Social Web Mining for Suicide Prevention

Completed

• Conditions: Suicidal Behaviors

• Registration Number: NCT04052477
• Lead Sponsor: University Hospital, Montpellier

Brief Summary

According to a recent and alarming WHO (World Health Organisation) report (September 4, 2014), one person dies of suicide every 40 seconds in the world. Suicide is the third-leading cause of death for 15- to 24-year-olds, according to the Centers for Disease Control and Prevention , after accidents and homicide.

This major public health issue need prevention strategies especially directed to at-risk populations. Since 2013, more than 2 billion users are enrolled in social networks such as Twitter or Facebook. Young adults (ages 18 to 29) are the most likely to use social media - fully 90% do.

Consequently, in this project, we focus on suicide prevention in social media network..

The aim of this project is the validation of the algorithm. This algorithm build a decision support system that monitor young people at-risk based on large volume of heterogeneous data collected through social media to improve suicide prevention.

Detailed Description

This study is composed of two steps :

1. 9 subjects were recruited. After patients agreement, computer scientists were accessing to patient social network profile. Computer scientists were not able to visualize the content of publications, just run the algorithm that will analyse the content of messages (text, frequency, emoticons...)

The algorithm defines the 3 most at-risk periods of suicide behaviors, on the next month. This result were compared to periods found by psychiatric interview. The psychiatrist then confirmed or not to LIRM whether periods found by the algorithm conrrespond to those defined by the psychitrist. No data of the social network were collected.

2. the 2nd step aim to improve the algorithm by collecting sociodemographic and clincal data related to patients included.

Study Timeline

• Study Start: 2017-01-01
• Primary Completion: 2017-06-01
• Study First Submitted: 2019-07-12
• Study First Submitted QC: 2019-08-08
• Study First Posted: 2019-08-09
• Study Completion: 2020-12-30
• Status Verified: 2021-09
• Last Update Submitted: 2021-09-01
• Last Update Posted: 2021-09-02

Recruitment & Eligibility

• Status: COMPLETED
• Sex: All
• Target Recruitment: 9

Inclusion Criteria

Not provided

Exclusion Criteria

Not provided

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
sensitivity of the algorithm	1 day	baseline

Trial Locations

Locations (1)

Uhmontpellier; 🇫🇷 Montpellier, France