)].

D14: day 14; D21: day 21; testosterone.pmc: baseline testosterone person-mean centered; cortisol.pmc: baseline cortisol person-mean centered; testosterone.pc: baseline testosterone person-centered; cortisol.pc: baseline cortisol person-centered.

Significant p values are highlighted in bold.

Table 2. Regression models predicting the acute cortisol response to the physical and psychological stressors.

	Model 1			Model 2			Model 3			Model 4
Predictors	Est (%)	95% CI	p Value	Est (%)	95% CI	p Value	Est (%)	95% CI	p Value	Est (%)	95% CI	p Value
(Intercept)	11.7	10.5, 13.0	<0.001	11.7	10.8, 12.7	<0.001	11.7	10.8, 12.7	<0.001	11.6	11.0, 12.2	<0.001
D14	−2.82	−3.80, −1.84	<0.001	−2.82	−3.80, −1.84	<0.001	−2.85	−3.95, −1.73	<0.001
D21	−0.84	−1.84, 0.16	0.102	−0.84	−1.84, 0.16	0.102	−0.82	−1.83, 0.19	0.112
Stressor [Psychological]	2.24	1.40, 3.08	<0.001	2.24	1.40, 3.08	<0.001	2.24	1.37, 3.12	<0.001
Testosterone.pmc				−0.13	−0.16, −0.10	<0.001	−0.13	−0.16, −0.10	<0.001	−0.13	−0.16, −0.10	<0.001
Cortisol.pmc				0.00	−1.79, 1.83	0.996	0.00	−1.79, 1.83	0.996	0.00	−1.79, 1.83	0.996
Testosterone.pc							0.01	−0.08, 0.09	0.861	−0.13	−0.21, −0.05	0.002
Cortisol.pc							−0.32	−1.16, 0.52	0.455	−0.63	−1.57, 0.32	0.192
Observations	180			180			180			180
Marginal R^2/Conditional R^2	0.177/0.476			0.430/0.477			0.429/0.475			0.298/0.319

Coefficients are expressed as a percentage [100×(exponentiate (coefficient)−1)].

D14: day 14, D21: day 21, testosterone.pmc: baseline testosterone person-mean centered; cortisol.pmc: baseline cortisol person-mean centered; testosterone.pc: baseline testosterone person-centered; cortisol.pc: baseline cortisol person-centered.

Significant p values are highlighted in bold.

Figure 3. Mediating effect of testing day on the baseline person-centered testosterone (testosterone.pc) relationship with the testosterone (A) and cortisol (B) change score measures. The coefficients in parentheses indicate the relationship between baseline testosterone.pc and each hormonal change score, controlling for testing day. *Denotes a significant effect p < 0.05.

Data availability statement

The study data are not publicly available due to privacy issues.