Why look to social media in youth suicide prevention work?
Suicide among young people in France
1Although the number of cases has fallen substantially over the last decade or so (by 30 percent between 2006 and 2016), suicide remains the second leading cause of death in France among young people aged between fifteen and twenty-five [1]. According to the most robust declarative survey data, more than one in ten high school students report having experienced suicidal ideation in the past twelve months [2], and almost 10 percent state that they have already tried to take their own lives [3]. Adolescent suicide attempts are among the strongest indicators of psychological distress and mental health disorders [4], but they are also associated with a higher risk of dying as a result of suicide, reporting a mental or physical health problem, exhibiting violent behavior, suffering from social alienation, and requiring social service support in adulthood [5].
Risks associated with exposure to online content
2The digital revolution set in motion almost thirty years ago has profoundly reshaped the youth suicide prevention landscape. Today, more than 2.5 billion people all over the world are connected through social networks [6]. In France, young people aged between fifteen and twenty-four spend over thirty minutes each day on social media [7]. As the lines between on- and off-line social spaces have been continuously redrawn, we have seen changes in the ways young people seek out information, communicate, present themselves to others, and interact with peers [8]. These new forms of socialization have contributed to the emergence of new threats to their mental health, unprecedented in both nature and scope. For example, the scientific community has studied how social media is used as a platform for broadcasting suicidal intentions and acts of self-harm, dramatized through emotionally charged iconography or videos [9–11]. This harrowing content is difficult to stamp out, and can act as a “hook” for the most vulnerable young people, who may see it as a model or an outstretched hand beckoning them down the same path [12]. Some studies have also pointed to a more overt malevolence in some corners of the online community. This may take the form of normalizing suicidal behavior or encouraging vulnerable individuals to hide their actions [13, 14], “trolling” (deliberately making incendiary remarks to derail conversations), or inciting suicide through online “challenges.” Between 10 and 40 percent of adolescents are reported to have experienced cyberbullying [15], found to double the risk of a suicide attempt [16]. As well as the actions of a few ill-intentioned users, vulnerable individuals can be pushed closer to the brink by the very structure or functional organization of these virtual networks. Those who post suicidal messages online tend to be connected to one another through a dense affiliate network (“liked” pages, “friends,” and so on) or through interactional activities (such as “retweets,” “likes,” and “comments”). Within these “suicidocosms,” characterized by strong homophilic tendencies (in other words, where users choose to identify with one another based on perceived commonalities), information circulates endlessly at tremendous speed, and is often highly emotionally charged. This gives rise to a phenomenon of collective rumination and reciprocal identification that can sow the seeds of suicide contagion [17].
The service desert
3In young people with suicidal ideation, access to specialist care is all the more vital given that, in this age group, suicides and attempted suicides are almost always associated with a mental health disorder [4]. Yet, there is a grave deficiency when it comes to the provision of targeted support. At the international level, it is believed that only a quarter to a third of young people presenting with signs of psychological distress receive appropriate care [18].
4This “service desert” is partly attributable to the fact that adolescents and young adults may be reluctant to seek the help they need or simply incapable of doing so [19, 20]. There are two sets of obstacles to consider here [21]:
- Personal obstacles are those that prevent young people from understanding the difficulties they are facing, articulating their suffering, and/or actively seeking support [22]. These obstacles can be all the more debilitating for those in deep distress, who may display signs of “denial,” refusing to acknowledge that they need help [23]. To a certain extent, personal obstacles are linked to the typical developmental changes we undergo in adolescence. This applies, for example, to “self-sufficiency bias,” which can cause young people to play down their need for support and overestimate their capacity to cope with their problems on their own [24]. The stigma attached to suicidal ideation, mental health problems, and psychiatric care is another example of a personal obstacle, and a particularly harmful one at that. Finally, beliefs and worries about treatments and mental health professionals, particularly in relation to confidentiality, trustworthiness, and competence, are especially inhibiting in young people [21, 23, 25]. This age group has a strong desire for emancipation and self-identity, and this can make individuals susceptible to perceive adult support as a form of subordination, alienation, or narcissistic threat.
- Structural obstacles have to do with factors such as the accessibility, cost, visibility, or practicality of support services. Long waiting lists for a first consultation, travel distances, the costs of services not covered by insurance, the complex maze of treatment options available, and the fact that young people often find it difficult to ascertain what is causing their distress are just some of the obstacles that can deter them from seeking help.
Is social media a solution to the service desert?
5In parallel to their off-line struggles to communicate their symptoms and take the first steps toward professional support, the last decade of literature suggests that young people often turn to the online social sphere in an effort to construct a new relationship to their psychological distress and their attempts to find relief. Carrying out online searches on mental health topics, sharing experiences to elicit peer support [26–28], or posting messages signaling an impending suicide attempt [29] have all become common practices among the “millennial” generation. [1] Because of the specific codes of online interaction and communication, young people who keep their suffering to themselves in the “real” world seem to have found their voices on social media, driven by complex motives that have to do with self-representation, help-seeking, and the need for affiliation [8]. In France, there is no other structured online resource for people in distress besides the crude alert systems offered by certain platforms [30, 31]. However, a series of authors have highlighted an untapped opportunity to harness social media as an access point to care for children and young adults experiencing suicidal thoughts [32, 33]. From a theoretical perspective, social media platforms present a number of qualities that could help circumvent the traditional obstacles to help-seeking described above. These qualities are summarized in Table 1.
Possible contribution of various social media communication and interaction norms to overcoming structural and personal barriers to help-seeking
Feature | Possible contribution to overcoming barriers to help-seeking |
---|---|
Round-the-clock availability | |
Platforms offer immediate, uninterrupted access, creating a virtually continuous opportunity to view and share content and to interact with other users [33]. | Overcoming structural barriers Can offset barriers associated with limited service hours and the lack of flexibility inherent in face-to-face care settings. Overcoming personal barriers Potential for more agile and timely responses, reflecting the minute-to-minute volatility of suicidal ideation in terms of severity and the motivation to seek support. Accommodates the immediacy demanded by executive dysfunction associated with suicidal thoughts and behavior. |
Spatial omnipresence | |
Accessible from any location, especially now that smartphones and mobile devices are so ubiquitous. | Overcoming structural barriers Can offset barriers linked to young people’s limited mobility, physical distance, and/or regional inequalities. Allows access to a potentially unlimited range of resources, regardless of geographic catchment area. Overcoming personal barriers Assures a level of discretion and freedom that can allay fears of social stigma and/or what friends and family may think. |
Zero cost | |
The most popular social media platforms are free for anyone to access and use. | Overcoming structural barriers Can offset barriers linked to the financial and organizational burden of accessing treatment, notoriously challenging for young people with limited resources and/or who are financially dependent on others. Overcoming personal barriers Conforms to the freeware model prevalent in online services. Lessens the motivational cost associated with financial cost, by making it easier for users to engage. |
Asynchrony | |
Messages shared on social media platforms are exempt from the implicit time pressure imposed by social norms governing turn taking in interactions. The user is free to respond in his or her own time [35]. | Overcoming structural barriers Allows the continuity of conversations to be maintained, by adapting to the rhythms of off-line life (for example, by accommodating mealtimes, classes, interruption by a third party, etc.). Overcoming personal barriers Helps circumvent mentalization, reflection, or verbalization difficulties that can be accentuated in the high-pressure context of a face-to-face interaction. Allows the conversational dynamic to be adapted to psychopathological processes that can impair coherent reasoning and engagement (psychomotor retardation, difficulty focusing, confusion, etc.). |
Impression of control | |
Having the ability to choose whether to respond to a message and the most opportune moment to do so, or to leave a conversation at any point, gives users a sense of control over their digital interactions [36]. | Overcoming personal barriers This sense of control is also a constitutive component of good mental health, and a necessary condition for engaging in outward-facing or subjectively risky behavior. Can help reassure young people who struggle to cope with demanding interpersonal interactions in off-line life. |
The public/private continuum | |
Certain social media interfaces are viewed as safe spaces or spaces of self-representation, while still being at least partially accessible to the digital community [37]. | Overcoming personal barriers Social media blurs the distinction between public and private, and this can unburden users of the need to define the purpose of their interactions. Messages can be sent to peers under the guise of a personal confidence. |
Dissociative pseudonymity or anonymity | |
Offers users the option of masking their identity with a virtual substitute that they entirely control, which may be more or less consistent with their real identity [38]. | Overcoming personal barriers Can help assuage the sense of vulnerability associated with sharing one’s inner life. Reduces the denotative weight of verbalizations (the things users say do not have to mean anything about their own worth, opinions, or character). This can lessen the personal commitment implied by speaking out. It alleviates both the risks of stigmatization linked to social exposure and fears that their confidence might be betrayed. |
Invisibility or immateriality | |
Social media platforms allow users to decide how much of their own physicality to reveal, by selecting the channels through which they wish to interact—voice and/or written word, and/or a photograph, and/or video, etc. | Overcoming personal barriers Neutralizes fears of being judged, rejected, or stigmatized because of one’s physical appearance (fears often exaggerated in those with low self-esteem). The absence of direct paraverbal cues relieves users of certain burdens of face-to-face communication, particularly in terms of the need for spontaneity. |
Possible contribution of various social media communication and interaction norms to overcoming structural and personal barriers to help-seeking
The Elios project
An ambitious undertaking
6The Elios project (Équipe en ligne d’intervention et d’orientation pour la prévention du suicide/Online Intervention and Referral Team for Suicide Prevention) marks the realization of an ambitious undertaking: to offer mental health professionals the opportunity to harness social media as a whole new site of suicide prevention. By creating a space with its own modes of interaction and communicative codes, the project provides a vehicle for drawing on digital technology’s potential to address the service desert facing young people experiencing suicidal thoughts or behavior. Founded on the pragmatic, evidence-based premise that millennials find it easier to make contact with mental health support through social media than by telephone or a face-to-face consultation, Elios aims to offer a true digital portal, where users can find support and be referred for appropriate treatment through a process designed to overcome the main barriers to help-seeking identified in the literature.
A clinical system
7Elios has a team of online clinicians (psychologists and nurses) who can be easily contacted on social media by any young person experiencing suicidal ideation. The typical usage scenario is depicted in Figure 1. The Papageno Program [2] will promote the system through an extensive publicity campaign that will run online, in the media, and with institutions that provide care for adolescents and young adults. As soon as a young person feels the slightest urge to seek help, he or she can contact the service simply by sending a message. Initially, individuals who contact Elios will be welcomed by a chatbot, which will provide encouragement, collect some useful initial information, and connect them with the team of online clinicians as soon as they begin work for the day. The team will then invite them to take part in a support program guided by two primary goals: relieving distress and suicidality, and ensuring a quick referral to mainstream mental health services if necessary. This program combines three complementary techniques:
- Support is designed to open up a reassuring relational space where online clinicians will try to establish trust, mainly by offering simple advice and encouraging users to verbalize their feelings.
- Motivational guidance is intended to resolve any ambivalence they may feel about seeking help, by tackling personal barriers and helping them channel their resources toward committing to treatment.
- Crisis intervention will be adapted to the urgency of each individual situation. For young people at low risk of attempting suicide, the priority will be to get them referred for treatment swiftly, by addressing structural barriers to accessing support. Online clinicians will determine the most appropriate organizations and health professionals in each case before reaching out to them proactively, providing them with details about the situation if necessary, and putting them in touch with the individual concerned. For those considered to be at a medium or high risk of suicide, the objective will be to initiate emergency or semi-emergency interventions to ensure their safety, defuse the suicidal crisis, and get an appropriate treatment package in place as quickly as possible. Crisis intervention measures will be underpinned by the principles of immediacy, brevity, and directivity [34].
User pathway and assessment in the Elios system

User pathway and assessment in the Elios system
A typical usage scenario is presented here in chronological order, moving from left to right: 1. The user (here, a young adult experiencing suicidal thoughts) learns about the system through the Elios public website or the promotional social media campaign led by the Papageno Program. 2. The user makes contact by sending a message to Elios’s account from the social media platform of his or her choice. 3. The Elios “chatbot” immediately responds, directing the user to a secure area of the Elios website. 4. Once there, the user enters his or her details, gives consent to participate in the study subject to meeting the inclusion criteria, and completes a self-administered questionnaire, as indicated by the protocol (M0 assessment). He or she will then be randomly allocated to one of two groups. 5. If allocated to the experimental group, the user will be put in touch with an online clinician, who will interact with him or her on the preferred social media platform. If allocated to the control group, he or she will be given details of support resources, without any further action. 6. Users in the experimental group will have an initial consultation with an online clinician, regular check-ins, and the option to contact the clinician directly for a period of three months. 7. Once three months have elapsed, all users will be contacted once again by the Elios chatbot and directed to the secure area of the Elios website. Here they will complete another self-administered questionnaire, as indicated by the protocol (M3 assessment).8The options available to online clinicians will necessarily be subject to the chosen platform’s inherent restrictions on communication. While providing an easy way to make initial contact, text interactions pose considerable obstacles to data collection and limit the options for intervention. Once the dialog is under way, online conversations should therefore guide the young person toward a more flexible and free-flowing communication channel (such as a telephone call, video call, or face-to-face consultation). This means that online clinicians are asked to negotiate a perpetual tension: the more the chosen form of communication constrains the scope for assessment and intervention, the more challenging it will be to get to know one another and establish an effective relationship [32]. The subsequent clinical approach must strike a delicate balance between maintaining trust and encouraging the user to overcome the barriers that led him or her to seek help through social media in the first place.
9Given the common ground between their respective missions, and with a view to sharing resources and expertise, the Elios system will be embedded in the VigilanS center [3] at the Centre hospitalier régional universitaire de Lille (CHU Lille) (Lille University Hospital Center). Specifically, a member of the VigilanS team will step in once a day as an online clinician, in addition to his or her usual role. To facilitate this, we will be increasing staffing levels on the Lille VigilanS team and providing training in the additional skills required to provide assessment, intervention, encouragement, and guidance through digital channels. In this way, we aim to preserve the distinctive character of each team while pooling their human, infrastructural, and logistic resources.
An evaluative research protocol
10To ensure the Elios system’s rigor, effectiveness, and safety, it will be rolled out initially under a research protocol designed to meet the very highest standards for evaluation reliability. The various stages of the protocol are outlined in Figure 1.
11The study’s main purpose is to demonstrate that the Elios system is more effective over a three-month period in reducing suicidality among adolescents and young adults who seek online support for suicidal ideation than simply referring them to other professional resources. To do so, Elios will be put to the test in a randomized controlled study, using two parallel intervention groups of equal size. It is expected that 386 young people (aged between eighteen and twenty-five) experiencing suicidal thoughts and behavior will consent to take part. Young people allocated to the control group will be given a list of approved resources for professional support, similar to the kind of list the most skilled might compile for themselves through independent online research. Those allocated to the treatment group will be put in touch with the online clinical team and provided with further support, as described above (“A clinical system”). All participants will be interviewed at the point of onboarding and again three months later, to assess their levels of suicidality. The study will be carried out under the authorization of an ethical review board and the Commission nationale de l’informatique et des libertés (CNIL) (National Commission on Information Technology and Liberty).
Digital support tools
12Elios will incorporate various innovative technological solutions to support and enhance the work of online clinicians and researchers. These will be developed through an iterative, multidisciplinary codesign process involving engineers, clinicians, and researchers. Focus groups will be carried out to gain insight into the expertise and practices of prospective online clinicians (members of the VigilanS team) and the needs and expectations of young people who might use the service. The information gathered will be fed into the next round of improvement. From the earliest design phase, Elios will adopt a rigorous approach to data security in compliance with General Data Protection Regulation (GDPR) and the applicable medical and legal regulations for clinical trials involving human subjects.
13Ultimately, the system will be built on three digital components:
- A digital platform. Designed as a true gateway to clinical services, the Elios platform will be built on a digital interaction application that will coordinate information streams from Elios accounts on various existing social media platforms. This will provide online clinicians with a secure, centralized, simple, and intuitive way to manage their interactions with the supported young person, through a dashboard offering a range of functionalities (Figure 1). In the first instance, they will be able to access a dynamic list and smart search tools allowing them to locate each user’s private space with ease. These spaces will consist of a standard “chat” interface where they can interact with the young person concerned, regardless of the social media platform he or she is using. Users will also be able to switch from one platform to another without interrupting the flow of the conversation. Over the course of their initial dialog, the online clinician will complete an ergonomically designed “clinical record” for each user, to help them determine the right clinical approach. This record will contain a variety of information, primarily sociodemographic data, personal and family history, past suicide attempts, previous clinical evaluations, and, if appropriate, a personalized safety plan. A systematic assessment of the urgency and level of suicide risk presented by the individual will feature prominently on each record. As well as granting access to each user’s private space, the dashboard will offer online clinicians a semiautomated planning tool to help them manage the contacts they need to make, prioritized according to each young person’s clinical situation. They will also find a display panel showing general usage statistics and a digital mailbox to notify them of professional correspondence and any adverse effects arising from the protocol. Finally, the dashboard will also enable authorized personnel to extract anonymized data for research purposes.
- The Elios website. The Elios website will gather assessment data in accordance with the research protocol through a dedicated secure area (see “an evaluative research protocol” above). The public interface will also serve as a communicative tool, ensuring that the system has a visible web presence.
- Automated chatbot. The Elios “chatbot” will allow the system to deliver a smooth, stand-by service with twenty-four-hour availability, so young people in need of support can receive an immediate automated response. Depending on whether the individual is participating in the research study and the time of day when contact is made, the chatbot will put him or her in touch with the Elios team or provide contact details for available services.
Discussion
14Currently under development, the Elios system is scheduled to launch in early 2021, when recruitment will begin for the clinical trial designed to assess its effectiveness. If proved effective, Elios could be introduced permanently as a suicide prevention and treatment access system within France’s conventional health care system. It could also be opened up to minors or adults over the age of twenty-five and to other situations besides suicidal crisis, such as psychological distress with no suicidal ideation.
15Elios will therefore be the very first preventative mental health support tool specifically designed to respond to the ways adolescents and young adults interact on digital platforms. It will provide a truly innovative, low-carbon, and intuitive solution for millennials struggling with suicidality. By offering immediate, interactive support, accessible from anywhere and at any moment, it will allow the health care system to harness whatever motivation to seek help that sufferers may have, and to lay the groundwork for an ongoing relationship with mental health services. Moreover, the system is a first for France in that it presents a positive alternative to the damaging content to which young people can be exposed online, by opening up a more promising avenue for support.
16In cases of suicidal behavior, Elios is rooted in a preventative approach that is currently underexplored by both mental health and psychiatric professionals and the public health system. The hope is that, thanks to its unlimited accessibility, young people experiencing suicidal ideation will take to it with the same spontaneity and ease they show when articulating their distress online. Elios will therefore expand the reach of youth suicide prevention beyond conventional prevention strategies. It will also cover the full spectrum of at-risk young people, from those who, aware that they are experiencing the first symptoms of suicidal ideation, do not feel that their case is serious enough to seek medical attention, to those who are on the verge of a suicide attempt and are making a last-ditch cry for help.
17Inevitably, the system’s very originality means that its development must venture into some untested regulatory and ethical terrain. All of the perennial questions relating to confidentiality, user safety, and data ownership come to the fore here in high definition, given the sensitivity of the data to be produced and collected. Both the research study and the operation of the system itself will be carried out within a robust legal and regulatory framework from the earliest stage in its development, in accordance with privacy/security-by-design principles, under authorization from the ethical review board and the CNIL, and in full compliance with the GDPR. There are a number of ethical considerations that traditional regulatory bodies have rarely encountered before, and these remain to be resolved. One of these issues, for example, concerns the use of digital platforms to collect sensitive personal data in a context where the subject’s life is at risk. In this respect, the Elios system is being developed in keeping with a firm commitment: to never leave a young person in great distress alone and unsupported in the face of online dangers, even if this means committing a limited and proportional breach of strict data confidentiality provisions.
Declaration of interests
18The authors have no interests to declare in connection with this article.
Notes
-
[1]
This term refers to those born between the early 1980s and the late 1990s—young enough when computers became widely accessible to have developed an intuitive fluency with technology and assimilated it into their everyday lives.
-
[2]
Nationwide program to prevent suicide contagion and raise awareness of the support and treatments available in France. Website: www.papageno-suicide.com.
-
[3]
Monitoring and remote intervention system aimed at preventing further suicide attempts among those released from hospital following a suicidal episode (www.dispositifvigilans.org).