Netflix’s show 13 Reasons Why, which focuses on teen suicide, has been linked to increased rates of suicide in a recent US study. Despite such controversy, Netflix has recently announced that it will be releasing a fourth series. But Professor Chris J. Ferguson says the data does not confirm such a finding and his research on fictional suicide media more generally demonstrates that such a link is not supported. [Header image credit: tedd4u, CC BY 2.0]
A study last year, funded by the US National Institute of Health (NIH), linked the show 13 Reasons Why to increased suicides in teens. The press release for the study claimed, “The findings of this study add to a growing body of information suggesting that youth may be particularly sensitive to the way suicide is portrayed in popular entertainment and in the media.” This study appeared to confirm widespread fears that fictional media can cause suicide contagion in youth. Only, not so fast, a close look at the data finds the case for contagion is not that strong after all.
If 13 Reasons Why caused suicide contagion, we’d expect it to be particularly strong in girls and young women, who would be more likely to model the female protagonist. We’d also expect to see a continuous influence in the months after the show’s release, given its constant availability in streaming format, and given not all suicide decisions are instantaneous. Lastly, we’d expect to see at least some increase in suicide among young adults, for whom the show was also popular.
However, the pattern of results from the NIH study didn’t show any of this. In fact, suicides for girls showed no increase. Rather male suicides increased, but only some of the time. Suicides were already increasing before the show came out. The authors attribute this to promotion, but this seems a stretch. Only three of the nine months following the show’s release showed an increase in suicide (one of which was December, seven months after the show’s release), and for girls, suicide rates actually decreased for one of the study months. The month after the show’s release (May) showed no increase in suicides. No effects were seen for young adults for any month, despite that we’d realistically expect to see some carry over into this age group.
This pattern of results is, frankly, a mess. These data should not have been selectively interpreted as demonstrating even correlational effects. Yet, in the press release and study the authors use causal language, which is inappropriate. The study does not adequately control for seasonal patterns in suicide, nor a year-to-year increase over the last few years, which has occurred for almost all age categories, not just teens. Nor does the study control for other events which may have influenced male suicides such as the suicide deaths of several popular male performers during the same period. There are good theoretical reasons to believe people are more likely to model real humans than fictional ones.
Other recent research has cast doubt on the suicide contagion by fictional media hypothesis. A study released almost simultaneously by the University of Pennsylvania revealed more nuanced effects. Overall, the show’s effect on most viewers was minimal, but for those with elevated suicide risk who watched the entire season, suicide risk decreased and willingness to help other suicidal individuals increased. However, more negative outcomes were found for at-risk individuals who stopped watching the show.
In early 2019, I published a meta-analysis of studies examining suicide-themed fictional media. Although the results of individual studies certainly vary, overall the body of literature was unable to support beliefs in fictional media suicide contagion.
Effects of 13 Reasons Why may be very individual, helping some and not helping others. Of course, the same might be said about almost any media. Making global statements of harm, particularly when supported by only wonky and inconsistent data is not helpful.
I don’t doubt the good intentions of the authors of the NIH funded study. However, they ignored red flags in their own data to promote a narrative of harm that can’t be supported. This follows a pattern of ecological fallacies (accidental correlations that occur between things that are unrelated like Nicholas Cage movies and swimming pool deaths) that have sometimes misinformed the public about media effects. Famously, rising crime in the 1970s promoted beliefs that media violence caused crime, beliefs that ultimately proved incorrect. For the moment, despite the scary headlines of last year, the evidence linking 13 Reasons Why to actual suicides in real life remains weak.
Post Script: Since this article was written in January 2020, a reanalysis of the NIH data was released. This reanalysis found that the original NIH study was likely in error and, with proper controls for seasonal and yearly patterns in suicide, 13 Reasons Why was not associated with an increase in suicides for either girls or boys.
This post was originally published in the Orlando Sentinel and has been reposted here with permission.
This post gives the views of the authors and does not represent the position of the LSE Parenting for a Digital Future blog, nor of the London School of Economics and Political Science.