Virtual stress testing: Analyzing endocrine, metabolic, cardiovascular, and psychological responses to the TSST-VR.

Background: The Trier Social Stress Test (TSST) is a widely used tool for inducing and measuring stress responses in a controlled environment. In this study, we aimed to explore the effectiveness of a virtual TSST (TSST-VR) in eliciting stress responses across multiple physiological and psychological markers. Methods: A sample of 24 participants underwent the TSST-VR, during which salivary cortisol, alpha-amylase (AA), blood glucose levels, heart rate (HR), root mean square of successive differences (RMSSD) as a measure of heart rate variability (HRV), and subjective stress ratings (NRS) were collected at multiple time points. Results: In a baseline-to-peak analysis, significant increases were observed in HR (MDiff = 13.04, 95 %-CI [8.19-17.90], p < .001), RMSSD (MDiff = 17.75, 95 %-CI [3.28-32.22], p < .001), AA (p < .001, r = 1.07), and NRS (p < .001, r = 1.31) measures following the TSST-VR. While no significant changes in cortisol levels were found in the baseline-to-peak analysis across all participants, a secondary cluster analysis identified distinct cortisol responders (baseline-to-peak rise >1.5 mmol/l). Within this group, high cortisol responders (HCR) showed significantly higher cortisol (Wald χ2(7) = 118.03, p < .001), HR (Wald χ2(8) = 17.91, p = .022), and AA levels (Wald χ2(7) = 17.13, p = .017) compared to low cortisol responders (LCR). Area-under-the-curve analysis further confirmed a more robust cortisol stress response in HCR. Conclusion: These findings suggest that the TSST-VR can effectively induce measurable stress responses and may provide insights into individual differences in physiological and metabolic stress reactions. The study highlights the potential of virtual stress paradigms in stress research and underscores the advantages of a virtual setting in terms of standardization and economic considerations.