Citation

Ahn CY, Lee JS. JMIR Res. Protoc. 2024; 13: e53597.

Copyright

(Copyright © 2024, JMIR)

DOI

10.2196/53597

PMID

38329791

Abstract

BACKGROUND: Nonsuicidal self-injury (NSSI) is a major global health concern. The limitations of traditional clinical and laboratory-based methodologies are recognized, and there is a pressing need to use novel approaches for the early detection and prevention of NSSI. Unfortunately, there is still a lack of basic knowledge of a descriptive nature on NSSI, including when, how, and why self-injury occurs in everyday life. Digital phenotyping offers the potential to predict and prevent NSSI by assessing objective and ecological measurements at multiple points in time.

OBJECTIVE: This study aims to identify real-time predictors and explain an individual's dynamic course of NSSI.

METHODS: This study will use a hybrid approach, combining elements of prospective observational research with non-face-to-face study methods. This study aims to recruit a cohort of 150 adults aged 20 to 29 years who have self-reported engaging in NSSI on 5 or more days within the past year. Participants will be enrolled in a longitudinal study conducted at 3-month intervals, spanning 3 long-term follow-up phases. The ecological momentary assessment (EMA) technique will be used via a smartphone app. Participants will be prompted to complete a self-injury and suicidality questionnaire and a mood appraisal questionnaire 3 times a day for a duration of 14 days. A wrist-worn wearable device will be used to collect heart rate, step count, and sleep patterns from participants. Dynamic structural equation modeling and machine learning approaches will be used.

RESULTS: Participant recruitment and data collection started in October 2023. Data collection and analysis are expected to be completed by December 2024. The results will be published in a peer-reviewed journal and presented at scientific conferences.

CONCLUSIONS: The insights gained from this study will not only shed light on the underlying mechanisms of NSSI but also pave the way for the development of tailored and culturally sensitive treatment options that can effectively address this major mental health concern. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/53597.

Language: en

Keywords

digital phenotype; digital phenotyping; ecological momentary assessment; EMA; emotion; emotions; heart rate; machine learning; mental health; mood; multilevel modeling; nonsuicidal self-injury; NSSI; predict; prediction; predictions; predictive; predictor; predictors; psychiatric; psychiatry; self-harm; self-injury; sleep; step; wearable; wearable device; wearables; wrist worn