Moira Mahoney , 2025-05-09 18:41:00
Key takeaways:
- Mood emerged as a stronger indicator of suicidal ideation than stress and sleep.
- Participants were significantly more likely to complete mood trackers than clinical assessments.
Researchers used data from the digital behavioral health platform NeuroFlow to identify an association between sleep, stress and mood and suicidal ideation, according to a study published in Journal of Technology in Behavioral Science.
Suicide is a leading cause of death worldwide and the age-adjusted suicide rate in the U.S. has increased over the past 20 years, despite efforts to address this issue, Tom Zaubler, MD, chief medical officer at NeuroFlow and colleagues wrote.
Data were derived from Kampa S, et al. J Technol Behav Sci. 2025;doi:10.1007/s41347-024-00478-8.
Further, individuals from historically marginalized groups, including young women, Black or Indigenous Americans and the LGBTQ+ community, are disproportionately affected by suicide.
“Psychiatric resources in the U.S. are scarce, and a majority of counties — especially in rural America — are critically underserved,” Zaubler, told Healio. “Clinicians in [rural] areas are doing heroic work, but the job of identifying and addressing suicide risk in a timely manner cannot be done manually.” he said.
Current detection of suicidal ideation (SI) — defined as thoughts about, consideration of or planning for self-inflicted harm — remains mostly restricted to primary care settings and is limited by patient under-reporting on standard assessments, natural variations in SI and the speed of suicide process, according to the researchers.
They added that digital behavioral health platforms such as NeuroFlow may improve SI detection by circumventing sociodemographic barriers and using ecological momentary assessments (EMAs) to continuously capture and deliver key physiological data such as change in sleep quality, therefore allowing clinicians to identify rising-risk situations and intervene to potentially save lives.
This inspired the researchers to conduct a retrospective database study of subjective mood ratings and their relationship to current SI risk based on concurrent self-harm assessments, all of which were collected via NeuroFlow.
The study included 30,725 individuals aged 18 years and older who were registered users of the NeuroFlow mobile app. Participants subscribed to daily trackers for sleep quality, stress, mood, pain impact and/or pain experience, which each utilized an 11-point ordinal scale from “awful” to “great.” Self-administered and provider-administered self-harm assessments, including the Patient Health Questionnaire9 (PHQ-9), were also collected and then recorded in NeuroFlow.
Overall, the participants collectively completed 5.7 million EMA trackers from May 2021 to August 2023. Users were significantly more likely to complete mood trackers than clinical assessments, with participants completing 18 trackers for every one clinical assessment on average, according to a press release from NeuroFlow.
According to exploratory binomial tests comparing severe vs. nonsevere EMA tracker scores, the researchers found strong associations between severe scores for sleep, stress and mood and SI expression within 30 days, with pain impact and experience associated to a lesser extent. Researchers noted this finding persisted on linear regression analyses.
Finally, the researchers performed a multilevel logistic regression controlled for EMA tracker and assessment history within individuals. In a model adjusted for age and sex, a one-point decrease in the mood EMA tracker value was linked to a 43% (adjusted OR = 1.43; 95% CI, 1.27-1.62) increase in SI risk. A similar increase in SI risk was observed for deterioration in stress (aOR = 1.29; 95% CI, 1.09-1.53) and sleep (aOR = 1.24; 95% CI, 1.11-1.38) scores.
The researchers noted several limitations to this study, including the potential inclusion of synthetic user profiles in the population.
“Our team believes that reducing suicide incidence is going to take advancements in technology that make it easier for clinicians to anticipate a patient’s moment of crisis,” Zaubler told Healio.
For more information:
Tom Zaubler, MD, can be reached at tomzaubler@neuroflow.com.
