A methodological proof-of-concept of a data-driven, personalized, blended digital health intervention for suicidal thoughts and behaviors: A case series

Abstract

Introduction

Suicidal thoughts and behaviors (STBs) are a leading cause of death in the United States. Individuals at high-risk for suicide vary greatly in their precedents to STBs, which hinders suicide prevention strategies. Personalized approaches to mapping individualized precedents to suicide ideation might increase the impact and efficiency of treatment.

Methods

The present study describes a personalized, blended digital health treatment that uses idiographic network models derived from ecological momentary assessment to inform treatment targets (PeRsonalizEd Clinical Intervention for Suicide Events; PRECISE). PRECISE includes skills from dialectical behavior therapy and safety planning, two existing evidence-based treatments. In this case series, participants (N = 5) at high-risk for suicide and completed a 6-week treatment which included 5×/day ecological momentary assessments as well as weekly coaching sessions. Outcomes were assessed at baseline, post-treatment, and six-weeks post-treatment.

Results

In the intent-to treat sample, three of the five participants (60%) completed the full treatment protocol. Participants attended an average of 4.4 coaching sessions (73.3%), adherence was excellent (98%), and satisfaction was also high (4.2 out of 5). The severity of suicidal thoughts and behaviors were reduced at both post-treatment and the 6-week follow-up (d_zs = −1.33 to −2.00).

Conclusions

PRECISE is an example of a blended digital health interventions that capitalizes on time series data to personalize interventions for suicidal thoughts and behaviors. Incorporating real-time data and idiographic models to inform clinical decision making are promising tools to improve suicide care. Lessons learned and future directions for implementation are discussed.