Media depictions of possible suicide contagion among celebrities: A cause for concern and potential opportunities for prevention

Introduction

The well-described phenomenon of suicide contagion or ‘copycat’ suicide, termed the Werther Effect, is thought to arise through social learning whereby a vulnerable person identifies with another person who has died by suicide and emulates his or her behaviour (Niederkrotenthaler et al., 2010, 2012). Research has repeatedly demonstrated that celebrity suicides (i.e. entertainers, politicians and other well-known individuals in the public eye) are particularly likely to be associated with increased subsequent suicide rates due to the high degree to which the general public identifies with and reveres celebrities (Niederkrotenthaler et al., 2012). Lending support to this theory is the corollary finding that publicized suicide deaths of infamous people such as criminals (i.e. people with whom most media consumers will not identify) are unlikely to be accompanied by an increase in suicide rates (Sinyor et al., 2018).

A previous meta-analysis demonstrated that suicide rates in a region increase by 0.26 per 100,000 population in the month following reports of a celebrity suicide (Niederkrotenthaler et al., 2012). Given that the yearly rate of suicide worldwide is just over 10 deaths per 100,000 population, this equates to a nearly 30% increase in deaths over 1 month ((World Health Organization (WHO), 2017). Estimates following the death of Robin Williams in 2014, for example, suggest that there were 1841 more suicide deaths than expected in the United States that year, a 10% increase (Fink et al., 2018). The notion that a single media event could have such a large, negative impact on mortality across the entirety of one of the most populous countries in the world is a finding that deserves greater attention from both public health advocates and those in the journalism community.

Media reports implying suicide contagion between celebrities

While suicide contagion from celebrities to the general public has been well described, contagion of suicide between celebrities has not been an area of focus in the research literature and may present both an important cause for concern but also an opportunity for intervention. This article identifies two recent cases in which media portrayals suggested this kind of contagion and comments on the potential implications for future suicide prevention efforts. Note that the information presented about each celebrity death derives exclusively from reporting in the popular press. On one hand, relying on third party reporting is a significant limitation in terms of potential reliability. On the other, we know that contagion often occurs via the mass media. Therefore, the ‘reality’ of events as presented in the news is relevant as it may have a potential impact on those exposed. Regardless, a note of caution is necessary because suicide is well known to be a multicausal event and media reporting immediately following a death will invariably include only partial information from sources that may be biased or misinformed. The background provided below therefore includes only selected aspects that were publicly reported for the respective individuals, which should be considered incomplete and even potentially inaccurate. These depictions are neither comprehensive nor sufficient to explain individual suicides, yet they illustrate a novel aspect of media reporting on celebrity suicide that has the potential to inadvertently result in harm and therefore should be considered a target of intervention.

Case #1: In May 2017, Chris Cornell, singer and Soundgarden frontman died by suicide by hanging (Media Reference 1). ¹ Chester Bennington, lead singer for the band Linkin Park, had toured with his friend Cornell and sang at his funeral prior to Bennington’s own suicide (Media Reference 2): ‘It didn’t take long for fans to link the tragic news to another devastating event: The death of Soundgarden lead singer Chris Cornell, one of Bennington’s closest friends, who committed suicide almost exactly two months [before Bennington]’ (Media Reference 2). Both men had reportedly struggled with depression as well as drug and alcohol use disorders (Media References 3–4). Shortly after Cornell’s death, Bennington posted the following about him on social media: ‘You have inspired me in many ways you could never have known … Your voice was joy and pain, anger and forgiveness, love and heartache all wrapped up into one. I suppose that’s what we all are. You helped me understand that’ (Media Reference 2). Two months later on what would have been Cornell’s 53rd birthday, Bennington also died by suicide by hanging (Media Reference 4). Coverage of these deaths may represent the most prominent example of media reporting portraying two celebrity friends and contemporaries dying by suicide by the same method in close temporal proximity.

Case #2: As mentioned, the Academy Award winning actor Robin Williams died by suicide by hanging in August 2014 (Fink et al., 2018; Media Reference 5). Williams was known to suffer from depression as well as an alcohol use disorder and there had also been speculation in the media that the comedian suffered from bipolar disorder, although that was something that he had previously denied (Media Reference 5). In June 2018, Kate Spade, the famous fashion designer, died by suicide by hanging using a scarf tied to a doorknob (Media References 6–7). According to her sister, she had suffered from depression, with a question about possible bipolar disorder, as well as problematic alcohol use (Media Reference 7). Spade’s sister noted that she had been ‘fixated’ on Robin Williams’ death at the time and ‘kept watching it and watching it [i.e. media coverage of his death] over and over. I think the plan was already in motion even as far back as then’ (Media Reference 7). As described above, this detail is difficult to corroborate as it apparently came from a single source. Yet, there is further evidence to suggest a possible role of contagion as Spade was the third prominent fashion designer to die by suicide by hanging in the span of 8 years. In 2010, Alexander McQueen died by hanging himself in his closet (Media Reference 8). Notably, McQueen’s career was launched when his clothing line was discovered by the English magazine editor Isabella Blow, who died by suicide herself in 2007 by self-poisoning. In 2014, L’Wren Scott, fashion designer and girlfriend of Mick Jagger, died by suicide also by hanging herself by a scarf tied to a doorknob (Media Reference 9). Finally, world-renowned chef and fellow-New Yorker Anthony Bourdain died by suicide only 2 days after Spade’s death (Media Reference 10). While media reporting did not suggest contagion between Spade and Bourdain, it did focus on the two deaths almost as a single event often within the same article given their close temporal proximity (Media Reference 10).

The suicide deaths described above represent two examples in which media reporting framed suicide deaths in a group of celebrities as possible suicide clustering. Both cases are notable in that coverage emphasized similarities between the deceased in terms of profession, clinical profile (at least according to what has been published in the popular press) and suicide method. Much of the research examining suicide contagion involves association studies. One of the drawbacks of these designs is that frequently they cannot prove exposure. But in this case, reports highlighted anecdotal evidence that both Chester Bennington and Kate Spade were aware of and influenced by the deaths of Chris Cornell and Robin Williams, respectively. Note that the way in which these two cases were presented also highlights the two potential models of suicide contagion through social learning. These are contagion among people that one knows (i.e. Cornell/Bennington) and via media reporting (i.e. Spade/Williams).

By necessity, most association studies also focus on the immediate time period following a widely covered suicide. This model is analogous to the case of Cornell/Bennington in which a second death follows days or weeks after the first. However, it is likely that media exposure can have an influence on certain people years later and, if the reporting of her sister’s assertions are accurate, this may have been true in the case of Kate Spade, who was exposed to and reportedly paid careful attention to coverage of Robin Williams’ death 4 years earlier.

While famous people occupy a social status apart from the average person, there is no reason to suspect that they are immune to social contagion. A wealth of research suggests that celebrities are strong vectors for transmission of suicidal behaviour to non-celebrities (Niederkrotenthaler et al., 2012) and the reports described here suggest that this may be true between celebrities as well. Whether or not media reports emphasizing contagion among celebrities are factually accurate, their existence and broad dissemination have important implications for public health and suicide prevention. Given the repeated finding that individuals emulate suicide after reports of a single celebrity death, there should be concern that reports of clustering of suicide deaths among multiple celebrities may lead to an enhanced Werther Effect and establish suicide clustering as a normative social event. For example, we are aware of anecdotal reports of suicide contagion among young rock music fans in multiple countries following the Cornell and Bennington suicides, although this has never been rigorously studied. If accurate, it would lend support to the concerns raised in this article. Furthermore, there is a potential for harm in any group that identifies with the celebrities who died, including those in similar age groups, sexes and professions. The question of how to prevent this from happening is a challenging one.

Potential efforts to mitigate and prevent harm

There are several potential options that would need to be thought out carefully before implementation. The first is to alert news media organizations of the potential harms of drawing connections between celebrity suicide deaths. This could be added to future updates of responsible reporting guidelines. Ideally the media would avoid drawing such connections, however, if they judge these details as a true and necessary aspect of a report, at a minimum journalists ought to be even more careful than usual in establishing the veracity of such claims and to take caution to avoid single sources or mere speculation. Regarding events that happen in close temporal proximity, existing media recommendations already emphasize that separate suicidal acts should not be reported in the same article (Niederkrotenthaler et al., 2010), a recommendation that may require further emphasis in discussions with media. The suicide prevention community could also engage proactively with celebrity unions, associations and professional organizations to develop action plans to be deployed in the event of a suicide in one of their members. For example, resources already exist to promote mental health in actors (Media Reference 11) and are being developed for other celebrity organizations such as the National Basketball Association in North America (Media Reference 12). These existing platforms could educate members who are active on social media or in the mainstream press following a death about recommendations for responsible media reporting (Media Reference 13). Of particular importance would be to encourage them to distinguish carefully between positive messages about the deceased and the tragedy of the death/its potential to have been avoided if help had been sought. Furthermore, crisis and mental healthcare resources could be disseminated among the celebrities themselves. Organizations could also be encouraged to solicit personal stories of help seeking and resilience from their members to be disseminated concurrent with coverage of a suicide death in order to promote the message that suicidal crises can be and are overcome.

One of the notable aspects of the suicide deaths described in the cases above is that all garnered substantial media attention, which may present an opportunity to enhance the ‘Papageno effect’, that is, contagion of positive coping (Niederkrotenthaler et al., 2010). More focus could be placed on celebrities who were able to successfully manage their suicidal ideation. For example, Lady Gaga spoke out at a recent awards ceremony honouring her mother stating ‘I have struggled for a long time, both being public and not public about my mental health issues or my mental illness. But, I truly believe that secrets keep you sick… Thank you Mom for not being afraid of my darkest thoughts and for doing what many don’t realize goes very far: just holding my hand… Suicidal ideation feels like a spell and we have to have empathy. Be kind and help each other break that spell and live and thrive’ (Media Reference 14). There are many other examples of some of the world’s most famous people finding paths to resilience when faced with suicidal thoughts, including J.K. Rowling (Media Reference 15) and Oprah Winfrey (Media Reference 16). While these stories already exist in the popular press, they could be deployed more strategically in an effort to combat potential harm following a celebrity suicide death and enhance a Papageno effect both among celebrities and members of the general public who identify with them.

Conclusion

The potential phenomenon observed in this paper deserves the attention of researchers, policymakers and mental health experts who should take a cautious approach in both interpreting these events and in contemplating a helpful response.

Supplemental Material

supplementary_references – Supplemental material for Media depictions of possible suicide contagion among celebrities: A cause for concern and potential opportunities for prevention

Supplemental material, supplementary_references for Media depictions of possible suicide contagion among celebrities: A cause for concern and potential opportunities for prevention by Mark Sinyor, Marissa Williams and Thomas Niederkrotenthaler in Australian & New Zealand Journal of Psychiatry