Document is current

Article History

Received: 18 March 2026

Accepted: 22 May 2026

First Online: 29 May 2026

Declarations

:

: This study is a retrospective, computational secondary analysis of three publicly available, de-identified passive-sensing datasets and did not involve any new data collection from, or interaction with, human participants by the authors. According to the institutional research-ethics policy applicable to the authors and to standard journal guidance, secondary analysis of publicly released, fully de-identified data does not constitute human-subjects research, and a separate ethics review for the present analysis was therefore not required. The original data collection for each dataset was approved by the responsible institutional review boards: the StudentLife and College Experience Study (CES) datasets were collected at Dartmouth College under research protocols reviewed and approved by the Dartmouth College Committee for the Protection of Human Subjects (Institutional Review Board) ; the GLOBEM datasets were collected by the University of Washington under research protocols reviewed and approved by the University of Washington Institutional Review Board (Human Subjects Division) . All experimental procedures in the original studies were performed in accordance with the relevant institutional guidelines and regulations and with the principles of the Declaration of Helsinki (World Medical Association). In the present work, we further adhered to the data use terms and access conditions specified by each data provider (in particular, the PhysioNet Credentialed Health Data License and Data Use Agreement for GLOBEM), and processed all data exclusively in de-identified form.

: This study did not involve any new interaction with, or recruitment of, human participants by the authors. All analyses were conducted on de-identified secondary data released to the public by the original StudentLife, College Experience Study (CES), and GLOBEM teams. As reported in the original publications and accompanying dataset documentation, written informed consent was obtained from all individual participants prior to the original data collection: from all Dartmouth College student participants for StudentLife and CES (Dartmouth College Institutional Review Board), and from all participants enrolled in the four-year data collection waves for GLOBEM (University of Washington Institutional Review Board). All participants were 18 years of age or older at the time of enrollment in the original studies; no participants under 16 years of age were involved, so additional informed consent from parents or legally authorized representatives is not applicable.

: Not applicable. The manuscript does not contain any individually identifiable data, images, videos, or case details of any participant. All datasets used in this work are publicly released in de-identified form by the original data providers, and only aggregated quantitative results are reported.

: Not applicable. This is a retrospective computational modeling study on publicly released, de-identified passive-sensing datasets and is not a clinical trial.

: The authors declare no competing interests.