D STRESS, an objective and efficient virtual reality-based tool to detect subthreshold PTSD in military population

Post-traumatic stress disorder (PTSD) is highly prevalent among military personnel. Subthreshold PTSD (sub-PTSD) is of particular concern because of its potential progression to full PTSD, its comorbidities (notably increased suicide risk), and its negative impact on operational capability. Current detection methods are limited, in particular due to stigma associated with mental health screening.

This study aims to validate an objective tool, D-STRESS, using physiological and behavioral data collected during virtual reality scenarios to detect sub-PTSD in military personnel.

Eighty-four active-duty participants were classified using CAPS-5 criteria into four groups: no-PTSD (CAPS = 0), mild sub-PTSD (0 < CAPS ≤ 1), moderate sub-PTSD (1 < CAPS < 2), and full PTSD (CAPS ≥ 2). Volunteers navigated an everyday-life virtual reality environment designed to highlight multisensory integration impairments associated with trauma-related fear conditioning. Data on heart rate variability (HRV), skin conductance, respiratory rate, and body temperature, along with behavioral measures and electroencephalographic activity, were recorded. Logistic regression models were constructed to generate risk scores.

The models achieved accuracy rates of 76-88% and AUC values of 0.73-0.88 in distinguishing moderate sub-PTSD from no-PTSD groups. The most discriminant variables were HRV parameters and skin conductance indices.

D-STRESS is a promising, objective and non-stigmatizing tool for the detection of sub-PTSD in military populations. By leveraging parasympathetic and sympathetic markers of autonomic nervous system function it offers an effective alternative to traditional methods. Further validation in larger and more diverse samples is essential to confirm its applicability and reliability.