Social media use tied to LESS suicide, self-harm among teenage girls

Analysis exposes the dangerous warping of pivotal Centers for Disease Control surveys on teens’ suicides, self-harm, and social media use

Ban teens under age 16 from social media to save girls? Consider the following:

Source: CDC, 2023.

You’d think these easily accessed numbers from the Centers for Disease Control’s go-to 2023 Youth Risk Behavior Surveillance (YRBS), our biggest and only comprehensive survey of 20,000 teenagers, would give anti-social-media crusaders and authorities pause. Especially since the smaller 2021 CDC survey showed a similar pattern.

But… did you see any headlines like, “Authorities say social media may deter girls from self-injury and suicide? More research needed”? If so, please send me links.

The CDC’s 2021 and 2023 surveys

Instead, wildly-hyped books, commentaries, and substack posts by social media blamers like psychologists Jonathan Haidt, Zach Rausch, and Jean Twenge extensively cite the CDC’s YRBS. Then – while not mentioning what the numbers actually show – they imply that social media must be causing more suicide and self-harm by girls in particular, justifying bans on under-age-16 use.

Let’s look at their logic. The 2021 CDC survey showed teen girls who spend lots of time (5 or more hours a day) in front of screens – which include TV and videos as well as PCs, laptops, tablets, and cellphones; just assume they’re all bad – report more frequent depression (47%) than those who rarely or never use screen media (less than 1 hour a day) (30%).

Social-media blamers cite these numbers to announce that (a) social media must be causing girls to be more depressed; (b) we know depression is strongly associated with self-harm and suicide, so, therefore, (c) social media use must be making more girls harm and kill themselves.

Correlation-equals-causation and post-hoc (“after this, therefore because of this”) logic can look convincing, especially if presented all by itself (ie., Sega’s bloody first-person Mortal Kombat was introduced in 1994; youth homicide rates fell; therefore, violent games reduce violence). Such indicators of potential causality certainly merit further investigation in cases where more data are not available.

But not in this case, because the CDC’s 2021 survey did ask those same girls about suicide and self-harm. Curiously, the social-media blamers who eagerly cited the survey’s numbers on girls’ depression somehow omitted adjacent numbers showing what those same girls said about suicide and self-injury.

Why? Let the reader guess.

The next two tables contrast the extremes of screen and social media use, which – if these really do drive teens’ troubles – should show the biggest divergences in outcomes:

Source: CDC 2021, 2023.

In both surveys, girls who used screen and social media the most reported more depression, but less self-harm and suicide attempt. By these numbers, we should prescribe more social media to self-destructive teens. Modestly, I’d think others must have noticed this anomaly, not just me. Wasn’t anyone scientifically (or even wtf-level) curious?

What explains this apparent contradiction?

After all, both surveys also showed girls who report frequent depression are around 4 times more at risk of suicide attempt and 3 times more of self-harm than are non-depressed girls.

Was, then, social media use intervening to deter depressed girls from going on to more rash acts? To examine this question, I combined the numbers into one table:

Again, compared to girls who rarely/never use social media, girls who reported using social media several times a day on the 2023 survey were 65% more likely to report depression, but 54% less likely to attempt suicide and 65% less likely to harm themselves.

Is social media actually deterring girls’ self-destruction?

I then examined suicide attempt and self-harm rates by social media use for three categories of girls: depressed versus non-depressed, abused by parents/household adults versus non-abused; and having addicted versus non-addicted parents/guardians. The results were again striking.

First, girls who reported depression, abusive adults, and addicted adults at home all were substantially more likely to use social media compared to girls who were not depressed and lived in healthy households. For example, girls with abusive parents/adults were twice as likely to use social media 5 or more hours a day versus rarely or never than were non-abused girls.

Second, not surprisingly, girls who were depressed, abused, and/or living with addicted adults were substantially – 3 to 6 times – more likely to attempt suicide and harm themselves compared to non-depressed girls in healthier households.

Third, reaffirming previous findings for all three categories of girls, those who used social media frequently were less at risk of suicide attempts and self-harm compared to girls who rarely or never used social media.

From those three findings, my conclusion was that depressed girls from troubled families were using social media to establish connections and to get help, actions that deterred them from going on to terrible acts like self-harm and suicide.

To some extent, numbers back that up. The 33% of girls living with drug/alcohol-abusing adults were substantially less likely to self-harm or attempt suicide if they used social media daily. However, depressed and abused girls were only slightly less likely to attempt suicide and showed mixed results for self harm the more they used social media.

Now, the mystery

My initial conclusion above turned out to be simplistic (damn kids). Surprisingly, the girls who proportionately benefited the most from frequent social media use were those who reported NOT being depressed.

This category of girls was puzzling. Overall, they reported few problems: just 2% harm themselves and 6% attempt suicide. True, that’s well below the rates of seriously depressed girls (6% and 24%, respectively), but why (by definition) would mentally healthy girls report any self-destruction?

These are not quirks of small samples. The 2023 CDC survey’s questions on these issues gained responses from nearly 5,000 girls – 2,116 who report frequently poor mental health, 1,757 who report poor mental health sometimes, and 1,083 who rarely or never suffer poor mental health.

So, a revised explanation…

Focusing only on non-depressed girls, those on social media 5 or more hours a day are 60% less likely to attempt suicide, and 89% less likely to self-harm, compared to non-depressed girls who rarely or never use social media. That’s as huge as it looks – always keeping in mind that non-depressed girls have low self-harm and suicide attempt rates to begin with.

Maybe, then, a fraction of non-depressed girls occasionally use self-harm and attempted suicide instrumentally to win attention, for whatever reason. In that case, girls who use social media frequently would have the means to get attention without committing rash acts, so they show less overt self-destruction.

Or, perhaps some girls who report good mental health on the 2023 survey were depressed and suffered more self-harm and suicide attempts in the past, used social media to find connections and help, and now are not depressed.

Whatever their thinking, it seems to work. The ultimately encouraging reality is that teen girls have far lower completed suicide and self-destructive death rates than the grownups lamenting girls’ behaviors, as CDC tabulations of suicide, drug-alcohol, and mental-disorder death rates per 100,000 population by age in 2023 show:

· Females: age 10-14 (2.9), 15-19 (9.1), 50-54 (Twenge’s age, 42.6)

· Males: age 10-14 (3.0), 15-19 (22.6), 60-64 (Haidt’s age, 115.9)

Grownups would help teenagers the most by attending to their own mental health, psychologist Christopher Ferguson points out. Of the staggering 41% of girls who told the 2023 CDC survey they had “lived with a parent or guardian who had severe depression, anxiety, or another mental illness, or was suicidal,” 58% are frequently depressed themselves. Perhaps many of the other 42% of these girls who now report not being depressed had issues with mentally ill and suicidal parents in the past that they now no longer have, and those with social media access were much more likely to recover.

Advocates and officials need to carefully consider the roles of social media, “self-harm,” and “suicide attempt” as strategies teen girls use to get attention and help for problems before they escalate to the deadly ends we see among grownups and males, which help keep girls’ completed suicide and self-inflicted death rates so low.

These counter-findings and complications present strong warnings against sweeping social media and smartphone bans and restrictions. Isn’t 2025 writ large teaching us the dangerous consequences of authorities acting rashly out of ignorance and the value of humility in taking a harder look at surprising realities?