Research shows media coverage of a celebrity suicide has the potential to increase the risk for contagion, especially among vulnerable individuals.
To reduce possible harmful effects, suicide reporting guidelines for media were developed with input from journalists, suicide prevention researchers, and those impacted by suicide.
Research suggests that when guidelines are followed and resources for suicide prevention and stories of recovery are shared, risk for suicide decreases.
A study published today in JAMA Network Open by researchers at Nationwide Children’s Hospital examined suicide reporting guideline adherence by the media for the suicide deaths of Kate Spade and Anthony Bourdain. The study found that some media coverage of the Spade and Bourdain deaths did not adhere to several of the suicide reporting guidelines.
Researchers examined a sample of print newspaper articles from across the United States that covered the Spade and Bourdain suicide deaths.
The articles followed an average of seven out of 14 specific guidelines, and only two of the 14 guidelines were adhered to by all selected newspapers for both deaths (avoid single-cause explanation for suicide death, and avoid referring to suicide as a “growing problem,” “epidemic,” or “skyrocketing”). Further, the study found that two guidelines were not followed at all by any of the selected publications (share a hopeful message that suicide is preventable, and convey suicide behaviors can be reduced with mental health support and treatment).
“Significant strides have been made in certain aspects of suicide reporting, but we still have a lot of work to do,” said Arielle Sheftall, Ph.D., principal investigator with the Center for Suicide Prevention and Research in the Abigail Wexner Research Institute at Nationwide Children’s and lead author of the study.
“Suicide needs to be reported on as a public health issue every single time rather than a story focused on the celebrity’s death and the method of that death. Most important, suicide prevention resources—places where at-risk individuals can seek help—are vital to include in reporting.”
Nationwide Children’s suicide prevention experts recommend that suicide reporting guidelines be reviewed before publication of a story and that the entire newsroom has access to and knowledge of the guidelines – from reporters on every beat, to editors, copywriters, photographers and headline writers.
Ideally, suicide reporting should be done by health reporters who are best positioned to contextualize the issue within the broader topic of mental health.
“When creating these reporting guidelines, we collaborated with journalists and individuals affected by suicide to make sure they were practical and accounted for key perspectives,” said John Ackerman, Ph.D., clinical psychologist and suicide prevention coordinator for the Center for Suicide Prevention and Research and senior author of the study.
“Many reporters in these newsrooms are reporting on suicide in a thoughtful way that respects the victims as well as the public, but other reporters in the same newsroom may not know these guidelines exist or why they are important.”
Because the study was blind, reviewers—who were experts in suicide prevention or news media – were not aware of the type of newspaper outlet or the reporter who wrote the story. Only the headline, body copy and photo were included in the review.
“You never know who is reading an article or which social media outlet has shared an article,” said Dr. Sheftall. “People who may be struggling with suicidal thoughts could be exposed to these articles, and that is why it is so crucial to follow all of the suicide reporting guidelines. We are not pointing fingers at journalists or their newsrooms, but encouraging them to become aware of the guidelines and understanding their nuances.”
“Beyond newsrooms, we can all do a better job talking about mental health and suicide in a responsible way. Parents, educators and community members can find additional resources for how to responsibly talk about suicide at OnOurSleeves.org,” said Dr. Ackerman.
More information: suicidology.org/wp-content/upl … ended-4-merged-1.pdf
reportingonsuicide.org/
Provided by Nationwide Children’s Hospital
————–
Suicide remains a serious public health issue and is a leading cause of death across the world [1,2]. Some research indicates that suicide rates significantly increase following the suicide of a prominent celebrity or peer: a phenomenon known as ‘suicide contagion’, ‘copy-cat suicides’ or the ‘Werther-effect’ [3,4].
Evidence suggests that suicide contagion may be particularly pronounced when media coverage of a peer/celebrity suicide is detailed and uses sensational language [5,6,7]. These studies theorize that some at-risk consumers of such media coverage identify with the celebrity/ peer and their struggles.
This is especially so when the consumer perceives demographic similarity and characterological homophily with the celebrity/peer [8].
Such affinity, combined with sensational and detailed media coverage, may lead some at-risk consumers to conclude that suicide is an acceptable solution to ongoing struggles, or even a heroic and decisive act worthy of emulation.
It is posited that this can increase suicidal behaviour in such at-risk individuals [4,6,9].
Interestingly, this theory is supported by several studies indicating that people who die by suicide after a celebrity/ peer suicide are significantly more likely to use the same suicide methods (and sometimes the same location), the details of which may have been learnt through sensational and detailed media coverage [7,10,11].
This ‘copy-cat’ behaviour suggests some level of affinity with the celebrity/ peer and their actions.
These issues recently made international news thanks to a well-publicized and carefully-crafted epidemiological study examining suicide contagion in the United States (U.S.) following Robin Williams’ suicide [12]. Robin Williams was a much-loved comedian and actor who died by suicide (suffocation through hanging) on August 12 2014. This study examined the monthly suicide count before and after this suicide, finding a 10% increase in suicides in the two months following his death.
A significant increase in the number of suicides by suffocation in men was also observed. The authors speculated that this increase could be partially attributed to detailed and glamourized U.S. media coverage of Williams’ suicide provoking copy-cat suicides, however the authors did not conduct a content analysis to examine media coverage.
In contrast, researchers have systematically examined the tone and content of Canadian media coverage of Robin Williams’ suicide in a focused study [13]. In this study, researchers assessed newspaper article fidelity to evidence-based recommendations for reporting suicide (see Table 1).
These recommendations were produced by the Canadian Journalism Forum on Violence and Trauma, in collaboration with the Canadian Broadcasting Corporation and the Mental Health Commission of Canada, published in a booklet called ‘Mindset’ [14].
Although produced in Canada, the Mindset recommendations overlap with other prominent suicide reporting guidelines, such as those produced by the World Health Organization (WHO) [15] and the U.S. based ‘Recommendations for Suicide Reporting’ [16]–a set of recommendations co-created by numerous stakeholders including U.S. government agencies, universities and non-governmental organizations that are publicly available on a dedicated website.
Common recommendations across these guidelines include ‘avoid language that sensationalizes or romanticizes the suicide’, ‘provide information about seeking help’ and ‘avoid using terms such as ‘commit suicide’ or ‘successful suicide”.
Interestingly, the Canadian research found that journalists strongly adhered to Mindset guidelines when reporting Robin Williams’ suicide [13].
In fact, study results showed that 85% of the newspaper articles included in the study applied at least 70% of the recommendations when covering Robin Williams’ suicide. Notably, 76% of articles did not go into details about the suicide method used and 86% did not use language that romanticized or glamourized the act.
These findings overlap with results from studies in other western jurisdictions. For example, a recent study assessed British media coverage of suicide, finding that 69% of articles did not go into detail about the method used, and that 83% avoided ‘sensationalist or irresponsible language’ [17].
Similar results were found for a study on the Irish media, finding that 80% did not go into detail about the method used, and 88% avoided sensationalized language [18].
In contrast, studies from non-western jurisdictions indicate more frequent violations of suicide reporting guidelines.
In South Korea, a study indicated that only 32% did not go into detail about the method used, though 76% avoided ‘sensational coverage’ [19].
Likewise, a study of the Pakistani media found that 66% of the articles did not go into details about the method used, and 99% of the articles used ‘careless language’ [20].
To our knowledge, there has not been any recent systematic analysis of U.S. media coverage of suicide and adherence to commonly-used suicide reporting guidelines. This is concerning as over 47 000 Americans died by suicide in 2017, and U.S. suicide rates have risen steadily since 1999 [21,22].
Moreover, Fink et al [12] speculate that U.S. media coverage of Robin Williams’ suicide may have been overly sensational and detailed, given the observed increase of suicides after Robin Williams’ death; however, this has not yet been studied systematically. Indeed, it is unknown whether the U.S. media is reporting suicide in a similar way to other comparable countries such as Canada. This is important to know given the national and international influence of the U.S. media and its potential contribution to suicide contagion.
All this raises an important research question: how did the U.S. media report the suicide of Robin Williams?
This research question propels the current study, which aims to formally examine the tone and content of U.S. newspaper coverage of Robin Williams’ suicide through deductive and inductive analysis.
The primary objective is to assess adherence to suicide reporting guidelines in U.S. newspapers after Robin Williams’ suicide through deductive assessment.
The secondary objective is to identify common themes inductively emerging in these newspaper articles which may not be captured by the Mindset guidelines. The tertiary objective is to compare patterns of coverage of Robin Williams’ suicide in the U.S. media as compared to the Canadian media.
Discussion
A key finding of this study is that U.S. newspaper articles about Robin Williams’ death have only moderate levels of adherence to core suicide reporting recommendations.
On the one hand, all articles applied at least 5 of the 12 Mindset recommendations. On the other hand, over half of the articles failed to apply at least 9 recommendations.
Perhaps most importantly, almost half (46%) ignored the recommendation ‘do not go into detail about the method used’, with one-quarter of the articles describing the suicide in an explicit and detailed manner, so much so that ‘highly-detailed description’ emerged as a theme in its own right in the qualitative analysis.
This recommendation to avoid detail about the method used is not only found in Mindset, but also in other guidelines such as the internationally-applicable WHO guidelines and the British Samaritans Media Guidelines for Reporting Suicide [15,33].
Indeed, this recommendation is considered essential for suicide prevention as several studies indicate a link between detailed reporting on a celebrity suicide method and subsequent suicide in the community by such method [5,34,35].
Of note, the 46% of U.S. articles violating this recommendation is considerably higher than that seen in similar media research from other jurisdictions mentioned in the introduction. For example, the media in Ireland (20%), Canada (24%), the U.K. (31%) and Pakistan (34%) had a much lower frequency of violations of this recommendation when reporting suicide per se [13,17,18,20].
Our other key finding was that several prominent themes emerged across the articles which were not adequately captured by the Mindset guidelines.
This included Williams’ struggles with mental illness and addiction, respectful remembrance of his life and work, as well as the need for open dialogue and discussion about suicide and mental health. All three of these themes could be considered helpful from a public health perspective. Adopting such angles has been identified in previous research as an important means of raising awareness of suicide prevention and reducing stigma associated with mental illness in the news [13,25].
Hence, using the tragedy of an individual suicide to open dialogue and discussion about suicide prevention and mental health in a news article could be a candidate recommendation in any future revision of Mindset and other suicide-reporting guidelines [25]. Likewise, utilizing ‘respectful remembrance’ in an article humanizes the deceased individual in a poignant and reflective manner and should be encouraged as part of responsible reporting.
An advantage of our methodology is that it allows direct comparison with the results from an analogous study examining adherence to Mindset recommendations in Canadian newspaper articles about Robin Williams’ suicide [13].
Overall, this comparison indicates that Canadian articles had much higher rates of adherence in comparison to the U.S. articles. For example, almost double the number of U.S. articles went into detail about the method used: 46% of U.S. articles vs 24% of Canadian articles.
Similarly, almost double the amount of U.S. articles (27%) romanticized the act, compared to 14% of the Canadian articles. Also, 27% of Canadian articles gave information about where and how to seek help, almost three times higher than U.S. articles (11%). In total, 85% of Canadian newspaper articles applied 70% or more of the 12 recommendations, almost double the figure (43%) seen in the U.S. articles.
Likewise, the methodology allows for a comparison of qualitative themes emerging from the Canadian and U.S. articles.
Again, differences were observed. In Canada, an emerging theme was ‘the need to open the conversation’, observed in 45% of articles. This is similar to the theme in the present study, ‘need to start conversation about suicide/ mental health’ which was observed in only 24% of U.S. articles. That said, the U.S. and Canadian newspapers shared some similar prominent themes, such as discussion of addictions and the dark side of comedy.
This pattern of reporting may have contributed to the increased suicide rates observed by Fink et al in the months following Robin Williams’ suicide [12]. As mentioned, detailed and romanticized description of celebrity suicides has been liked to suicide contagion in previous studies [5,7,34,35].
The fact that Williams was a beloved national figure, combined with this mixed pattern of reporting, may have negatively affected vulnerable readers who identified with Robin Williams and his struggles, especially in readers perceiving demographic similarity and characterological homophily [8].
Such vulnerable readers may have concluded from newspaper articles that suicide is an acceptable (or even heroic) solution to ongoing struggles. This could have contributed to the observed increase in suicide, however such links are speculative and further real-time research is necessary to explore the relationship between celebrity suicide, associated media coverage and subsequent community suicides.
These findings have numerous public health implications. First, they suggest the need to revise existing suicide reporting guidelines. Notably, there have been some changes to U.S. suicide reporting guidelines since Williams’ death.
The widely-used Associated Press Stylebook now recommends “suicide stories, when written should not go into detail on methods used” [36]. Similar changes could be made to the on-line ‘Recommendations for Reporting on Suicide’ mentioned in the introduction.
The guideline ‘do not go into detail about the method used’ is not explicitly stated in these recommendations, but implicitly alluded to across each guideline. For instance, these guidelines suggest that journalists write “Kurt Cobain dead at 27” instead of “Kurt Cobain used shotgun to commit suicide”, without reference to the higher-order principle ‘do not go into details about methods used’ [16].
In light of our findings and the previously described research on contagion, these recommendations could be re-written to include a specific bullet point ‘do not go into details about the method used’.
Similarly, the qualitative analysis of both the U.S. and Canadian data indicates that many journalists used the tragedy of Robin Williams suicide to open helpful dialogue and discussion about suicide prevention, mental health stigma and wider issues of health care. From a public health perspective, such efforts are welcome as they can raise awareness and catalyze change [25,37].
As such, future revisions of guidelines may include a new bullet point along the lines of ‘do discuss suicide prevention, mental health and health care where appropriate.’ Such a positively-worded ‘do’ recommendation encourages creative freedom among journalists, and may be especially important in guidelines such as Mindset where the majority of existing recommendations are constraining negatively-worded ‘do not’s’.
Second, the results suggest the need for strong and pro-active outreach work with U.S. journalists to encourage them to better follow core suicide reporting recommendations. Results from the present study, as well as related evidence, suggests that in particular, U.S. journalists should be encouraged to
(i) not go into detail about method used;
(ii) avoid romanticizing the act and
(iii) include helpline numbers, on-line resources and further information about where to seek help. Journalists should also be encouraged to continue the observed trend to use suicide reporting as an opportunity to open dialogue and discussion about wider issues of suicide prevention and mental health, which can raise awareness and foster help-seeking [38].
Such action could be inspired by ongoing approaches employed in Canada. In collaboration with numerous organizations, the Mental Health Commission of Canada has created and implemented a number of programs to improve media reporting of mental illness and suicide.
These include
(i) the creation and distribution of Mindset to over 5 000 newsrooms and journalists;
(ii) the creation of a freely-available on-line course on best practice mental health and suicide reporting for current journalists and journalism students; and
(iii) regular visits to major journalism schools to conduct panel-based interactive educational sessions aimed at students and faculty.
Recent research suggests that these activities may have been successful in significantly improving Canadian media coverage of mental illness and suicide in recent years [13, 25, 39, 40].
There are some limitations to this study. First, the study only included ten newspapers, a small proportion of the newspapers published in the United States.
Articles in other newspapers or focused on a specific region may have given different results. That said, we included best-selling national and regional newspapers, which at least give an indication of general patterns in the U.S. suicide reporting landscape. Moreover, the number of newspapers in the present study is similar to that seen in the analogous studies from other jurisdictions mentioned in the introduction [13,17,18,19,20].
Second, we assessed U.S. articles’ adherence to recommendations created and published in Canada (Mindset).
This was done to provide a valid and direct comparison between the two jurisdictions, and the Mindset recommendations do overlap considerably with other suicide reporting guidelines such as those produced by the World Health Organization. That said, Mindset has been widely distributed to Canadian journalists, and not to U.S. equivalents, perhaps explaining greater adherence in Canada.
Third, a single coder (the first author) was involved in the granular-level analysis of all the data.
This involved the exercise of subjective judgment to evaluate the presence or absence of themes, introducing the possibility of observer bias. However we took many steps to reduce such bias.
Firstly, the first author is a highly-trained analyst with considerable experience analyzing media coverage of suicide and mental illness [25,26], who was closely supervised by the second author who helped resolve ambiguities during the various stages of data analysis. Secondly, there was substantial agreement when a sub-sample of articles were double coded by the two authors following training.
Thirdly, the coding and judgment calls were not complex in nature, involving simple yes/no responses to the presence or absence of simple words or themes; such non-complex judgements do not require complex coding procedures [41]. Moreover, using a single well-trained and well-supervised coder in such studies is often considered an advantage as it is cost-effective and reduces inter-rater variability, as long as this is accompanied by a check on reliability from a second coder, which was the case in the present study [32]. Likewise, this approach is commonly-used when analyzing media stories about suicide and mental illness [13,24,25]. Still, there may have been some minor observer bias in the study.
Finally, we focused on a small selection of newspapers within the mainstream media. This focus on one segment of the media (mainstream newspapers) is limiting given the diversity of media in the modern world. Reports in other forms of mainstream media such as television or radio may have given different results. Additionally, analysis of social media is vitally important to better understand suicide contagion [42] but was not conducted in the present study.
Indeed, there is a dire need for further research on the role of social media in suicide contagion as there has been a massive growth in on-line citizen journalism, user-generated content and widely shared social media commentaries on current events [43].
As such, journalists are no longer the sole gatekeepers to knowledge and information about suicide. Improving professional journalists’ reporting of suicide remains critical, however this is one part of an increasingly complex jigsaw. There is a need for further research to understand how to positively affect citizen ‘reporting’ and social media sharing about suicide.
Conclusions
This study suggests that U.S. newspapers moderately adhered to best practice recommendations when reporting Robin Williams’ suicide. Many recommendations were consistently followed, however others were frequently flouted, including core recommendations not to go into details about the methods used and not to romanticize the act. The patterns of reporting in the U.S. media differed to patterns of reporting in the Canadian media, with Canadian newspapers having much higher adherence to suicide reporting guidelines. This pattern of reporting in the U.S. media may have contributed to suicide contagion in the U.S. in the months following Williams’ suicide. As such, new interventions targeted at the U.S. media and journalists may be necessary to improve suicide reporting, which may in turn reduce suicide contagion
