Intended for healthcare professionals

  1. David Gunnell, professor1 2,
  2. Lucy Biddle, senior lecturer1 3
  1. 1Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, UK
  2. 2National Institute of Health Research Biomedical Research Centre at the University Hospitals Bristol NHS Foundation Trust and the University of Bristol, Bristol, UK
  3. 3National Institute for Health Research Applied Research Collaboration West (NIHR ARC West), University Hospitals Bristol NHS Foundation Trust, Bristol, UK
  1. Correspondence to: D Gunnell D.J.Gunnell{at}bristol.ac.uk

Detail, sensationalism, and accounts of the method used are unnecessary and harmful

News coverage of television presenter Caroline Flack’s recent death by suicide once again raises public health concerns about media reporting of suicide. Freedom of the press is one of the fundamental pillars of a democratic society, but is regulation around some aspects of the media reporting of suicide required?

Niederkrotenthaler and colleagues’ systematic review in The BMJ (doi:10.1136/bmj.m575)1 suggests that the answer could be “yes.” The team synthesised findings from 31 studies investigating associations between media reporting of deaths by suicide and population suicide rates. Most studies looked at the reporting of deaths of celebrities by suicide. The phenomenon has been studied in Europe, Asia, North America, and Australia, highlighting the international extent of concerns. Suicide rates increased by 13% (95% confidence interval 8% to 18%) on average in the period (median 28 days) following media reports of the death of a celebrity by suicide.

This effect is substantial. In the United Kingdom, where 6507 people died by suicide in 2018 (542 per month),2 a 13% increase would amount to around 70 additional deaths. In the five months following the death of the international celebrity Robin Williams by suicide, deaths by suicide increased by almost 10% (n=1841) in the United States.3

It could be argued that producing one summary estimate of risk from such a diverse range of studies is misleading. There was marked variation between study findings and evidence of publication bias. Further, any association between the reporting of a celebrity’s death and wider suicide rates will be confounded by the popularity of the celebrity, which influences both the extent of media reporting and the reader’s emotional connection to the death. Even so, the estimate reported by Niederkrotenthaler and colleagues will help give media outlets a clearer sense of the potential effect of their reporting.

Media reporting of suicide methods is a particular concern, because it risks making specific methods “cognitively available.” The authors found that reporting the method of suicide used by a celebrity was associated with a 30% increase in deaths by the same method. This proportion is consistent with studies in which survivors of serious suicide attempts describe media reports as influencing their choice of method.4 If reporting of suicide methods promotes a shift from less lethal to more lethal methods, this could increase overall suicide rates. Suicide attempts are often one-off events; those who survive can receive appropriate treatment for their suicidal thoughts and behaviour; those using a highly lethal method get no such opportunity.

Although the review found no increase in the rate of death by suicide following media reports of non-celebrity suicides, this is not grounds for complacency. A US study5 found evidence that news reporting of suicides could trigger suicide clusters in young people; a higher risk was associated with front page reporting, description of the suicide method, and detailed accounts of the suicide. These concerns are perhaps particularly salient in relation to recent high profile news reporting of deaths of students by suicide in the UK.6

Concerns about media reporting of suicide are not new. In 1841, the British epidemiologist William Farr wrote: “no fact is better established in science than that suicide . . . is often committed from imitation”.7 Farr called for an end to the reporting of “dramatic tales of suicide”; a call in keeping with current guidance from the Samaritans and World Health Organization.89

Media coverage of suicide is important. Suicide is a leading cause of premature death and responsible reporting can lead to greater public understanding. Indeed, some reporting (such as descriptions of people overcoming a suicidal crisis) could have a beneficial effect on suicide rates.10 A tension, however, remains between the so-called public interest and the interest of the public. Detail, sensationalism, and accounts of the method used are unnecessary. Reports of deaths of celebrities from heart attacks do not report which coronary artery was occluded.

Easy access to online information presents particular hazards. People can now read a news story about a method specific death by suicide, and then learn how to use that method from sites such as Wikipedia.11 Hits on Wikipedia’s page about suicide by hanging increased sixfold on the day when details of Caroline Flack’s death were released.

Following the death of a celebrity by suicide, potentially harmful information could flood into individuals’ news feeds as it is reposted, begins trending, and is commented on. Exposure is repeated and accompanied by a new layer of unregulated social discourse on suicide, the effect of which is not easily quantified but likely contributes to Niederkrotenthaler and colleagues’ suggested explanations for a rise in suicide rates following media reporting: increased identification with the deceased person, and normalisation of suicide as a solution to adversity.

Journalists, news editors, and social media platforms must be made to consider more carefully the costs to population health, and impacts on families and friends,12 of sensationalist, detailed reporting of these tragic deaths.

Acknowledgments

We thank Jacks Bennett for comments on an earlier draft.

Footnotes

  • Competing interests: The BMJ has judged that there are no disqualifying financial ties to commercial companies. The authors declare the following other interests: DG is a member of the Department of Health’s National Suicide Prevention Strategic Advisory Group (England), and is a member of Samaritans Policy, Partnerships and Research Committee and Movember’s Global Advisory Committee.

  • Provenance and peer review: Commissioned, not peer reviewed.

References

Log in

Log in through your institution

Subscribe

* For online subscription