Conflicting research released last month gave us a very unclear answer about whether simply watching or being exposed to a television show about teen suicide — Netflix’s 13 Reasons Why (13RW) — results in an increase in actual teen suicide in real life.
One study found a correlation (not a causal relationship) between the two, while another study found declines in suicidal thoughts and self-harm behaviors.
So what’s the real story?
13 Reasons Why is a Netflix television series that explores the first-hand account of a fictional teenage girl’s life and eventual suicide. The second season delved deeper into the aftermath of all the people touched by and potentially responsible for the teen’s suicide. Researchers, psychologists, parents, and experts were all concerned that after the release of the first season of the show, teens would be inclined to greater suicidal and self-harm behavior.
The media suicide effect (or Werther effect as coined by David Phillipps in 1974) has been well-researched, as summarized by Hawton & Williams (2002):
The impact of the media on suicidal behaviour seems to be most likely when a method of suicide is specified—especially when presented in detail—when the story is reported or portrayed dramatically and prominently—for example with photographs of the deceased or large headlines—and when suicides of celebrities are reported.
Younger people seem to be most vulnerable to the influence of the media, although limited evidence also shows an impact on elderly people.
In short, there is legitimate concern that the suicide rate tends to increase after the publicity of a suicide, especially that of prominent people or celebrities.
What research hasn’t looked into is whether this effect extends to fictional portrayals of suicide, such as the one in 13RW.
The University of Pennsylvania Study
Today, we have two conflicting studies that provide some possible answers to this question.
The first study, conducted by a team of international researchers (Arendt et al., 2019) led by a University of Pennsylvania professor, recruited subjects who were “young adults (ages 18–29; N = 729) with access to Netflix who completed surveys shortly before and one month after the release of the show’s second season.” Subjects were recruited from online gaming sites; women were over-represented due to methodological limitations (using a third-party audience survey panel).
The researchers also carefully considered the fact that there may be a self-selection bias when studying this issue, since only some people watch the show. And, importantly, “[t]hose who choose to watch in the first place are likely to differ from those who do not. To overcome this selection effect requires analyses that can compare viewers with non-viewers while holding constant predispositions to watch.” So the researchers took this into account when designing their study.
Subjects took a battery of surveys before season two was released, and then took the same surveys four weeks after the release of the second season of 13 Reasons Why.
The researchers found some unexpected and seemingly contradictory results: “our findings suggest that over the course of a month, the show exerted a beneficial effect on students who watched the entire show.” And even better, those at highest risk for suicide benefited the most by watching the entire season:
[… A]mong this higher-risk group, we observed unexpectedly that those who watched the entire season showed less suicidality than those who did not watch at all. (Emphasis added.)
Contrary to many people’s expectations, these researchers show that 13RW actually can help people at greatest risk for suicide. But only if they watch the entire season.
For those who didn’t watch the entire season, the findings are more grim: “For those who only watched some of the second season, we found that the experience predicted elevated suicide risk, especially among current students.” The researchers hypothesized that those who didn’t stick with the second season were exhibiting a sign of distress. When those people stopped watching the show, it may have been as a coping mechanism to stop their own upset and reduce the risk of harm.
The “full dose” effect seems to have outpaced the effect that emerged from only watching some of Season 2. This finding may also be consistent with the results of Zimerman et al. (2018), who found that those who watched the entire first season and who reported prior suicidal ideation were more likely to report improvement in their mental state than a reduction.
The show had no significant impact on people who had a lower initial risk of suicide.
Overall, this was very good research with solid methodological design. It didn’t have the best, most randomized sample in the world, which resulted in an over-representation of women and people who visit gaming sites (who may be different than those who don’t).
Does a Warning Make a Difference?
Netflix failed to include any type of pre-show warning about the content the viewer was about to view in Season 1. The lack of a “viewer warning card” led to an outcry among suicide experts, school psychologists, and well-meaning advocates. So in Season 2 Netflix included such a warning, which begs the question, did it help to keep vulnerable viewers from watching the show? Not according to this research: “Our analysis of the warning that Netflix put out prior to the second season suggests that it may have mainly served to increase viewing.” File under “Good intentions not borne out by the data.”
The Ohio State University College of Medicine Study
Unlike the UPenn study, Bridge et al.’s (2019) study simply looked at the association between suicide data in the U.S. (from data provided by Centers for Disease Control and Prevention) before and after Season 1 of 13RW. The researchers hypothesized that the release of 13RW would have an “immediate and sustained impact on suicide rates in youth and emerging adults.”
Season 1 was released on March 31, 2017, and the researchers found statistically significant increases in the 10-to-17 year age group of suicide deaths during the months of April, June, and December. This means that after its release, the increases happened in only 3 of the 9 months during the study period.
To make this result seem stronger than it is, the researchers also proposed that since March’s suicide death rate was also statistically higher, the mere promotion of the show and the release of its trailer also contributed to this increase. If you feel like the researchers were reaching with this — that a mere advertisement can also significantly contribute to suicide deaths — you wouldn’t be wrong.
But this is not what the media highlighted. Instead of noting that the researchers did not find a sustained impact on suicide rates, they went with the highlights — that 13RW was associated with increased suicide deaths among those in one age group, 10-to-17 year olds. When teasing out the data by gender, the researchers furthermore found that this increase did not apply to girls — only boys.
The researchers offered no explanation as to why their study did not find increased suicide deaths during the months of May, July, August, September, October or November. And why December was the outlier, given that the series was released 7 months prior to then.
This data clearly shows there was no sustained effect of the release of the show on suicide rates in this age group, nor was the effect found in girls. This is particularly odd, given that the protagonist in 13RW is a girl. And while there was indeed an association of increased suicide deaths in the month immediately following Season 1’s release, it did not continue into the next month — which is also odd.
Data correlation studies such as this can be difficult to generalize from, since they seek to associate a population trend with a single variable from our environment — the release of a television show. Such a study needs us to suspend our understanding about the hugely complex social lives that teenagers live nowadays — that they could be so readily influenced by a single fictional account — as well as ignore national and international events happening in the world around us (which could offer alternative explanations for the uptick in suicide deaths).
What 13RW Does for Suicide Awareness
The stronger of the two studies is the University of Pennsylvania study, since it was designed from the ground-up to examine the effects of 13RW. The second study was a retrospective examination of data, and while we shouldn’t ignore the uptick in suicide deaths in the month following the series’ release, we also shouldn’t panic about it. As the UPenn study demonstrated, the series may have a beneficial effect for those most at risk. But it may also negatively impact those who don’t take in the entire show’s message — which is ultimately uplifting and important.
In short, it’s not surprising to find that a television show may have an impact on people’s perception of a topic. But at the same time, we shouldn’t over-emphasize the impact such a show can have while ignoring the impact that others can provide to help a person with suicidal thoughts and feelings. If our country had a working mental health system, people who are suicidal wouldn’t be forced to turn to volunteer-run helplines as their first-line treatment option (which is a travesty). If people who were concerned about a friend simply asked them about how they’re really doing — and if concerned about suicide, asking specifically about suicidal thoughts — we could do a better job in making a dent in the suicide rates in this country.
If you or someone you know may be considering suicide, contact the National Suicide Prevention Lifeline at 1-800-273-8255 (En Español: 1-888-628-9454; Deaf and Hard of Hearing: 1-800-799-4889) or the Crisis Text Line by texting HOME to 741741.
Arendt, F. Scherr, S., Pasek, J., Jamiseon, P.E., & Romer, D. (2019). Investigating harmful and helpful effects of watching season 2 of 13 Reasons Why: Results of a two-wave U.S. panel survey. Social Science & Medicine. In press.
Bridge, J.A. et al. (2019). Association Between the Release of Netflix’s 13 Reasons Why and Suicide Rates in the United States: An Interrupted Times Series Analysis. Journal of the American Academy of Child & Adolescent Psychiatry. In press.