JMIR Publications

ABSTRACT

Background:

Social media platforms have been increasingly used to express suicidal thoughts, feelings, and acts, raising public concerns over time. A large body of literature has explored the suicide risks identified by people’s expressions in social media space. However, there is less assertive to conclude that social media provides public surveillance for suicide without being able to align suicide risks detected in social media space with actual suicidal behaviours. Corroborating this alignment is a crucial foundation for suicide prevention and intervention through social media and for estimating and predicting suicide in countries with no reliable suicide statistics.

Objective:

This study aims to corroborate whether the suicide risks identified in social media space align with actual suicidal behaviours. This aim is achieved by tracking suicide risks detected by 62-million tweets posted in Japan over a 10-year period and assessing the locational and temporal alignment of such suicide risks with actual suicide behaviours, recorded in the national suicide statistics.

Methods:

This study utilizes a human-in-the-loop approach to identify suicide risk tweets posted in Japan in the period from January 2013 to December 2022. This approach involves keyword-filtered data mining, data scanning by human efforts, and data refinement via an advanced natural language processing model termed the Bidirectional Encoder Representations from Transformers. The tweets-identified suicide risks are then compared with actual suicide records in both temporal and spatial dimensions to validate if they are statistically correlated.

Results:

Twitter-identified suicide risks and actual suicide records are temporally correlated by month in the 10 years from 2013 to 2022 (correlation coefficient= 0.533; p<0.001); this correlation coefficient is higher at 0.652 when we advance the actual suicide records one month earlier to compare with the actual suicide records. These two indicators are also spatially correlated by city with a correlation coefficient of 0.699 (p<0.001) for the 10-year period. Among the 267 cities with the top quintile of suicide risks identified from both tweets and actual suicide records, 73.5% (196 cities) overlapped.

Conclusions:

Social media platforms provide an anonymous space where people express their suicidal thoughts, ideation, and acts. Such expressions can serve as an alternative source to estimating and predicting suicide in countries without reliable suicide statistics. It can also provide real-time tracking of suicide risks, serving as the early warning for suicide committed. The identification of areas where suicide risks are highly concentrated is crucial for place-based mental health planning, enabling suicide prevention and intervention through social media in a spatially and temporally explicit manner. Clinical Trial: N/A