JMIR Publications

ABSTRACT

Background:

Social media is now a normalized context where people express their feelings in real-time. These platforms are increasingly showing their potential to detect the mental health status of the population. Suicide prevention is a global health priority and efforts towards early detection are starting to develop, although there is a need for more robust research.

Objective:

This study aims to explore the emotional content and relevant risk factors of Twitter posts in Spain and their relationships with the severity of the risk of suicide at the time of writing the tweet.

Methods:

Through Twitter's public API, 2,509 tweets containing specific lexicon relating to suicide were filtered. Expert psychologist judges were trained to independently evaluate these tweets. Each tweet was evaluated by 3 experts. Tweets were filtered by experts according to their relevance to the risk of suicide. In the tweets, the experts evaluated: 1) the severity of the general risk of suicide and the risk of suicide at the time of writing the tweet (real-time risk); 2) the emotional valence and intensity of 5 basic emotions; 3) relevant personality traits; and 4) other relevant risk variables such as helplessness, desire to escape, perceived social support, and intensity of suicidal ideation.

Results:

8.61% (N=216) of tweets were considered suicidal by most experts. Results show correlations between the severity of the risk of suicide at the time and sadness (0.266; P<.001), joy (-0.234; P=.001), general risk (0.908; P<.001), and intensity of suicidal ideation (0.766; P<.001). The severity of real-time risk was significantly higher in people who expressed feelings of defeat and rejection (P=.003), a desire to escape (P<.001), a lack of social support (P=.03), helplessness (P=.001), and daily recurrent thoughts (P=.007). In the multivariate analysis, the intensity of suicide ideation was a predictor for the severity of suicidal risk at the time (0.311; P=.001), as well as being a predictor for fear (-0.009; P=.01) and emotional valence (0.007; P<.009). The model explained 75% of the variance.

Conclusions:

This exploratory study suggests that it is possible to identify emotional content and other risk factors in suicidal tweets with a Spanish sample. Emotional analysis and, especially, the detection of emotional variations may be key for real-time suicide prevention through social media.