Dying online: live broadcasts of Chinese emerging adult suicides and crisis response behaviors

Predicting negative attitudes towards suicide in social media texts: prediction model development and validation study

Background

Implementing machine learning prediction of negative attitudes towards suicide may improve health outcomes. However, in previous studies, varied forms of negative attitudes were not adequately considered, and developed models lacked rigorous external validation. By analyzing a large-scale social media dataset (Sina Weibo), this paper aims to fully cover varied forms of negative attitudes and develop a classification model for predicting negative attitudes as a whole, and then to externally validate its performance on population and individual levels.

Methods

938,866 Weibo posts with relevant keywords were downloaded, including 737,849 posts updated between 2009 and 2014 (2009-2014 dataset), and 201,017 posts updated between 2015 and 2020 (2015-2020 dataset). (1) For model development, based on 10,000 randomly selected posts from 2009 to 2014 dataset, a human-based content analysis was performed to manually determine labels of each post (non-negative or negative attitudes). Then, a computer-based content analysis was conducted to automatically extract psycholinguistic features from each of the same 10,000 posts. Finally, a classification model for predicting negative attitudes was developed on selected features. (2) For model validation, on the population level, the developed model was implemented on remaining 727,849 posts from 2009 to 2014 dataset, and was externally validated by comparing proportions of negative attitudes between predicted and human-coded results. Besides, on the individual level, similar analyses were performed on 300 randomly selected posts from 2015 to 2020 dataset, and the developed model was externally validated by comparing labels of each post between predicted and actual results.

Results

For model development, the F1 and area under ROC curve (AUC) values reached 0.93 and 0.97. For model validation, on the population level, significant differences but very small effect sizes were observed for the whole sample (χ 2 1 = 32.35, p < 0.001; Cramer's V = 0.007, p < 0.001), men (χ 2 1 = 9.48, p = 0.002; Cramer's V = 0.005, p = 0.002), and women (χ 2 1 = 25.34, p < 0.001; Cramer's V = 0.009, p < 0.001). Besides, on the individual level, the F1 and AUC values reached 0.76 and 0.74.

Conclusion

This study demonstrates the efficiency and necessity of machine learning prediction of negative attitudes as a whole, and confirms that external validation is essential before implementing prediction models into practice.

Epidemiological characteristics and behaviors of online broadcast suicidality in China: implications for targeted prevention strategies

Objectives

Suicide is a global health concern, exacerbated by stigma around mental illnesses. Online platforms like Twitter and Sina Weibo have seen a rise in "online broadcast suicide," where individuals share suicidal thoughts and actions. However, there is limited understanding of the epidemiological characteristics, particularly in China. This study aims to analyze the demographics and behaviors of individuals engaging in online broadcast suicide in China to inform targeted prevention strategies.

Methods

A total of 525 incidents were identified through systematic retrieval of relevant news reports from online sources. Subsequently, a content analysis was performed on these reports to extract detailed information on the characteristics of each individual incident.

Results

Among the incidents analyzed, the male-to-female ratio was 1:1.6, with an average age of 23.1 ± 5.9 years. Approximately 71.9% took place in Southern China. Unemployment was reported in 15.0% of incidents. Relationship breakup (62.3%) was cited as the leading cause of suicide. Wrist cutting (58.2%) emerged as the predominant suicide method, and home (36.2%) was the most common location for these tragic events. Instant messaging apps were the primary platforms (54.7%) for conveying suicidal thoughts and actions. Additionally, among the 525 incidents examined, 12.0% disclosed having a mental disorder, and 7.6% had a history of prior suicide attempts. Significant variations were observed across age, gender, region, and occupation categories.

Conclusion

This study emphasizes the importance of developing suicide prevention programs for internet users. Besides, interventions should be customized to meet the specific needs of various populations.

Mind the gap: Exploring differences in suicide literacy between cybersuicide and offline suicide

Introduction

The highly public nature of cybersuicide contradicts long-held beliefs of offline suicide, which may cause differences in the way people perceive and respond to both of them. However, knowledge of whether and how suicide literacy differs between cybersuicide and offline suicide is limited.

Methods

By analyzing social media data, this paper focused on livestreamed suicide and aimed to compare suicide literacy between cybersuicide and offline suicide on three aspects, including false knowledge structure, extent of association with stigma, and linguistic expression pattern. 7,236 Sina Weibo posts with relevant keywords were downloaded and analyzed. First, a content analysis was performed by human coders to determine whether each post reflected suicide-related false knowledge and stigma. Second, a text analysis was conducted using the Simplified Chinese version of LIWC software to automatically extract psycholinguistic features from each post. Third, based on selected features, classification models were developed using machine learning techniques to differentiate false knowledge of cybersuicide from that of offline suicide.

Results

Results showed that, first, cybersuicide-related posts generally reflected more false knowledge than offline suicide-related posts ( χ 1 2 = 255.13, p < 0.001). Significant differences were also observed in seven false knowledge types. Second, among posts reflecting false knowledge, cybersuicide-related posts generally carried more stigma than offline suicide-related posts ( χ 1 2 = 116.77, p < 0.001). Significant differences were also observed in three false knowledge types. Third, among established classification models, the highest F1 value reached 0.70.

Discussion

The findings provide evidence of differences in suicide literacy between cybersuicide and offline suicide, and indicate the need for public awareness campaigns that specifically target cybersuicide.

Facebook and Suicidal Behaviour: User Experiences of Suicide Notes, Live-Streaming, Grieving and Preventive Strategies-A Scoping Review

BACKGROUND

Facebook represents a new dimension for global information sharing. Suicidal behaviours and attempts are increasingly reported on Facebook. This scoping review explores the various aspects of suicidal behaviours associated with Facebook, discussing the challenges and preventive measures.

METHODS

PubMed, Google Scholar, and Scopus were searched for related articles published in English up to October 2021, using different combinations of "Facebook" and "suicide". A group of experts comprising consultant psychiatrists screened the records and read the full-text articles to extract relevant data. Twenty-eight articles were chosen as relevant and included in the review under four selected themes.

RESULTS

Facebook impacts on suicidal behaviours in different aspects. Announcing suicides through sharing notes or personal information may lead to the prediction of suicide but be harmful to the online audience. Live-streaming videos of suicide is another aspect that questions Facebook's ability to monitor shared contents that can negatively affect the audience. A positive impact is helping bereaved families to share feelings and seek support online, commemorating the lost person by sharing their photos. Moreover, it can provide real-world details of everyday user behaviours, which help predict suicide risk, primarily through novel machine-learning techniques, and provide early warning and valuable help to prevent it. It can also provide a timeline of the user's activities and state of mind before suicide.

CONCLUSIONS

Social media can detect suicidal tendencies, support those seeking help, comfort family and friends with their grief, and provide insights via timelining the users' activities leading to their suicide. One of the limitations was the lack of quantitative studies evaluating preventative efforts on Facebook. The creators' commitment and the users' social responsibility will be required to create a mentally healthy Facebook environment.

Motivations for Online Expression, Willingness of Online Help-Seeking, and the Risk of Suicide Among Hong Kong Youths: A Mixed-Methods Study

Despite widespread interest in social media use, little is known about the heterogeneity of underlying motivations for online engagement among at-risk adolescents. This study adopted a mixed-methods approach to evaluate the motives of online expression among Hong Kong youths aged 15-19, as well as their relationships to online help-seeking willingness and suicide risks. In total, 6 focus groups and 12 individual interviews were conducted (N = 40) to elicit information regarding online expression purposes and to construct questionnaire items. The cross-sectional survey (N = 1,676) was undertaken subsequently to determine the prevalence of motives, and their correlations with online help-seeking willingness and suicide risks. Qualitative interviews revealed two key themes: major motivations for online expression, and features of online expression and help-seeking among suicidal youths. Quantitative analyses demonstrated the prevalence of each online expression purpose, with most reporting on "emotional expression" and "life sharing and documentation," but variations identified by gender, school academic banding, and suicide-related experience. Notably, an ordinal logistic regression model revealed that motives of "emotional expression" and "image building" were associated with increased suicide risks, while the purpose of "life sharing" was associated with decreased suicide risks. Those who expressed online for "self-expression" and "friend making" showed a stronger willingness for online help-seeking. These findings provide a more holistic overview of the purposes of online expression, which suggested the likelihood of mental health issues and the need for support. Such insight may facilitate the development of more effective suicide prevention strategies and online intervention services.

The Study of Para-Social Interaction With E-Word-of-Mouth for Influencer Marketing by Complex Computing

The purpose of this study was focused on exploring the relationship among the fans’ preferences, fans’ para-social interaction, and fans’ word-of-mouth. A survey consisted of 21 items based on the literature review and developed by this study. An online survey was distributed to the users of YouTube in Taiwan. A total of 606 valid samples was collected by survey. The instrument passed the reliability and validity test. Further, the data process applied the PLS (partial least squares) regression analysis methodology. The result shows that the ‘attractive’ impacted ‘para-social interaction’, ‘e-word-of-mouth’, and ‘preferences of fans’ positively. In addition, the para-social interaction plays an important role as a mediator between influencer’s attractiveness, w-word-of-mouth, and preferences of fans. Some suggestions were provided for social media influence’ related studies as reference.

Stigmatizing Attitudes Across Cybersuicides and Offline Suicides: Content Analysis of Sina Weibo

Background

The new reality of cybersuicide raises challenges to ideologies about the traditional form of suicide that does not involve the internet (offline suicide), which may lead to changes in audience’s attitudes. However, knowledge on whether stigmatizing attitudes differ between cybersuicides and offline suicides remains limited.

Objective

This study aims to consider livestreamed suicide as a typical representative of cybersuicide and use social media data (Sina Weibo) to investigate the differences in stigmatizing attitudes across cybersuicides and offline suicides in terms of attitude types and linguistic characteristics.

Methods

A total of 4393 cybersuicide-related and 2843 offline suicide-related Weibo posts were collected and analyzed. First, human coders were recruited and trained to perform a content analysis on the collected posts to determine whether each of them reflected stigma. Second, a text analysis tool was used to automatically extract a number of psycholinguistic features from each post. Subsequently, based on the selected features, a series of classification models were constructed for different purposes: differentiating the general stigma of cybersuicide from that of offline suicide and differentiating the negative stereotypes of cybersuicide from that of offline suicide.

Results

In terms of attitude types, cybersuicide was observed to carry more stigma than offline suicide (χ²₁=179.8; P<.001). Between cybersuicides and offline suicides, there were significant differences in the proportion of posts associated with five different negative stereotypes, including stupid and shallow (χ²₁=28.9; P<.001), false representation (χ²₁=144.4; P<.001), weak and pathetic (χ²₁=20.4; P<.001), glorified and normalized (χ²₁=177.6; P<.001), and immoral (χ²₁=11.8; P=.001). Similar results were also found for different genders and regions. In terms of linguistic characteristics, the F-measure values of the classification models ranged from 0.81 to 0.85.

Conclusions

The way people perceive cybersuicide differs from how they perceive offline suicide. The results of this study have implications for reducing the stigma against suicide.

Self-harm and suicidal content online, harmful or helpful? A systematic review of the recent evidence

Purpose

There are calls for greater regulation of online content related to self-harm and suicide, particularly that which is user-generated. However, the online space is a source of support and advice, including an important sharing of experiences. This study aims to explore what it is about such online content, and how people interact with it, that may confer harm or offer benefit.

Design/methodology/approach

The authors undertook a systematic review of the published evidence, using customised searches up to February 2021 in seven databases. The authors included empirical research on the internet or online use and self-harm or suicide content that had been indexed since 2015. The authors undertook a theoretically driven narrative synthesis.

Findings

From 4,493 unique records, 87 met our inclusion criteria. The literature is rapidly expanding and not all the evidence is high quality, with very few longitudinal or intervention studies so little evidence to understand possible causal links. Very little content online is classifiable as explicitly harmful or definitively helpful, with responses varying by the individual and immediate context. The authors present a framework that seeks to represent the interplay in online use between the person, the medium, the content and the outcome.

Originality/value

This review highlights that content should not be considered separately to the person accessing it, so online safety means thinking about all users. Blanket removal or unthinking regulation may be more harmful than helpful. A focus on safe browsing is important and tools that limit time and diversify content would support this.

Social Media Applications: A Potential Avenue for Broadcasting Suicide Attempts and Self-Injurious Behavior

Social media is a mainstay in American culture, but it’s impact on mental health continues to evolve. Social media has revolutionized the way people share information and communicate. It is particularly popular among adolescents and young adults ages 18-24. There has been increasing concern that the internet and social media may be influencing suicidal behavior and self-harm in this population. Publications focusing on the impact of social media use on suicidal/self-injurious behavior are beginning to emerge in the literature. This clinical vignette chronicles the events of a suicide attempt documented on the social media application Snapchat. It facilitates a discussion about the role social media played in influencing this behavior.

Social Media Use and Deliberate Self-Harm Among Youth: A Systematized Narrative Review

BACKGROUND

Suicide is now the 2^nd leading cause of death among adolescents and young adults. Social media's influence on youth suicidal risk or attenuation of risk is a novel and rapidly expanding topic of research that requires attention from a broad range of mental health services professionals. We aimed to provide an updated review of social media-related risk and protective factors to youth deliberate-self harm (DSH) to guide mental health services professionals in offering care and support to youth vulnerable to suicide.

METHODS

Studies on which primary research was conducted that evaluated young people's use of social media platforms related to DSH were systematically searched via Scopus and identified through expert recommendation and the Association for Computing Machinery's digital library of conference materials. The search focused on the timeframe June 2014 to September 2019, to offer an update since the time the most recent systematic reviews on this topic concluded their literatures searches. Quality was reviewed using the Mixed Methods Appraisal Tool (MMAT).

RESULTS

A total of 38 articles were eligible, and 24 articles rated as high quality were included in a narrative review. Of the included articles, 19 explored DSH risk and 8 explored DSH protection. Most articles reported on cross-sectional quantitative or qualitative studies. Opportunities for both risks and benefits were explored, and potential influences of social media use were considered for subgroups of youth who may be especially vulnerable to suicide.

CONCLUSIONS

In the relatively short period of review, the association between social media use and youth DSH was tested in population-based studies, offering preliminary evidence for suicide prevention and treatment efforts. Further research is needed to understand the contribution of social media use to youth DSH, particularly among youth most vulnerable to suicidal risk.

Suicide on Facebook-the tales of unnoticed departure in Bangladesh

Facebook has transformed social communication and offers the opportunity to share personal thoughts to people including suicide ideas, plans and attempts. Suicide after Facebook posts has been reported in different parts of the world and it has become a potential area of research for suicide prevention. The analysis of Facebook posts prior to suicide or Facebook live streaming may help in understanding the etiological factors, patterns of communication and possible prevention approaches for a particular community. However, there is a dearth of evidence about suicide incidents after Facebook posts and Facebook live streaming in low and middle-income countries. This study aims to explore the trends and phenomena of suicide after Facebook posts and live streaming in Bangladesh. We conducted an online search using the Google, Facebook and five daily online newspaper archives from 15th August to 15th September 2019. Two research assistants independently conducted the initial searching to find out people who committed suicide after Facebook posts or live streamed suicide in Bangladesh and documented 21 cases. After further evaluation of each of the 21 cases we confirmed 19 cases that met the selection criteria. All of them were under 35-years of age. We observed sucide after Facebook posts were more common in male(78%) e and students. Hanging was the most frequently used method of suicide followed by poisoning. Their Facebook posts and livestream videos indicated relationship problems, academic stress and mental disorders were the common stressors for their suicide. This study lays the foundation for the future researchers to work on suicidal posts on Facebook in Bangladesh and develop culture-specific, real-time suicide preventive systems using a social media platform.

How Health Communication via Tik Tok Makes a Difference: A Content Analysis of Tik Tok Accounts Run by Chinese Provincial Health Committees

During the last two decades, social media has immersed itself into all facets of our personal and professional lives. The healthcare sector is no exception, with public health departments now capitalizing on the benefits that social media offers when delivering healthcare education and communication with citizens. Provincial Health Committees (PHCs) in China have begun to adopt the micro-video sharing platform, Tik Tok, to engage with local residents and communicate health-related information. This study investigates the status quo of official Tik Tok accounts managed by PHCs in mainland China. In total, 31 PHC accounts were analyzed during August 2019, while the top 100 most liked micro-videos were examined using content analysis. Coding included three major aspects: Quantified Impact, Video Content, and Video Form. 45.2% (n = 14) of PHCs had official Tik Tok accounts. A limited number of accounts (n = 2) were yet to upload a micro-video, while most (n = 9) had uploaded their first micro-video during 2019. For the top 100 most liked micro-videos, a sharp difference was observed in terms of number of Likes, Comments and Reposts. Videos containing cartoons or documentary-style content were most frequently watched by citizens. Similarly, content that promoted professional health or provided knowledge of diseases was frequently viewed. Content containing original music, formal mandarin language, subtitles, and which lasted less than 60 s, were most frequently followed. It is considered a missed opportunity that most PHCs struggle to take advantage of the Tik Tok platform, especially given its growing popularity and daily increase in account creation.

A Psycholinguistic Analysis of Responses to Live-Stream Suicides on Social Media

Live-stream suicide has become an emerging public health problem in many countries. Regular users are often the first to witness and respond to such suicides, emphasizing their impact on the success of crisis intervention. In order to reduce the likelihood of suicide deaths, this paper aims to use psycholinguistic analysis methods to facilitate automatic detection of negative expressions in responses to live-stream suicides on social media. In this paper, a total of 7212 comments posted on suicide-related messages were collected and analyzed. First, a content analysis was performed to investigate the nature of each comment (negative or not). Second, the simplified Chinese version of the LIWC software was used to extract 75 psycholinguistic features from each comment. Third, based on 19 selected key features, four classification models were established to differentiate between comments with and without negative expressions. Results showed that 19.55% of 7212 comments were recognized as "making negative responses". Among the four classification models, the highest values of Precision, Recall, F-Measure, and Screening Efficacy reached 69.8%, 85.9%, 72.9%, and 47.1%, respectively. This paper confirms the need for campaigns to reduce negative responses to live-stream suicides and support the use of psycholinguistic analysis methods to improve suicide prevention efforts.

Mining social networks to improve suicide prevention: A scoping review

Attention about the risks of online social networks (SNs) has been called upon reports describing their use to express emotional distress and suicidal ideation or plans. On the Internet, cyberbullying, suicide pacts, Internet addiction, and "extreme" communities seem to increase suicidal behavior (SB). In this study, the scientific literature about SBs and SNs was narratively reviewed. Some authors focus on detecting at-risk populations through data mining, identification of risks factors, and web activity patterns. Others describe prevention practices on the Internet, such as websites, screening, and applications. Targeted interventions through SNs are also contemplated when suicidal ideation is present. Multiple predictive models should be defined, implemented, tested, and combined in order to deal with the risk of SB through an effective decision support system. This endeavor might require a reorganization of care for SNs users presenting suicidal ideation.

A systematic analysis of online broadcasts of suicidality in China

INTRODUCTION

This study attempts to analyze the epidemiological characteristics of online broadcasts of suicidality in China.

METHODS

A total of 193 incidents were identified by systematically searching for relevant news reports on the internet. A content analysis on these news reports was conducted to obtain information on the characteristics of each individual incident.

RESULTS

Of these incidents, the ratio of males to females was 1:1.7 (age: 24.6 ± 6.1 years old). Approximately 45.4% and 70.8% of incidents were located in the region of Eastern China and Southern China, respectively, and 15.7% of incidents were unemployed. The most common suicide method was wrist cutting (57.5%), and the most common suicide location was at home (35.2%). Instant messaging apps (52.8%) were most frequently used for broadcasting suicidal thoughts and actions. Relationship breakup (61.1%) was reported as the leading cause of suicide. Furthermore, of the 193 incidents, 19 indicated that they suffered from a mental disorder, and 12 revealed that they had a history of suicide attempts. There were significant gender, age, regional, and occupational differences across the categories.

DISCUSSION

This study indicates that in China, suicide prevention programmes need to target internet users.

Suicidality, psychopathology, and the internet: Online time vs. online behaviors

This study investigated whether several psychopathology variables, including suicidality, could predict the time people spend using the internet (hours online). Next, we examined a specific at-risk population (suicidal individuals) by their online behaviors, comparing suicidal individuals who went online for suicide-related purposes with suicidal individuals who did not go online for suicide-related purposes. An anonymous online sample of 713 (aged 18-71) reported hours online, psychiatric histories, and completed several standardized scales. After accounting for age and education, hierarchical regression modeling showed that the assessed psychopathology variables, including suicidality, did not explain significant variance in hours online. Hours online were better predicted by younger age, greater willingness to develop online relationships, higher perceived social support, higher curiosity, and lower extraversion. Suicidal participants, who did or did not go online for suicide-related purposes, did not differ on hours online. Multiple regression modeling showed that those who went online for suicide-related purposes were likely to be younger, more suicidal, and more willing to seek help from online mental health professionals. These findings revealed that hours online are not a valid indicator of psychopathology. However, studying online behaviors of specific at-risk groups could be informative and useful, including for suicide prevention efforts.

Assessing Suicide Risk and Emotional Distress in Chinese Social Media: A Text Mining and Machine Learning Study

BACKGROUND

Early identification and intervention are imperative for suicide prevention. However, at-risk people often neither seek help nor take professional assessment. A tool to automatically assess their risk levels in natural settings can increase the opportunity for early intervention.

OBJECTIVE

The aim of this study was to explore whether computerized language analysis methods can be utilized to assess one's suicide risk and emotional distress in Chinese social media.

METHODS

A Web-based survey of Chinese social media (ie, Weibo) users was conducted to measure their suicide risk factors including suicide probability, Weibo suicide communication (WSC), depression, anxiety, and stress levels. Participants' Weibo posts published in the public domain were also downloaded with their consent. The Weibo posts were parsed and fitted into Simplified Chinese-Linguistic Inquiry and Word Count (SC-LIWC) categories. The associations between SC-LIWC features and the 5 suicide risk factors were examined by logistic regression. Furthermore, the support vector machine (SVM) model was applied based on the language features to automatically classify whether a Weibo user exhibited any of the 5 risk factors.

RESULTS

A total of 974 Weibo users participated in the survey. Those with high suicide probability were marked by a higher usage of pronoun (odds ratio, OR=1.18, P=.001), prepend words (OR=1.49, P=.02), multifunction words (OR=1.12, P=.04), a lower usage of verb (OR=0.78, P<.001), and a greater total word count (OR=1.007, P=.008). Second-person plural was positively associated with severe depression (OR=8.36, P=.01) and stress (OR=11, P=.005), whereas work-related words were negatively associated with WSC (OR=0.71, P=.008), severe depression (OR=0.56, P=.005), and anxiety (OR=0.77, P=.02). Inconsistently, third-person plural was found to be negatively associated with WSC (OR=0.02, P=.047) but positively with severe stress (OR=41.3, P=.04). Achievement-related words were positively associated with depression (OR=1.68, P=.003), whereas health- (OR=2.36, P=.004) and death-related (OR=2.60, P=.01) words positively associated with stress. The machine classifiers did not achieve satisfying performance in the full sample set but could classify high suicide probability (area under the curve, AUC=0.61, P=.04) and severe anxiety (AUC=0.75, P<.001) among those who have exhibited WSC.

CONCLUSIONS

SC-LIWC is useful to examine language markers of suicide risk and emotional distress in Chinese social media and can identify characteristics different from previous findings in the English literature. Some findings are leading to new hypotheses for future verification. Machine classifiers based on SC-LIWC features are promising but still require further optimization for application in real life.