Can suicide risk and emotional states be detected in Twitter and Instagram posts of individuals at risk?
| ISRCTN | ISRCTN60734624 |
|---|---|
| DOI | https://doi.org/10.1186/ISRCTN60734624 |
- Submission date
- 09/05/2022
- Registration date
- 08/06/2022
- Last edited
- 08/08/2022
- Recruitment status
- No longer recruiting
- Overall study status
- Completed
- Condition category
- Mental and Behavioural Disorders
Plain English Summary
Background and study aims
With over 800,000 deaths every year, suicide has become a global public health priority and affects especially the young population. Social media has already proven its potential for real-time detection of mental health disorders such as depression or post-traumatic stress disorder. Recent studies are now revealing the potential of social media in suicide prevention. These new approaches are developing real-time identification of individuals at risk, either by people or by computer language processing systems. The objective of this study is to continue these efforts for further validation, in this case, in Spanish social media.
The present study is the continuation of the Tweetcare project and aims to compare Twitter and Instagram posts of people who have a history of suicide behavior from individuals with no reported suicidal symptoms in the last year to clarify if suicide risk can be detected by social media.
Who can participate?
Adults aged 18 years and over who are active users of Twitter and Instagram.
What does the study involve?
The content of the publication of social media profiles will be analyzed with a computer tool developed by "machine learning" techniques. This tool allows the detection of emotions and potential risk posts from users on the social networks Twitter and Instagram through artificial intelligence. If during data computer processing, a profile is detected and classified as high risk, we will contact the participant to be informed of this fact to contact their reference health services.
What are the possible benefits and risks of participating?
There are no known benefits to participants
There are no known risks to participants
Where is the study run from?
University of Zaragoza (Spain)
When is the study starting and how long is it expected to run for?
March 2021 to May 2023.
Who is funding the study?
This project is not funded by any institution or company.
Who is the main contact?
Bárbara Oliván-Blázquez, bolivan@unizar.es,
Yolanda López-Del-Hoyo, yolandal@unizar.es
Contact information
Principal Investigator
Calle de Violante de Hungría, 23
Zaragoza
50009
Spain
 ORCID ID |
0000-0001-6565-9699 |
|---|---|
| Phone | +34 636 44 13 15 |
| bolivan@unizar.es |
Principal Investigator
Calle de Violante de Hungría
23
Zaragoza
50009
Spain
| ORCID ID |
0000-0002-1690-4130 |
|---|---|
| Phone | +34 630 03 01 25 |
| yolandal@unizar.es |
Study information
| Study design | Observational cross-sectional cohort study |
|---|---|
| Primary study design | Observational |
| Secondary study design | Cohort study |
| Study setting(s) | Internet/virtual |
| Study type | Screening |
| Participant information sheet | Not available in web format, please use contact details to request a participant information sheet. |
| Scientific title | Can suicidal behavior be detected by analysing the emotional content and suicide risk assessments of Twitter and Instagram posts of individuals with a history of suicidal behavior in the last year, compared with individuals reporting nonsuicidal behavior? |
| Study acronym | #SocialMediaCare |
| Study hypothesis | 1. Descriptive analysis of the emotional states and suicide risk in the social media posts in a population with no reported suicidal behavior and a population with reported suicidal behavior in the last year. 2. Descriptive analysis of the emotional states and suicide risk in the social media posts in a population with no reported suicidal behavior but depression or other mental pathologies and a population with reported suicidal behavior in the last year. 3. There will be differences in the emotional content of the posts in the population with no reported suicidal behavior and the population with reported suicidal behavior in the last year. 4. There will also be differences between individuals with no reported suicidal behavior but depression or other mental pathologies and the population with reported suicidal behavior in the last year. 5. There will be differences in the suicide risk assessment of the posts in the population with no reported suicidal behavior and in the population with reported suicidal behavior in the last year. 6. There will also be differences in the population with no reported suicidal behavior but depression or other mental pathologies and the population with reported suicidal behavior in the last year. |
| Ethics approval(s) | Approved 20/04/2022, Clinical Research Ethics Committee of Aragón (Comité Ético de Investigación Clínica de Aragón, CEIC Aragón, Instituto Aragonés de Ciencias de la Salud (IACS), Centro de Investigación Biomédica de Aragón (CIBA), Avenida San Juan Bosco 13, Zaragoza 50009, Spain; +34 976 71 5836; ceica@aragon.es), CEICA number PI21-164 |
| Condition | Suicide risk |
| Intervention | Participants will complete an online survey and agree to the extraction of information from their social media posts. Enrolment to survey end will be approximately 15 min. When their profiles are not public, there will be a second phase in which we will contact participants to ask for permission to view their content from the 12 months prior to registration. There is no follow-up. The total duration of the participation for those who need to give permission is no longer than 60 days. Emotional content and suicide risk in the online social media posts of healthy individuals, individuals with depression or other mental pathologies, and a population with reported suicidal behavior in the last year based on automatic classification and expert evaluations. |
| Intervention type | Other |
| Primary outcome measure | 1. Suicide risk of individuals is measured using SBQ (Suicidal Behaviors Questionnaire) and Plutchik Suicide Risk Scale 2. Suicide risk in the content of social media posts is measured on a scale of severity of suicide risk at the time of writing the tweet (real-time risk): from 0 (no existent risk) to 4 (extreme risk), by the computerized automatic classification system and also assessed by experts |
| Secondary outcome measures | 1. Emotional content analysis, including 5 basic emotions: anger, sadness, joy, surprise, and fear, will be measured on a scale of 0-100, by the computerized automatic classification system and also assessed by experts. 2. Emotional valence, measured on a scale 0 to 100, by the computerized automatic classification system and also assessed by experts 3. Severity of general suicide risk, measured on a scale from 0 (no existent risk) to 4 (extreme risk), by the computerized automatic classification system and also assessed by experts. 4. Daily time spent on social media, measured on a self-reported scale. 5. Psychopathology in the last year, measured on a self-reported item. |
| Overall study start date | 01/03/2021 |
| Overall study end date | 01/05/2023 |
Eligibility
| Participant type(s) | All |
|---|---|
| Age group | Adult |
| Lower age limit | 18 Years |
| Sex | Both |
| Target number of participants | 225 |
| Participant inclusion criteria | 1. People aged 18 years or over 2. People with the ability to read and write in Spanish 3. Active users or have used Twitter and/or Instagram in the last year |
| Participant exclusion criteria | 1. People aged 17 years or under 2. People with cognitive impairment or impaired reality judgment. |
| Recruitment start date | 25/06/2022 |
| Recruitment end date | 01/05/2023 |
Locations
Countries of recruitment
- Spain
Study participating centre
Zaragoza
50003
Spain
Sponsor information
University/education
Avenida San Juan Bosco, 13
Zaragoza
50009
Spain
| Phone | +34 976 71 6818 |
|---|---|
| info@iisaragon.es | |
| Website | https://www.iisaragon.es/ |
"ROR" |
https://ror.org/03njn4610 |
Funders
Funder type
University/education
No information available
Results and Publications
| Intention to publish date | 01/01/2024 |
|---|---|
| Individual participant data (IPD) Intention to share | Yes |
| IPD sharing plan summary | Available on request |
| Publication and dissemination plan | Planned publication in a high-impact peer-reviewed journal |
| IPD sharing plan | Current IPD sharing statement as of 08/08/2022: The datasets generated during and analysed during the current study will be available upon request only for replicability purposes. An anonymized dataset will be accessible after the study is completed. To request it please contact prevencionsuicidio@unizar.es. Previous IPD sharing statement: It is highly probable that, because of the nature of the data, the datasets analysed during the current study will be available only upon request and only for replicability purposes. Nevertheless, data-sharing plans for the current study are yet unknown and will be made available at a later date. |
Editorial Notes
08/08/2022: IPD sharing statement added.
15/06/2022: The recruitment start date was changed from 10/05/2022 to 25/06/2022.
20/05/2022: Trial’s existence confirmed by Clinical Research Ethics Committee of Aragón.
