Social Media and Suicide: A Review of Technology-Based Epidemiology and Risk Assessment

Abstract

Introduction:

Suicide is a significant public health problem among teenagers and young adults in the United States, placing significant stress on emergency departments (EDs) to effectively screen and assess for the presence of suicidality in a rapid yet efficient manner.

Methods:

A literature search was performed using PubMed and MEDLINE with the following terms: “Social media,” “Suicide,” “Facebook^®,” “Twitter^®,” “MySpace^®,” “Snapchat^®,” “Ethics,” “Digital Media,” and “Forums and Blog.” Data were extracted from each article, specifically the sample size, study setting, and design. Only English-language studies were included. We reviewed the reference lists of included articles for additional studies, as well. Abstracts, unpublished data, and duplicate articles were excluded.

Results:

A total of 363 articles met our initial criteria. Studies older than 10 years and/or in a language other than English were removed. After review, a total of 31 peer-reviewed articles were included in the study. Teenagers and young adults often fail to disclose risk factors to physicians, despite sharing them with the public on social media platforms such as Facebook and Twitter. Therefore, physician access to a patient's social media can assist in identifying suicidal ideation and/or acts.

Conclusions:

Viewing a patient's social media accounts can help ED physicians gain perspective into his or her mental health status and identify those at risk for suicide; however, ethical and privacy concerns associated with this method of data gathering make implementation of such a practice controversial. To justify its use, formal prospective studies analyzing if and how physician access to a patient's social media influences care should be performed.

Introduction

With an incidence of 800,000 cases each year, suicide is the second leading cause of death in people between the ages of 15 and 29 globally.¹ The interpersonal-psychological theory of suicide states that individuals who manifest both a feeling of isolation and burdensomeness (i.e., the desire to commit suicide) and possess the ability and means to lethally injure themselves (i.e., the capacity to carry out the act) are at highest risk for completed suicide.² In addition, those who have attempted suicide in the past are much more likely to attempt (and succeed) in the future.² A history of psychiatric illness, alcohol or substance abuse, interpersonal loss, childhood maltreatment, or family history of suicide are factors associated with an increased risk for suicide.³

Given the increasing incidence and widespread impact, suicide has become a significant public health concern, and preventive measures are being put into place to help lower the likelihood of suicide in both high- and low-risk populations. At present, suicide can be combated on a community level by increasing suicide awareness and educating individuals about high-risk behaviors and encouraging them to identify at-risk patients.^1,3 Patients who may have unexpressed suicidal thoughts and no clear risk factors pose a major challenge to suicide prevention strategies.

The digital revolution jump-started an unprecedented spread of information and dramatically increased global communication. This interaction has manifested in different forms, molding, and being molded by the context under which they emerged. Text messages provide a quick and easy way to directly communicate. Social networking services, or social media, serve to facilitate interaction with one another in different spheres, typically based on common interests. Each site is distinct with regard to their media offering. For example, Twitter^® caters to an audience seeking to share quick, concise thoughts; Instagram^® and Snapchat^® are photo and video based; and Facebook is a combination of multiple media forms. With such variety and ubiquity, there is now an unprecedented access to media and each other at all times.⁴

Recent studies have discussed how digital media can promote, instigate, and encourage violence such as cyberbullying and gang violence, especially in the preteen and teenage population.^5
–7 Similarly, digital media use impacts suicide rates, in some cases encouraging the act.^8

–12 Several studies demonstrated a correlation between an increase in social media use and risk of suicide.^13

–16 Headlines on large media outlets implicate cyberbullying as the cause for many suicides among young teenagers.¹⁷ Other articles discussed how the use of text messages encouraged teen suicides by facilitating the rapid development of suicide pacts and helping to hasten the planning process.^18,19

The aim of this study was to review literature looking at relationships between suicide and social media and to discuss the potential role social media can play in assessing and preventing suicide risk and attempts in a clinical setting.

Methods

A literature search was performed using PubMed and MEDLINE with the following terms: “Social media,” “Suicide,” “Facebook^®,” “Twitter^®,” “MySpace^®,” “Snapchat^®,” “Ethics,” “Digital Media,” and “Forums and Blog.” A total of 363 articles were found. Studies older than 10 years and/or written in a language other than English were removed. Boolean operators were used to combine search terms. Data were extracted from each article, specifically the sample size, study setting, and design. A total of 31 peer-reviewed articles were included in this review. All types of studies and designs, including case reports and case series, were considered for inclusion. The bibliographies of included studies were reviewed to provide supporting information regarding suicide, current assessment tools, and social media.

Results

Media and Suicide Exposure

Media give rise to exposure to suicidal encouragement

Ueda et al.²⁰ analyzed the relationship between user reaction to a celebrity suicide on Twitter and suicide rates. The research determined that, after a celebrity suicide, the incidence of suicide in the general population increased solely if the death elicited a significant reaction from Twitter users.²⁰ Conversely, suicides that did not garner significant social media attention had little impact on suicide rates in the general population. Thus, it is not the absolute number of suicides, but rather the platform on which they are presented that can serve as a catalyst for emulation suicide. Emulation, or copycat suicide, was first introduced as the Werther effect after Goethe's The Sorrows of Young Werther, wherein the main character's suicide sparked numerous subsequent suicide attempts and led to the book's banning in many European countries.²¹

In addition, social media sites have given a voice to pro-suicide groups, and studies have demonstrated how themes of normalization, glorification, and self-mutilation and depression acceptance were dominant within subpopulation-specific social media sites. An exploratory study of Tumblr^® content demonstrated that 9% of the website's postings were related to depression, self-harm, and suicidal ideation, many of which encouraged and normalized this behavior. A large percentage of these posts were written from anonymous accounts, giving rise to bolder statements and behavior from users.²² Suicide and nonsuicidal self-harm (NSSH) in similar groups on Facebook are sometimes glorified as methods of self-expression, rather than serious public health concerns. Shared personal beliefs and credos such as nihilism are prevalent on these pro-suicide pages.²⁰ Similarly, the connectivity of the internet allows for the facilitation of suicide pacts whereby suicidal individuals find a sense of community.¹⁹

Forums and blogs can be used to share suicidal methods

The internet and social media provide easy access to information pertaining to suicide methods.²³ A systematic review examining how the internet affects risk for self-harm and suicide in youth found that internet forums discussing self-harm and suicide often focused on execution and techniques.²⁴ Furthermore, there is little to no regulation of the content provided on these sites.²⁵ Between 2007 and 2014, the number of websites that included information on suicide methodology tripled, whereas websites designed to provide support to those contemplating suicide decreased by half.²³

Expression of Suicidality through Media

People are likely to express distress online surrounding similar topics

Systematic review of Marchant et al. revealed that young people who use social media are more likely to communicate their emotional distress online than to an adult or peer.²⁵ A study examining phrases posted by at-risk South Korean adolescents on social media found that an average of 19% of adolescents expressed suicidality online, most commonly relating to academic performance, self-image, bullying, and health concerns. School was the most commonly reported stressor.²⁶ According to demographic research, most of the suicidal comments were made by men, and many made references to relationships with family and romantic partners.²⁷

Expression of distress is variable

Studies examining variability in suicide notes demonstrated that subjects who left a paper note scored higher on the Beck Suicide Intent Scale and were older. Notes posted to social media were more common in the younger population and among those with a substance abuse history.²⁸ An analysis of Twitter postings revealed that two phrases, “want to die” and “want to commit suicide,” correlated with an increased odds ratio of suicidal ideation, self-harm, suicidal plan, and suicide attempt compared with Twitter users who did not use these phrases.²⁹ “Vaguebooking,” or posting intentionally vague comments regarding one's mental health to subtly bring attention to it, is also associated with significantly increased rates of loneliness and suicidality.³⁰

Predicting Suicidal Risk with Digital Footprint

Predictability of risk in populations who have committed suicide

Although suicide rates do not seem to linearly correspond to the frequency of the tweets, research demonstrates a regional association between prevalence of high-risk tweets and age-adjusted suicide rates, with the Rocky Mountains demonstrating the greatest prevalence of high-risk tweets and suicide rates.³¹ Ren et al. compared public blogs of individuals who committed suicide to the blogs of randomly chosen individuals and found that, among those who committed suicide, their postings contained significantly more negative emotion and language.²⁹

Tools used to assess for risk

Various tools have been developed to identify suicide risk. A study examining the American Association of Suicidology's mnemonic IS PATH WARM (Suicidal Ideation, Substance Abuse, Purposelessness, Anxiety, Trapped, Hopelessness, Withdrawal, Anger, Recklessness, and Mood Change) demonstrated it was helpful in distinguishing between risk for behavior classified as NSSH and true suicidal ideation, thus making it a useful tool for prioritizing individuals by severity of suicide risk.³² Other currently used suicide assessment tools include the Suicide Assessment Five-step Evaluation and Triage (SAFE-T) and the Columbia Suicide Severity Rating Scale (C-SSRS).^33,34 SAFE-T was developed by the United States Department of Health and Human Services as a way to measure an individual's suicidality.³⁴ The five steps include assessing risk and protective factors, suicidal inquiry, determination of risk with appropriate intervention, and documentation of findings and actions taken.³⁴ Similarly, the C-SSRS, created by the National Institute of Mental Health and approved as a diagnostic tool by the Food and Drug Administration, evaluates an individual for suicidal risk looking at ideation, preparation, and attempts as measures.³³ All these current tools are time consuming and limited by the need for an evaluator to administer them (Table 3).

There is recent literature on the development of technology-based assessments for suicide risk among social media users. Researchers in China developed a digital “suicide dictionary” containing 2,168 words pertaining to suicide and applied it to almost 5,000 posts on Sina Weibo^®, a Chinese microblogging server similar to Twitter. To assess the efficacy of this dictionary, posts were assessed and evaluated for suicide risk using both the Suicide Probability Scale (SPS) and the newly developed dictionary. The dictionary yielded comparable results to expert evaluator assessments of posters' risk for suicide.³⁵ Similar algorithms were applied to Twitter and demonstrate that posts concerning for suicide risk contain a higher word count, are more likely to be written in the first person, and are more focused on death.^36,37 Additional research evaluated methods to classify posts pertaining to suicide and exclude references to suicide news reports and memorials, as well as flippant suicidal declarations.³⁸ The study found that differentiating between real suicidal ideation and nonsuicidal declarations was the most significant challenge for the algorithm. Nonetheless, the classification algorithm agreed with human evaluators 85% of the time, making it relatively reliable³⁸ (Table 1).

Table 1.

Selected Review of Studies on Risk of Suicide or Self-Harm Behaviors on the Internet

STUDY	DESIGN	NO (MESSAGE OR PATIENT)	FINDINGS	CONCLUSIONS
Jiang et al.¹⁹	Retrospective review	62	Suicide victims were mainly people in their 20s–30s (84.1%), with the overall male-to-female ratio being 2.3:1. In total, 87.1% suicide victims completed suicide in closed hotels or rental homes or apartments, whereas charcoal-burning suicide accounted for 80.6% of cases	People who commit suicide because of an internet suicide pact are more likely to be men between their 20s and 30s. Charcoal-burning suicide in sealed hotels or rental housing was the most prevalent observed way of suicide in terms of internet suicide
Ueda et al.²⁰	Retrospective review of tweets or internet posts	>1,000,000	After a celebrity suicide, the number of suicides increased only if the death supported a substantial reaction from Twitter users. However, minimally reported suicides gained less attention, and, therefore, had a minimal impact on the suicide rates of the general population	Promotes the idea that the mode or platform on which suicides are presented can instigate suicides in return
Cavazos-Rehg et al.²²	Exploratory review	3,360	Nine percent of material posted on Tumblr^® demonstrated postings that were related to depression, self-harm, and suicidal ideation	There is a need for a Tumblr intervention dedicated toward suicide prevention since a great amount of depression and suicidal content was observed on the platform
Bidde et al.²³	Retrospective review	8,525	There was a growth of blogs and discussion outlets from 2007 to 2014 with an increase in hits relating to websites that detailed suicide methods. Specifically, hits specifically for suicide sites increased from 19% to 23%, whereas hits for formal help decreased from 13% to 6.5%	There should be an improved implementation of suicide reporting guidelines for sites
Daine et al.²⁴	Systematic review	51	There are correlations between internet exposure and violent modes of self-harm	Internet use can have both positive and negative effects on young people who are already at risk for suicide or committing self-harm
Marchant et al.²⁵	Systematic review	51	There is a relationship between internet use and self-harm/suicidal behavior; it was more commonly found among people who used the internet often and/or visited websites that included self-harm or suicide content	There are many cons associated with internet use (normalization, triggering, competition, contagion) and also many pros or benefits (crisis support, decrease social isolation, outreach). The younger generation seems to increasingly use social media to portray their distress, especially to their peers
Song et al.²⁶	Systematic review	99,693	A lower employment rate, higher rental prices index, and more bullying were associated with internet searches containing suicide-related words	Pressure related to school or education-related activities seems to be the biggest contributor to Korean adolescents' suicide risk
Cash et al.²⁷	Exploratory review	1,038	Content analysis was done on public profiles on MySpace^® and it was found that comments were significantly associated with hopelessness, despair, and desperation	Adolescents use public websites to display comments about their suicidal thoughts, behaviors, and possible intentions
Barrett et al.²⁸	Prospective mixed methods	1,435	New media note-leaving was associated with younger age and substance use; both risk factors for repeated self-harm. However, suicidal intent remained highest in paper note-leavers.	People who leave paper notes are at greatest risk for suicide; however, notes left on media are also correlated to suicide and self-harm risk. Clinicians should seek into media use to assess self-harm when conducting emergency department assessments
Berryman et al.³⁰	Correlational study	467	Social media use was not an accurate predictor of impaired mental health functioning. But, vague booking was predictive of suicidal ideation, suggesting this specific action could be a predictor for future problems	Social media use cannot be used to successfully suggest lack of healthy mental functioning, except for vague booking
Ren et al.²⁹	Experimental study	1,800	Emotion transition traits are related and associated with suicide risk	Blogs should be checked to further comprehend the relationship between emotions and thoughts for these authors
Burnap et al.³⁸	Prospective mixed methods	6,846	A number of machine classification models was made with the goal of distinguishing texts on Twitter related to suicide, with discriminating content based on suicidal ideation, reporting of a suicide, and support	The most important predictive principle component of the suicidal ideation class was found, but sustained classification procedures should be done to provide insight

Implementation of tools for prevention

Using social media as a suicide prevention tool is also a common phenomenon, and some websites and servers contain pages on suicide prevention and resources for those in crisis. In addition, a few platforms (notably Facebook and Twitter) have teams who deal with individuals whose postings are overtly suicidal. These teams reach out to the poster and provide support and resources.^38

–41 A similar program implemented on Sina Weibo used directed messaging and chatboxes to interact with potentially suicidal individuals in an effort to motivate them to seek help⁴² (Table 2).

Table 2.

Barriers to Suicide Prevention and Suggested Troubleshooting

BARRIERS TO SUICIDE PREVENTION	SUGGESTED
Suicidal encouragement from media	Suicides committed by celebrities should not be continuously reported or mentioned to the general public. With less attention devoted to making the public aware of suicides or serious self-harm committed by famous individuals, the risk of consequential suicides can be lessened
Online sharing of suicidal methods	Although many forums and blogs allow users to filter out posts based on their own needs, more regulation toward posts that detail execution and techniques for suicidal acts should be implemented. Websites that seem to have content focused on sharing suicide methodology should be downregulated by the owners themselves
Social media suicide normalization and glorification	Work with companies such as Tumblr and Facebook to monitor posts related to increased risk of suicide or self-harm. Individuals can be allowed to share their personal beliefs, but social media postings detailing a serious health dilemma should be monitored and evaluated
Regional or community association to suicide	Within certain regions, suicide prevention efforts should be increasingly implemented. Local efforts should be made to enhance efforts toward diminishing the attraction to suicide

Inclusion of Media in the Patient Profile

Evidence of successful use of media in patient record

There have been case reports detailing the use of social media as a means by which to assess suicide risk in the emergency department (ED). One such notable report described a young man who was altered, disoriented, and not providing a clear history to his ED providers. Physicians successfully identified him by his MySpace page and learned from his postings that he was experiencing severe psychosocial stress. This prompted a psychiatric evaluation, and it was subsequently determined that his altered mental status and disorientation had been the result of an intentional overdose of muscle relaxants.⁴³

Another case report documented the case of a medical student in his 20s with a long history of depression. He had a history of suicidal behavior, his first instance a thwarted attempt to jump off of a balcony. When the patient failed to match into a residency program, he attempted suicide by intentional medication overdose. Before doing so, he posted goodbye messages on his Facebook page, and an alarmed friend notified the police. The evaluating medical team sought and was granted permission to view his Facebook page, and the information therein helped them pinpoint the exact time of ingestion, thus aiding them in his treatment.⁴⁴

When an adolescent male using Snapchat, a picture and video-based communication app, broadcasted his suicide, his ex-girlfriend captured an image of a bottle of acetaminophen he sent her, and she then shared it with his parents. Using that picture, the treating physician was able to ascertain both the time and dosage of the ingestion. The patient underwent a speedy and uneventful recovery in large part because of the information gleaned from his own social media.⁴⁵

Ethics regarding patient privacy and sharing of information

Despite evidence that data mining a patient's social media can yield valuable information regarding risk and potential for suicidal behavior, questions pertaining to privacy in the digital age remain. These ethical concerns were demonstrated by the case of a 13-year-old girl who posted pictures with distinctly suicidal themes to her social media account. Her parents shared the images with the evaluating physician to assist in obtaining psychiatric evaluation and care. Questions then arose because her account had been set to “private.” Although she was a minor, and her mother brought the pictures to the physician's attention, the physician did not directly seek consent from the patient to view the “private” social media content, thus raising the question as to whether or not this constituted a violation of privacy.⁴⁶ In addition, consent to access an individual's specific social media account (i.e., Facebook) can result in indirect access to other social media accounts (i.e., Instagram) without their consent, adding to the difficulty navigating such situations.⁴⁷ The balance between providing the best possible care for suicidal individuals and the protection of their online privacy is difficult to achieve. Although there is enacted legislation regarding privacy on social media and the handling of protected health information (PHI) separately, no legislation addressing the intersection of the two currently exists.

The Health Insurance Portability and Accountability Act (HIPAA) contains specific language and policies regarding the proper handling of private patient information when it pertains to mental health and suicide risk. When a medical or mental health professional deems a patient to be at risk for hurting themselves or others, HIPAA allows for private patient information to be shared with individuals who may be able to intervene and lessen the risk, even without the explicit consent of the patient.⁴⁸ Information can be shared with family members, other health care professionals, and law enforcement. With respect to patients who are acutely incapacitated and unable to provide consent, such as those who have attempted suicide, relevant information regarding the patient can be shared with family members who are able to make medical decisions on the patient's behalf.⁴⁸ The use of technology and sharing patients' private social content for purposes of data gathering and medical decision-making is not covered under HIPAA, which was enacted in 1996, well before the current digital age (Table 3). In addition, the Declaration of Helsinki, adopted in 1964 and amended in 2013, outlines the ethical guidelines for medical research involving human subjects.⁴⁹ It contains similar wording to HIPAA, stressing the importance of patient consent while at the same time ensuring that the ultimate goal of research and treatment is patient health.

Table 3.

Selected Review of Commonly Used Tools to Assess for Risk and Their Purpose

COMMONLY USED TOOLS	ORIGIN	PURPOSE
IS PATH WARM (Suicidal Ideation, Substance Abuse, Purposelessness, Anxiety, Trapped, Hopelessness, Withdrawal, Anger, Recklessness, and Mood Change) Mnemonic	American Association of Suicidology	Differentiate between real suicidal ideation and non-suicidal self-harm (NSSH), in order to help rank the severity of an individual's risk for suicide
Suicide Assessment Five-Step Evaluation Triage	United States Department of Health and Human Services	Measure individual risk for suicide in a more detail-oriented and specified manner
Columbia Suicide Severity Rating Scale (C-SSRS)	National Institute of Mental Heath	Evaluate a person's suicidal risk by looking at factors such as ideation, preparation, and attempts as factors
Suicide Probability Scale	John Cull and Wayne Gill	To allow practitioners in determining the risk, especially for adolescents and college populations

There has been increased interest in legislation addressing social media privacy in recent years. From 2012 to 2014, 17 states enacted legislation regarding the access of social media accounts, private and public, by employers and educational institutions.⁵⁰ The Uniform Employee and Student Online Privacy Protection Act, adopted by three states as of 2018, aims to protect the individual, whether student or employee, from giving up their online privacy to anyone without their explicit consent.⁵⁰ This act does not include any wording addressing the patient–provider relationship and also explicitly places the burden of consent in the hands of the individual,⁵⁰ which is contradictory to the access granted by HIPAA.

The dilemma faced by medical professionals and researchers in this new area is whether to follow legislation regarding social media privacy OR legal statutes pertaining to the handling of PHI, because the existing laws exclude and contradict one another. At present such ethical challenges must be assessed subjectively based solely on professional judgment on a case-by-case basis. A 2018 study in Norway interviewed public health nurses about their handling of difficult health dialogues with adolescents, including unhealthy home situations and suicidal ideation. When navigating social media use, the nurses reported that they did not utilize any specific guidelines but instead relied on their own expertise and personal experiences.⁵¹ Dilemmas that could not be resolved by one provider were discussed in groups to reach a general consensus on a course of action.⁵¹ Variability in responses to ethical challenges such as these could be ameliorated by the implementation of legislation.

Discussion

A review of the current available literature demonstrates numerous barriers to suicide prevention efforts including media reports, social media use, the ethics of physician access to a patient's social media, and geography (Table 2).

Media Coverage

It is well-established that the amount of media coverage given to celebrity suicides correlates with increased rates of suicide and NSSH in the general population. School psychologists, guidance counselors, and health care providers should be particularly attuned to the attitude and emotions among their young contacts after a celebrity suicide. Furthermore, this association highlights the need for reevaluation of the current method of reporting celebrity death.²⁰ Media outlets should be mindful of the total amount of time spent covering the death and/or time at which said reports are released, taking into consideration periods of high readership among younger adults. In addition, when reporting a celebrity suicide, numbers for crisis hotlines should be provided in the event that viewers and readers are actively experiencing suicidal ideation. That said, unilateral development of regulations in the name of public safety runs the risk of being considered overly paternalistic and essentially censorship. It will continue to be challenging to strike an appropriate balance between public safety and freedom of the press and the public's right to information.

Social Media Use

Approaching websites and blogs containing content on suicide methodology presents similar issues to regulating media coverage of suicides. Deleting or restricting access to these sites could be seen as an infringement of freedom of speech and of the press. Therefore, websites could be encouraged to monitor for certain flagged words, similar to the work of Lv et al. in China on Sina Weibo.³⁵ Individual blogs with concerning content could then be marked with a content warning, requiring internet users to consent to viewing potentially distressing material.²³ Furthermore, companies should be encouraged to follow Facebook's efforts to link individuals manifesting suicidal ideation to appropriate resources.³⁹

Physician Access to Patient Social Media

As demonstrated by the aforementioned case reports,^44,45 physician access to a patient's social media can provide crucial information needed for the management of incapacitated or uncooperative patients. From an ethics standpoint, providers could argue the need to use all available information, including social media, to best treat a patient. Nevertheless, this clear benefit has to be balanced with the patient's right to autonomy and privacy. For example, if a patient is presenting for evaluation of an upper respiratory infection but is found to have social media postings indicating depressive symptoms, is it appropriate for the physician to broach the subject in that setting? Furthermore, how would the physician justify the data mining in this clinical scenario? What insights could be gleaned from a patient's social media that might aid in the diagnosis and treatment of their respiratory symptoms? It is hard not to interpret this as an invasion of privacy, and similar real-world scenarios could potentially damage the doctor–patient relationship.

Geography

With the demonstrated regional association of suicide risk,³⁰ providers in these areas, including psychiatrists, ED physicians, and primary care physicians, should be encouraged to become familiar with the variety of suicide risk assessment tools (Table 3). At the simplest level, mnemonics such as IS PATH WARM alert providers to a variety of manifestations of suicidal ideation, prompting further investigation.³⁵ With a positive IS PATH WARM, providers should consider conducting SAFE-T,³⁴ C-SSRS,³³ or SPS.³⁵ Formal assessments are often viewed as inherently time and labor intensive; however, they can be tailored to a variety of clinical environments. For example, on reviewing these three options, a provider could pick one with which he/she feels would best fit his/her practice. This would allow for mastery of a specific screening model in the hopes of improving patient care. In addition, all three of the aforementioned assessments take roughly 5–10 min at most and can provide invaluable information.

Limitations and Ongoing Research

At present, there are no clinical trials investigating the physician's use of social media to screen and assess for a patient's suicide risk. The use of retrospective reviews to correlate an individual's social media postings and their suicidal actions is difficult because of the ubiquitous use of pseudonyms and the inherent anonymity of the internet. An in-depth retrospective review would require access to the user's legal name, and then correlate to hospital and/or death records. Therefore, the aforementioned recommendations are based on case reports and demonstrate the clear need for formal studies.

At a small departmental level, physicians could be provided patient scenarios where an individual presents with vague somatic symptoms possibly related to depression (myalgia, headache, etc.) but no clear expressed suicidal ideation. A subset of these physicians could then be granted “access” to predesigned social media postings demonstrating active suicidal ideation. The care plans for physicians with access to social media would be compared with those of their colleagues without access. This could, in the future, support the implementation of formal prospective studies whereby patients voluntarily grant a hospital center access to their social media, and its influence on care plans is assessed.

Conclusion

Suicide is a leading cause of preventable death throughout the world today, with wide-reaching community and societal impacts. Screening for suicidal risk in the clinical setting is both necessary and challenging, becoming even more difficult with changes to technology and the way we communicate. This review of articles deduces that using social media can augment existing information and present key missing data when evaluating patients for suicidality and risky behaviors. This can aid in the prevention of suicide and also provide information vital to the care of patients who have attempted suicide. However, the benefits of using a patient's private posts must be tempered against the ethical questions surrounding privacy in the digital age. These concerns require further legal and ethical discussions regarding the physician's use of social media as a clinical tool.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

World Health Organization. Suicide. 2017. Available at http://who.int/mediacentre/factsheets/fs398/en/ (last accessed June 29, 2018 ).

Van Orden

, Witte

, Gordon

, Bender

, Joiner

. Suicidal desire and the capability for suicide: Tests of the interpersonal-psychological theory of suicidal behavior among adults. J Consult Clin Psychol, 2008; 76:72–83.

Centers for Disease Control and Prevention. Suicide: Risk and protective factors. 2017. Available at https://cdc.gov/violenceprevention/suicide/riskprotectivefactors.html (last accessed June 29, 2018 ).

Villanti

, Johnson

, Ilakkuvan

, Jacobs

, Graham

, Rath

. Social media use and access to digital technology in US young adults in 2016. J Med Internet Res, 2017; 19:e196.

Çelik

, Kalenderoğlu

, Almiş

, Turgut

. Copycat suicides without an intention to die after watching tv programs: Two cases at five years of age. Noropsikiyatri Ars, 2016; 53:80–81.

Upton

, Sung

, Ranney

, Patel

, Kelley

, Eschmann

, et al. Social media as a vector for youth violence: A review of the literature. Comput Human Behav, 2014; 35:548–553.

Bostic

Does social media perpetuate youth violence?. 2014. Available at http://yvpc.sph.umich.edu/social-media-perpetuate-youth-violence/ (last accessed June 29, 2018 ).

Associated Press. Rise in teen suicide connected to social media popularity: Study. 2017. Available at https://nypost.com/2017/11/14/rise-in-teen-suicide-connected-to-social-media-popularity-study/ (last accessed June 29, 2018 ).

Chuck

Is social media contributing to rising teen suicide rate?. 2017. Available at https://nbcnews.com/news/us-news/social-media-contributing-rising-teen-suicide-rate-n812426 (last accessed June 29, 2018 ).

10.

Mehta

No is social media contributing to the rise of teen suicides? Available at https://healthify.us/healthify-insights/is-social-media-contributing-to-the-rise-of-teen-suicides (last accessed June 29, 2018 ).

11.

Robinson

, Rodrigues

, Fisher

, Bailey

, Herrman

. Social media and suicide prevention: Findings from a stakeholder survey. Shanghai Arch Psychiatry, 2015; 27:27–35.

12.

Eggertson

. Social media embraces suicide prevention. CMAJ, 2015; 187:E333.

13.

Luxton

, June

, Fairall

. Social media and suicide: A public health perspective. Am J Public Health, 2012; 102(Suppl 2):195–200.

14.

Shah

, Bhandarkar

. The relationship between general population suicide rates and educational attainment: A cross-national study. Suicide Life Threat Behav, 2009; 39:463–470.

15.

Associated Press. Social media blamed for increase in teen suicides in the US. 2017. Available at http://dailymail.co.uk/health/article-5079715/Rise-teen-suicide-social-media-coincide-link.html (last accessed June 29, 2018 ).

16.

Kelly

Teen suicide rate suddenly rises with heavy use of smartphones, social media. 2017. Available at https://washingtontimes.com/news/2017/nov/14/teen-suicides-rise-with-smartphone-social-media-us/ (last accessed June 29, 2018 ).

17.

Negi

Teens turn to social media for attention, even in death. 2017. Available at http://baltimoresun.com/news/opinion/oped/bs-ed-op-0619-social-media-suicide-20170615-story.html (last accessed June 29, 2018 ).

18.

LeBlanc

The text messages that led up to teen's suicide. 2017. Available at http://cnn.com/2017/06/08/us/text-message-suicide-michelle-carter-conrad-roy/index.html (last accessed June 29, 2018 ).

19.

Jiang

, Xu

, Liao

, Zhang

. Analysis of internet suicide pacts reported by the media in mainland China. Crisis, 2017; 38:36–43.

20.

Ueda

, Mori

, Matsubayashi

, Sawada

. Tweeting celebrity suicides: Users' reaction to prominent suicide deaths on Twitter and subsequent increases in actual suicides. Soc Sci Med, 2017; 189:158–166.

21.

Phillips

. The influence of suggestion on suicide: Substantive and theoretical implications of the Werther effect. Am Sociol Rev, 1974; 39:340–354.

22.

Cavazos-Rehg

, Krauss

, Sowles

, Connolly

, Rosas

, et al. An analysis of depression, self-harm, and suicidal ideation content on Tumblr. Crisis, 2017; 38:44–52.

23.

Biddle

, Derges

, Mars

, Heron

, Donovan

, Potokar

, et al. Suicide and the internet: Changes in the accessibility of suicide-related information between 2007 and 2014. J Affect Disord, 2016; 190:370–375.

24.

Daine

, Hawton

, Singaravelu

, Stewart

, Simkin

, Montgomery

. The power of the web: A systematic review of studies of the influence of the internet on self-harm and suicide in young people. PLoS One, 2013; 8:e77555.

25.

Marchant

, Hawton

, Stewart

, Montgomery

, Singaravelu

, Lloyd

, et al. A systematic review of the relationship between internet use, self-harm and suicidal behaviour in young people: The good, the bad and the unknown. PLoS One, 2017; 12:e0181722.

26.

Song

, Song

, Seo

, Jin

. Data mining of web-based documents on social networking sites that included suicide-related words among Korean adolescents. J Adolesc Health, 2016; 59:668–673.

27.

Cash

, Thelwall

, Peck

, Ferrell

, Bridge

. Adolescent suicide statements on MySpace. Cyberpsychol Behav Soc Netw, 2013; 16:166–174.

28.

Barrett

, Shetty

, Broadbent

, Cross

, Hotopf

, Stewart

, et al. “He left me a message on Facebook”: Comparing the risk profiles of self-harming patients who leave paper suicide notes with those who leave messages on new media. Br J Psychiatry Open, 2016; 2:217–220.

29.

Ren

, Kang

, Quan

. Examining accumulated emotional traits in suicide blogs with an emotion topic model. IEEE J Biomed Health Inform, 2016; 20:1384–1396.

30.

Berryman

, Ferguson

, Negy

. Social media use and mental health among young adults. Psychiatr Q, 2018; 89:307–314.

31.

Jashinsky

, Burton

, Hanson

, West

, Giraud-Carrier

, Barnes

, et al. Tracking suicide risk factors through Twitter in the US. Crisis, 2014; 35:51–59.

32.

McSwain

, Lester

, Gunn

. Warning signs for suicide in internet forums. Psychol Rep, 2012; 111:186–188.

33.

Giddens

, Sheehan

. The Columbia-Suicide Severity Rating Scale (C-SSRS): Has the “Gold Standard” become a liability?. ICNS, 2014; 11:66–80.

34.

Suicide Prevention Resource Center. Suicide assessment five-step evaluation and triage (SAFE-T) pocket card. 2009. Available at http://sprc.org/resources-programs/suicide-assessment-five-step-evaluation-and-triage-safe-t-pocket-card ((last accessed June 29, 2018 ).

35.

, Li

, Liu

, Zhu

. Creating a Chinese suicide dictionary for identifying suicide risk on social media. PeerJ, 2015; 3:e1455.

36.

O'Dea

, Larsen

, Batterham

, Calear

, Christensen

. A linguistic analysis of suicide-related Twitter posts. Crisis, 2017; 38:319–329.

37.

Cheng

, Li

, Kwok

C-L

, Zhu

, Yip

. Assessing suicide risk and emotional distress in Chinese social media: A text mining and machine learning study. J Med Internet Res, 2017; 19:e243.

38.

Burnap

, Colombo

, Amery

, Hodorog

, Scourfield

. Multi-class machine classification of suicide-related communication on Twitter. ONSEM, 2017; 2:32–44.

39.

Facebook. Suicide prevention. 2018. Available at https://facebook.com/help/594991777257121/ (last accessed June 29, 2018 ).

40.

Twitter. About self-harm and suicide. 2018. Available at https://help.twitter.com/en/safety-and-security/self-harm-and-suicide (last accessed June 29, 2018 ).

41.

American Psychiatric Association. Using social media to prevent suicide. 2016. Available at https://psychiatry.org/news-room/apa-blogs/apa-blog/2016/08/using-social-media-to-prevent-suicide (last accessed June 29, 2018 ).

42.

Tan

. Designing microblog direct messages to engage social media users with suicide ideation: Interview and survey study on weibo. J Med Internet Res, 2017; 19:e381.

43.

Dissanayake

, Nasr

. Betrayed mood in public view: Taking a MySpace history. West J Emerg Med, 2014; 15:31–34.

44.

Ahuja

, Biesaga

, Sudak

, Draper

, Womble

. Suicide on facebook. J Psychiatr Pract, 2014; 20:141–146.

45.

Chhabra

. Snapchat toxicology: Social media and suicide. Ann Emerg Med, 2016; 68:527.

46.

Bennett

, Pourmand

, Shokoohi

, Shesser

, Sanchez

, Joyce

. Impacts of social networking sites on patient care in the emergency department. Telemed J E Health, 2014; 20:94–96.

47.

Hunter

, Gough

, O'Kane

, McKeown

, Fitzpatrick

, Walker

, McKinley

, Lee

, Kee

. Ethical issues in social media research for public health. Am J Public Health, 2018; 108:343–348.

48.

U.S. Department of Health and Human Services. Information related to mental and behavioral health. 2017. Available at https://hhs.gov/hipaa/for-professionals/special-topics/mental-health/index.html (last accessed June 29, 2018 ).

49.

World Medical Association. WMA Declaration of Helsinki-ethical principles for medical research involving human subjects. 2013. Available at https://wma.net/policies-post/wma-declaration-of-helsinki-ethical-principles-for-medical-research-involving-human-subjects (last accessed August 21, 2018 ).

50.

The National Conference of Commissioners on Uniform State Laws. Employee and student online protection act. 2018. Available at http://uniformlaws.org/Act.aspx?title=Employee%20and%20Student%20Online%20Privacy%20Protection%20Act (last accessed August 22, 2018 )

51.

Laholt

, McLeod

, Guillemin

, Beddari

, Lorem

. Ethical challenges experienced by public health nurses related to adolescents' use of visual technologies. Nurs Ethics, 2018; [Epub ahead of print]; DOI: 10.1177/0969733018779179.