Correction: Emotional Distress During COVID-19 by Mental Health Conditions and Economic Vulnerability: Retrospective Analysis of Survey-Linked Twitter Data With a Semisupervised Machine Learning Algorithm.

[This corrects the article DOI: 10.2196/44965.].

Emotional Distress During COVID-19 by Mental Health Conditions and Economic Vulnerability: Retrospective Analysis of Survey-Linked Twitter Data With a Semisupervised Machine Learning Algorithm.

BACKGROUND: Monitoring the psychological conditions of social media users during rapidly developing public health crises, such as the COVID-19 pandemic, using their posts on social media has rapidly gained popularity as a relatively easy and cost-effective method. However, the characteristics of individuals who created these posts are largely unknown, making it difficult to identify groups of individuals most affected by such crises. In addition, large annotated data sets for mental health conditions are not easily available, and thus, supervised machine learning algorithms can be infeasible or too costly. OBJECTIVE: This study proposes a machine learning framework for the real-time surveillance of mental health conditions that does not require extensive training data. Using survey-linked tweets, we tracked the level of emotional distress during the COVID-19 pandemic by the attributes and psychological conditions of social media users in Japan. METHODS: We conducted online surveys of adults residing in Japan in May 2022 and collected their basic demographic information, socioeconomic status, and mental health conditions, along with their Twitter handles (N=2432). We computed emotional distress scores for all the tweets posted by the study participants between January 1, 2019, and May 30, 2022 (N=2,493,682) using a semisupervised algorithm called latent semantic scaling (LSS), with higher values indicating higher levels of emotional distress. After excluding users by age and other criteria, we examined 495,021 (19.85%) tweets generated by 560 (23.03%) individuals (age 18-49 years) in 2019 and 2020. We estimated fixed-effect regression models to examine their emotional distress levels in 2020 relative to the corresponding weeks in 2019 by the mental health conditions and characteristics of social media users. RESULTS: The estimated level of emotional distress of our study participants increased in the week when school closure started (March 2020), and it peaked at the beginning of the state of emergency (estimated coefficient=0.219, 95% CI 0.162-0.276) in early April 2020. Their level of emotional distress was unrelated to the number of COVID-19 cases. We found that the government-induced restrictions disproportionately affected the psychological conditions of vulnerable individuals, including those with low income, precarious employment, depressive symptoms, and suicidal ideation. CONCLUSIONS: This study establishes a framework to implement near-real-time monitoring of the emotional distress level of social media users, highlighting a great potential to continuously monitor their well-being using survey-linked social media posts as a complement to administrative and large-scale survey data. Given its flexibility and adaptability, the proposed framework is easily extendable for other purposes, such as detecting suicidality among social media users, and can be used on streaming data for continuous measurement of the conditions and sentiment of any group of interest.

Reflection of Suicidal Ideation in Terms Searched for by Japanese Internet Users.

Background: The relationship between search terms and suicidal ideation is not precisely known. Aims: This study examined the relationship between the types of terms used in web searches and the level of suicidal ideation among individuals who conduct such searches. Methods: A total of 508 Internet users completed a suicidal ideation scale and conveyed to us the words they used when searching for a consultation site. The site was run by the authors using Internet advertisement listings, to consult people via e-mail who had searched for suicide-related words. We divided the users into three groups based on the most salient search terms: suicide method terms, suicide but no method terms, and other terms. Results: The scores on the suicide ideation scale were compared using one-factor analysis of variance, and differences were found between the groups. Users searching for suicide method-related terms had the strongest suicidal ideation. Limitations: This study was limited to users who contacted online consultation sites because of their help-seeking intentions. Conclusion: It is especially important for support organizations to identify Internet users who directly search for suicide methods.

Changes in Suicide Ideation Among Users of Online Gatekeeping Using Search-Based Advertising.

Recently, the practice of "online gatekeeping" related to suicide prevention-placing online advertisements directed at users who search the web for suicide-related terms and leading them to email-based consultations-has been implemented. However, the effectiveness of this practice has not been examined. Therefore, we quantitatively examined the effectiveness of suicide-related online gatekeeping. This was a non-randomized, single-arm study of online gatekeeping among Japanese Internet users. Outcomes were measured at the beginning of the use of the consultation service (T1) and approximately four weeks later (T2). The primary outcome measured was suicidal ideation. In total, 167 participants completed two assessment surveys; approximately 80% were women, and 59.3% were young people (in their twenties or younger). Data analysis of 167 online gatekeeping service users showed that the mean suicidal ideation at T2 was statistically significantly lower than that at T1, and its effect size was small to medium (d = 0.37, 95% confidence interval: 0.15-0.59). Online gatekeeping is a promising suicide-prevention tool, although further research is warranted to examine its effectiveness using a randomized controlled trial.HighlightsWe examined the effectiveness of email-based consultations for suicide prevention.About 4 weeks later, users' suicidal ideation was significantly reduced (d = 0.37).This was a single-arm study and requires a more rigorous study design in the future.

Relationship between Beck Hopelessness Scale and suicidal ideation: A short-term longitudinal study.

We aimed to clarify if the Beck Hopelessness Scale (BHS) predicted change in suicidal ideation in the general population. This study used a prospective cohort design (4-week survey interval) (N = 3,295). The online survey covered the BHS, suicidal ideation, depression/anxiety tendencies, and demographic details. Hierarchical multiple regression analysis showed hopelessness was a significant predictor of change in suicidal ideation from T1 to T2 (B = 0.07, 95% confidence interval: 0.05-0.09). Although high hopelessness was related to a short-term rise in suicidal ideation, the risk for suicide cannot be judged adequately by hopelessness from a clinical perspective.

Pre-pandemic suicidal ideation and COVID-19 preventive behavior: a prospective cohort study.

We investigated the association between suicidal ideation and COVID-19 preventive behaviors using a prospective observational longitudinal design (N = 6683). The baseline survey was 24 January 2020 (before COVID-19 pandemic, T1) and the follow-up survey was between April 27 and 30, 2020 (during COVID-19 pandemic, T2). The sample was the general adult population in Japan. Adjusted for demographic and socioeconomic factors, high suicidal ideation at T1 was associated with significantly low levels of engagement in COVID-19 preventive behaviors at T2. Individuals with strong suicidal ideation before the pandemic were less likely to engage in COVID-19 preventive behaviors.

Characteristics of Train Stations Where Railway Suicides Have Occurred and Locations Within the Stations.

Background: To devise effective railway suicide countermeasures, it is necessary to identify stations where suicide is likely to occur. Aim: We explored the characteristics of stations where railway suicides have occurred and locations within the stations. Method: (Study 1) Using suicide data from between April 2014 and September 2019 provided by a major railway company in Japan, station-specific suicide was modeled as an outcome variable in a multivariate Poisson regression model. (Study 2) With railway company staff, we visited stations where suicide frequently occurs and conducted fieldwork. Results: (Study 1) Our estimation using a Poisson regression model revealed that railway suicides were more frequent when stations were serviced by passing trains, had a large number of passengers, and were located near psychiatric hospitals. (Study 2) Of 50 suicides, 48.0% occurred in front of benches or waiting rooms, 26.0% occurred at the front end of the platform, 24.0% occurred at the entrance to the platform, and 22.0% occurred at a blind spot for the train driver. Limitations: All data were provided by one railway company in Japan, limiting the generalizability of the results. Conclusion: Stations where suicide occurs frequently have distinct characteristics. Focusing on suicide hotspots may aid suicide prevention.

Short-Term Effect of the COVID-19 Pandemic on Suicidal Ideation.

Background: Few studies have examined the effect of pandemics on suicide-related outcomes. Aims: We examined whether suicidal ideation levels among the general population changed owing to the COVID-19 pandemic by tracking individuals between January and April 2020. Method: We used a prospective observational longitudinal design (n = 6,683) and stratified sampling to conduct online surveys of the general adult population in Japan before (baseline) and during the pandemic (follow-up). Results: Suicidal ideation levels were significantly lower during than before the pandemic; however, the effect size was very small. Participants who were younger, with unstable employment, without children, with low income, and receiving psychiatric care were more likely to have higher suicidal ideation levels during the pandemic. Limitations: Because this was an Internet survey and subject to selection bias, the sample was not necessarily representative of the Japanese population. At the time of the survey, COVID-19 cases and deaths in Japan were relatively lower than in other developed countries. The dropout rate may have affected the results. Conclusion: Although the short-term impact of the COVID-19 pandemic on suicidal ideation is limited, relatively young and economically vulnerable individuals are more likely to show exacerbated suicidal ideation during the pandemic.

What character strengths are protective factors for suicidal ideation?: A cross-sectional study in Japan.

One factor that has been studied as a protective factor for suicide is character strength and there are many studies that examined the relationship between gratitude and suicide. The problem of character strength research on suicide prevention is that the link between various character strengths and suicide has not been comprehensively examined. We explored which character strength is associated with suicidal ideation. The study is a cross-sectional, self-administered questionnaire survey. Participants were 2,000 Internet users in Japan aged 20 years or older. The online survey covered suicidal ideation, character strengths and demographic details. A hierarchical multiple regression analysis was performed with the suicidal ideation as the dependent variable and character strengths as the independent variable. A significant regression equation was found (F (22, 1977) = 64.65, P < 0.001), with an R2 of 0.42. 'Engagement' (B = -0.38, P < 0.001), 'Love' (B = -0.33, P = 0.001), and 'Risk-taking' (B = 0.40, P < 0.001) were significant predictors of suicidal ideation. The positive trait of being able to actively engage with people and build a loving relationship could be a protective factor for suicide.

Relationship between annual household income and suicidal ideation: a cross-sectional study.

Previous studies have shown that high income was associated with low risk of death by suicide; however, the extent to which household income reduces suicidal ideation remains unclear. We performed a survey to examine the relationship between annual household income and suicidal ideation and determine the level of annual household income at which suicidal ideation no longer decreased as income increased. In total, 94,131 members of Internet survey panels were recruited as participants by a major Japanese Internet survey company in 2016. The participants completed a self-administered questionnaire consisting of items pertaining to demographics, annual household income, and suicidal ideation. We performed logistic regression analysis with suicidal ideation as the dependent variable and income level as the independent variable. The results showed that odds ratios for suicidal ideation decreased as annual household incomes increased, with sex, age, educational level, marital status, and parenthood adjusted for; however, once annual household incomes exceeded 8 million JPY (approximately 77,700 USD), the risk of suicidal ideation no longer decreased. We conclude that suicidal ideation was negatively associated with annual household income and 8 million JPY (approximately 77,700 USD) is a threshold beyond which further increases in income no longer reduce individuals' suicidal ideation.

Preferences for suicide prevention strategies among university students in Japan: a cross-sectional study using full-profile conjoint analysis.

In Japan, implementation of suicide prevention policies is mainly supported by public funds. However, it is unclear what kinds of suicide prevention strategies people prefer. We conducted a survey to clarify people's preferences regarding suicide prevention strategies adopted in Japan. This was a cross-sectional self-administered questionnaire survey. Participants were recruited through a lecture held by the first author at their university. We distributed questionnaires to 324 people present at the lecture; 249 completed questionnaires were included in the final analysis. We estimated suicide prevention strategy preferences using full profile conjoint analysis. For all six prevention strategies, the inclusion of each strategy in the policy profile was statistically significantly related to the positive evaluation of the overall policy profile. Marginal Willingness to Pay (MWTP) for restriction of access to means showed the highest value of the six suicide prevention strategies. MWTP was the lowest for public awareness activities. Preferred suicide prevention strategies in Japan are restriction of access to means and enhancement of psychiatric services. The results of this study indicate the strategies that are preferred have a high level of evidence of suicide prevention and do not directly intervene in a person's free will to die by suicide.

Impact of educational intervention on willingness-to-pay for suicide prevention: a quasi-experimental study among university students in Japan.

Many individuals show negative attitudes toward the use of taxes for suicide prevention. Activities that enhance knowledge and awareness of suicide and suicide prevention may increase willingness to pay (WTP) for suicide prevention. WTP is the amount that a consumer will pay for a product or service. The present study examined the influence of educational activities on enhancing knowledge and awareness of suicide and its prevention on WTP. We conducted a quasi-experimental study to examine the influence of lectures on suicide by comparing the amount of change in WTP between two groups over the same period: an intervention group (n = 92) comprising students who participated in the lectures and a control group (n = 128) comprising general university students recruited through an Internet-based survey. A t-test showed that the amount of change was significantly larger in the intervention group (t (152.31) = 2.25, p = .026). Ordinal logistic regression analysis showed that increased WTP was significantly correlated with an annual household income of JPY 4-6 million or higher. It may be appropriate to conclude that participation in lectures about suicide is effective in increasing WTP for suicide prevention.

Appropriate Targets for Search Advertising as Part of Online Gatekeeping for Suicide Prevention.

BACKGROUND: Gatekeeper training is an effective suicide prevention strategy. However, the appropriate targets of online gatekeeping have not yet been clarified. AIM: We examined the association between the outcomes of online gatekeeping using the Internet and the characteristics of consultation service users. METHOD: An advertisement to encourage the use of e-mail-based psychological consultation services among viewers was placed on web pages that showed the results of searches using suicide-related keywords. All e-mails received between October 2014 and December 2015 were replied to as part of gatekeeping, and the obtained data (responses to an online questionnaire and the content of the received e-mails) were analyzed. RESULTS: A total of 154 consultation service users were analyzed, 35.7% of whom were male. The median age range was 20-29 years. Online gatekeeping was significantly more likely to be successful when such users faced financial/daily life or workplace problems, or revealed their names (including online names). By contrast, the activity was more likely to be unsuccessful when it was impossible to assess the problems faced by consultation service users. CONCLUSION: It may be possible to increase the success rate of online gatekeeping by targeting individuals facing financial/daily life or workplace problems with marked tendencies for self-disclosure.

The relationship between attitudes toward suicide and willingness to pay for suicide prevention: a cross-sectional study in Japan.

There are gaps in our knowledge of the role attitudes toward suicide play in determining people's willingness to participate (WTP) for suicide prevention. We conducted a large nationwide cross-sectional study with the aim of clarifying the relationship between WTP for reducing suicide risk and attitudes toward suicide. Ordinal logistic regression analyses (n = 1771) showed that there were significant associations of WTP for suicide prevention with 'Suicide as a right' (ß = -.15, 95% CI: -.25 to -.04, p = .006), 'Preventability/readiness to help' (ß = .81, 95% CI: .69-.94, p < .001) and 'Common occurrence' (ß = .32, 95% CI: .19-.46, p < .001). 'Incomprehensibility/unpredictability' did not show an association with WTP. Taxpayer acceptance for suicide prevention is more likely to be achieved through provision of information that increases endorsement of 'preventability/readiness to help' and 'common occurrence' factors, and decreases 'suicide as a right' scores.

Willingness to pay for suicide prevention in Japan.

The present study aimed to establish willingness to pay for suicide prevention among taxpayers in Japan. We conducted an internet-based questionnaire survey using a double-bounded dichotomous choice contingent valuation, and analyzed data for 956 participants. The median willingness to pay to reduce the mortality risk from suicide by 25% was JPY 1,572 ($13.67 USD). Being married was significantly associated with willingness to pay. The willingness to pay to reduce the mortality risk from suicide may be lower than that to reduce other mortality risks such as traffic accidents.

Suicide prevention through online gatekeeping using search advertising techniques: a feasibility study.

BACKGROUND: Nurturing gatekeepers is an effective suicide prevention strategy. Internet-based methods to screen those at high risk of suicide have been developed in recent years but have not been used for online gatekeeping. AIMS: A preliminary study was conducted to examine the feasibility and effects of online gatekeeping. METHOD: Advertisements to promote e-mail psychological consultation service use among Internet users were placed on web pages identified by searches using suicide-related keywords. We replied to all emails received between July and December 2013 and analyzed their contents. RESULTS: A total of 139 consultation service users were analyzed. The mean age was 23.8 years (SD = 9.7), and female users accounted for 80% of the sample. Suicidal ideation was present in 74.1%, and 12.2% had a history of suicide attempts. After consultation, positive changes in mood were observed in 10.8%, 16.5% showed intentions to seek help from new supporters, and 10.1% of all 139 users actually took help-seeking actions. CONCLUSION: Online gatekeeping to prevent suicide by placing advertisements on web search pages to promote consultation service use among Internet users with suicidal ideation may be feasible.

The association of suicide-related Twitter use with suicidal behaviour: a cross-sectional study of young internet users in Japan.

BACKGROUND: Infodemiology studies for suicide prevention have become increasingly common in recent years. However, the association between Twitter use and suicide has only been partially clarified. This study examined the association between suicide-related tweets and suicidal behaviour to identify suicidal young people on the Internet. METHODS: A cross-sectional survey was conducted using Internet survey panels (n=220,848) comprising users in their 20s, through a major Japanese Internet survey company. Final analyses included the data of 1000 participants. RESULTS: Of the participants (n=1000) used in the final analysis, 61.3% were women and the mean age was 24.9 years (SD=2.9, range=20-29). Logistic regression analyses showed that tweeting "want to die" and "want to commit suicide" was significantly related to suicidal ideation and behaviour. Lifetime suicide attempts, the most powerful predictor of future suicide out of all suicidal behaviours, were more strongly associated with tweeting "want to commit suicide" than tweeting "want to die". Having a Twitter account and tweeting daily were not associated with suicidal behaviour. LIMITATIONS: An online panel survey has some inherent biases, such as coverage bias. Respondents were already registered as members of a particular Internet survey company in Japan, which limits the possibility of generalization. CONCLUSIONS: Twitter logs may be used to identify suicidal young Internet users. This study provides a basis for the early identification of individuals at high risk for suicide.

The impact of suicidality-related internet use: a prospective large cohort study with young and middle-aged internet users.

BACKGROUND: There has been no study that has allowed clear conclusions about the impact of suicide-related or mental health consultation-related internet use. AIM: To investigate the impacts of suicide-related or mental health consultation-related internet use. METHODS: We conducted prospective observational longitudinal study with data collection at baseline screening (T0), 1 week after T0 (T1) and 7 weeks after T0 (T2). Participants with a stratified random sampling from 744,806 internet users were 20-49 years of age who employed the internet for suicide-related or mental health consultation-related reasons and internet users who did not. The main outcome was suicidal ideation. Secondary outcome measures comprised hopelessness, depression/anxiety, and loneliness. RESULTS: The internet users who had employed the internet for suicide-related or mental health consultation-related reasons at T0 (n = 2813), compared with those who had not (n = 2682), showed a significant increase in suicidal ideation (ß = 0.38, 95%CI: 0.20-0.55) and depression/anxiety (ß = 0.37, 95%CI: 0.12-0.61) from T1 to T2. Those who disclosed their own suicidal ideation and browsed for information about suicide methods on the web showed increased suicidal ideation (ß = 0.55, 95%CI: 0.23-0.88; ß = 0.45, 95% CI: 0.26-0.63, respectively). Although mental health consultation with an anonymous other online did not increase suicidal ideation, increased depression/anxiety was observed (ß = 0.34, 95%CI: -0.03-0.71). CONCLUSIONS: An increased suicidal ideation was observed in the young and middle-aged who employed the internet for suicide-related or mental health consultation-related reasons. Mental health consultation via the internet was not useful, but those who did so showed worsened depression/anxiety.

The effect of suicide-related Internet use on users' mental health.

BACKGROUND: Previous studies have shown that suicide-related Internet use can have both negative and positive psychological effects. AIMS: This study examined the effect of suicide-related Internet use on users' suicidal ideation, depression/anxiety tendency, and loneliness. METHOD: A two-wave panel study of 850 Internet users was conducted via the Internet. RESULTS: Suicide-related Internet use (e.g., browsing websites about suicide methods) had negative effects on suicidal ideation and depression/anxiety tendency. No forms of suicide-related Internet use, even those that would generally be considered positive, were found to decrease users' suicidal ideation. In addition, our results suggest that the greater the suicidal ideation and feelings of depression and loneliness of Internet users, the more they used the Internet. CONCLUSION: Since suicide-related Internet use can adversely influence the mental health of young adults, it is necessary to take measures to reduce their exposure to such information.