Understanding Patterns Before They Become Outcomes.

Studies establish HRV as a measurable suicide risk biomarker. Wilson et al. (Heart Rate Variability and Suicidal Behavior, 2016) found that suicide attempters show measurably lower HRV during stress, reflecting a physiological failure in emotional regulation. Sheridan et al. (HRV and Its Ability to Detect Worsening Suicidality in Adolescents, 2021) demonstrated that a wrist wearable detected HRV changes tracking directly with clinical suicide severity scores. Lee et al. (Association of Resting HRV With Proximal Suicidal Risk, 2021) confirmed HRV deviation as a transdiagnostic suicide risk marker across 1,461 patients regardless of diagnosis.

Three peer-reviewed studies establish the clinical foundation. Bishop et al. (Sleep Disorders and Suicide Attempts Following Discharge from Residential Treatment, 2023) found that sleep disorders accelerate suicide attempt timelines in veterans, and that sleep medicine treatment measurably reduces that risk. McCarthy et al. (Sleep and Timing of Death by Suicide Among U.S. Veterans, 2019) found veterans are eight times more likely to die by suicide during nighttime waking hours. Yu et al. (Depressive Symptoms as a Mediator Between Excessive Daytime Sleepiness and Suicidal Ideation Among College Students, 2022) found disrupted sleep predicts suicidal ideation independent of depression across nearly 7,000 students.

Shen et al. (Passive Sensing for Mental Health Monitoring Using Machine Learning With Wearables and Smartphones) and Sheikh et al. (Wearable, Environmental, and Smartphone-Based Passive Sensing for Mental Health Monitoring) reach the same conclusion: passive sensing of heart rate, sleep, movement, and social behavior produces reliable digital biomarkers for depression, anxiety, PTSD, and bipolar disorder. Both studies confirm the technology works. Both demand larger populations, stronger privacy frameworks, and external validation to save lives at scale.