Schools opened, suicide attempts in girls skyrocketed.
New CDC data go against a pervasive narrative.
Content warning: suicide
After a year of speculation over a brewing mental health crisis among kids and adolescents, in June the Centers for Disease Control and Prevention finally released data showing that starting in February of 2021, the number of suspected suicide attempts had dramatically increased among girls ages 12-17 in the United States as compared to 2019. You could almost hear a collective, we told you so, from pundits who had spent the year decrying the closure of schools as Covid-19 tore through the country. To them, these statistics were proof positive that closing schools had brutalized kids.
Here’s the snag: the rate of suicide attempts appears to have been inversely related to school closures. That is, when schools closed, suspected suicide attempts actually decreased. When schools reopened, suspected suicide attempts skyrocketed in lockstep.
The rate of suicide attempts appears to have been inversely related to school closures.
Take a look at the animation that we created for Inside Medicine below
using CDC data and detailed statistics on US school closures during the Covid-19 pandemic.‡ It illustrates the association dramatically. When Covid-19 erupted on American soil in March of 2020, schools closed en masse. Right when that happened, suicide attempts among teen girls dropped. Things get a little noisy, but the overall correlation is uncanny. Schools opened more in the fall, suicide attempts rose. Schools closed for winter vacation, suicide attempts plunged for a time. The 2021 portion of the graph is particularly startling. As the percent of open schools approached 75% of normal, the number of suicide attempts increased suddenly.
While an increase in suspected suicide attempts was not observed among adolescent boys during the same period as attempts rose among girls, the earlier association between school closures and a decrease in suicide attempts was observed in boys too, just as it had been in girls.
While these graphs tell us nothing about causation, we ran the numbers and found the degree of association to be “very strong,” which is the highest category statistically possible.‡
You may have noticed that the initial decline in suicide attempts at the outset of the pandemic in the United States in March of 2020 was not any kind of all-time low. In fact, suicide attempts in adolescent girls dropped to rates similar to those seen in the summer of 2019. That’s not unexpected. Suicide death rates (for which we have decades of reliable data) among teens and young adults predictably fall during summer months, and also in December—that is, times when schools are the least in session. So, what our graphs above show, an apparent inverse correlation between school closings and suicidal behavior, is not unique to the Covid-19 pandemic.
•••
Does any of this match the prevailing narrative around school closures during the Covid-19 pandemic? Not in my experience. In fact, the most common narrative I have seen, coming from concerned parents and academics alike, is that school closures fueled an increase in suicidal behavior among students. This had been an untested assertion. If anything, the data here imply the exact opposite.
But is that really the case? Did school closures decrease suicidal behavior? Did opening schools after a long break create or even contribute to a large surge in suicide attempts in adolescent girls?
The most likely scenario is that suicidality is simply too complex a phenomenon to neatly align with any one particular policy or explanation, especially during a once-in-a-century disruption to life as we knew it. While many people will see the graphs above as evidence that, in actuality, school closures reduced suicidality among teens and that opening schools after the deadliest winter in modern American history increased teen suicidality, the truth could be far more mundane than that: that pandemic-related school closures and suicidal behaviors among adolescents have no direct cause-and-effect relationship. To be clear, these graphs do not prove that opening schools made more adolescent girls more suicidal—though that would be an understandable interpretation, given the breathtaking trends they illustrate. But they certainly do show, and rather impressively, that closing schools did not by any means increase suicidal behavior one bit.
In writing this, I have found the surprising reverse correlation between suicidal behavior and time spent in school—both prior to the Covid-19 pandemic and during it—gnawing at me. Could it be that the pressures around school itself are among the most important stressors related to suicidality among teens? If that’s so, the underlying reasons could be related either to academic or social pressures. Regardless, no one would argue that we should do away with in-person learning just because more time spent in the classroom appears to be associated with increased rates of attempted or completed suicide. But it is an upsetting insight, nonetheless. The problem of adolescent suicide demands a comprehensive approach. Suicide deaths among adolescents rose around 50% over the last decade, from around 60 deaths per one million teens annually to around 90 suicide deaths per one million teens annually. The Covid-19 pandemic did not create this problem, but it might yet teach us something about it.
•••
The effect that Covid-19 school closures has had on children and adolescents is unknown. Some have argued that we have forfeited nearly a year of learning, especially among certain vulnerable populations. But it’s unclear what exactly those loses are, how to quantify them, and whether they are insurmountable. For example, children with cancer often miss many months or more of their normal schooling. But survivors don’t seem to have any long-term differences in scholastic achievement as compared to their healthy peers. Whatever setbacks pediatric cancer survivors face in their education are apparently not permanent in a majority of cases. However, these children often have the benefit of hospital and community-based support systems, designed to keep them from falling too far behind. This then quickly becomes a question of resources and priorities.
The United States has not handled Covid-19 mitigation correctly, especially with respect to children. Can you name a project more essential to the long-term success of our communities than education? And yet in far too many regions of the country, restaurants and bars opened while schools remained shuttered.
Failing to control Covid-19 should be considered a national stain. Even setting the school issue aside, I can think of almost nothing that has harmed children more than our ineptitude in controlling Covid-19 during 2020 and the winter of 2021. Hundreds of children in the United States died of Covid-19, thousands more suffered complications, and tens of thousands of parents of school-aged children died. This is an American tragedy; American in particular because we possess the financial resources to have done far better. Few policymakers were willing or able to do much to make schools safe, nor our communities at large. The story of suicidal behavior and school closures changes none of that.
•••
Suicide attempts and suicide deaths tell us something about the mental health of the population. But they do not tell us everything. In a future post, I’ll discuss child and adolescent mental health during the Covid-19 pandemic in general.
Please leave a comment below on this difficult subject.
Suicide is preventable. If you or someone you know needs to talk to someone, call the National Suicide Prevention Hotline at 800-273-8255, 24 hours a day.
‡Special thanks to Dr. Kristen Panthagani for data visualization created for this Inside Medicine article and to Dr. Chengan Du for statistical analysis.
•••
Works referenced/further reading:
CDC study on suspected suicide 2019-2021: https://www.cdc.gov/mmwr/volumes/70/wr/mm7024e1.htm
US school closures during the pandemic datasets: https://osf.io/tpwqf/
Data on Multisystem inflammatory syndrome in children: https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2780861?utm_source=twitter&utm_campaign=content-shareicons&utm_content=article_engagement&utm_medium=social&utm_term=061121#_blank
Study on parental deaths during the Covid-19 pandemic: https://jamanetwork.com/journals/jamapediatrics/fullarticle/2778229?guestAccessKey=43aa3546-0434-4397-b992-5e4806ea7953&utm_source=For_The_Media&utm_medium=referral&utm_campaign=ftm_links&utm_content=tfl&utm_term=040521
https://brief19.com/2021/04/05/brief/breaking-covid-19-has-created-thousands-of-newly-single-parents-and-orphans
Study on pediatric cancer survivors and academic achievement: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6558589/