Mike Males, Principal Investigator, YouthFacts.org| July 2026

A better question: Why did a few nations like the United States, United Kingdom, Australia, and Japan suffer teen suicide increases while dozens more like France, Spain, Italy, Korea, had declines?

Ban-teens advocates constantly trumpet that social media must be torturing vastly more teenaged girls into self-destruction all over the globe. Psychologists Jonathan Haidt, Zach Rausch, and Jean Twenge advance the following logic: if girls’ suicide rates are rising everywhere, that means we can’t blame some random factor unique to this or that country. No, they argue: we should blame the one factor common to all nations… social media! What else could possibly be instigating internationally rising suicide among girls?

I investigated that argument and found one big problem: it isn’t true. As social media use exploded among teenage girls across the world over the last three decades, here’s what really happened:

· 27 of the 41 European, Latin American, East Asian, and North American nations with complete data show DECREASED suicide rates by girls age 15-19.

· 30 of these 41 nations show DECREASED suicide rates among supposedly social-media-devastated girls age 10-14 (Table 1).

The real issue is the opposite: what is the problem specific to the few nations that suffered teen suicide increases while the large majority showed declines? By their own logic, social media must NOT be the culprit.

Table 1. Suicide trends by teenage girls ages 10-14 and 15-19 by nation, 1995 to 2023

Source: Global Burden of Disease, University of Washington School of Medicine Institute for Health Metrics and Evaluation (2026), Data tool. Suicides are deaths from “self-harm.” Rates are suicides per 100,000 population by country. “Social media era” is 1995 through 2023, with all countries supplying data for all 28 years. Rates and changes are calculated using standard regression that incorporates all 28 years of data into a linear trendline. Trends derived from WHO mortality data are similar. See note on method.*

Equally emphatic, during the post-1995 rise of social media…

· 34 of 41 nations show decreased suicide rates among boys age 15-19

· 36 of 41 nations show decreased suicide rates among boys age 10-14

… a trend ban-teens crusaders tacitly concede, even if they publicly pretend otherwise.

Since large majorities of nations show decreased suicide among teens as social media use mushroomed, why are so many pushing measures to ban teens from social media? Why aren’t those few nations with suicide increases instead investigating troubling factors unique to themselves? That question answers itself.

I suspect those who analyze teen suicide internationally already know these realities. The most recent posting by ban-teens-from-social-media advocates on Jonathan Haidt’s substack is, “Teen girls’ suicide rates are rising in many countries around the world.” The weasel-word “many” could mean anything.

Why not a straightforward statement: How many countries have these advocates analyzed using standard, consistent methods, and of these, how many showed suicide increases versus decreases?

The posting dodges that issue by citing secondary articles that compare varying, arbitrarily selected recent years (ones over-influenced by the COVID-19 suicide anomalies) to arbitrarily selected base years (some, like 2010, well into the online era, when 90% of teens already used social media).

Why the weird methods and inconsistencies?

Evaluating how teen suicide, male, female, younger, older, changed during the rise of social media involves straightforward analytical procedures. First, a responsible analyst would identify those nations with reasonably complete death registration systems. I found 41 nations (total population 1.5 billion) that met that criterion, excluding tiny principalities. Some authoritarian and developing countries merit skepticism. Reasonable analysts could add a couple or subtract a couple of nations.

Second, define a consistent time period allowing comparisons of the most recent suicide rates decades into the social media era versus those of the most recent time period before social media emerged. I chose 1995 as the base year when very few teens went online and social media did not exist. Reasonable analysts could choose a base year a bit earlier or later, but not as late as 2005, 2010, or 2015, when nearly all teens were already online.

The best method would then be used to answer the research question: Did increased social media use accompany (that is, vary with) increased, decreased, or no change in suicide? In this case, the standard method is linear regression, which incorporates suicide rates for every year into a trendline that most accurately reflects (that is, minimizes the error in) overall trends. (If the analyst is studying subtrends, polynomial expansion might be better, but never just picking arbitrary years to compare.)

Finally, honestly report the results upfront

My analysis using Institute for Health Metrics and Evaluation international data for 1995-2023 shows that as teenagers’ social media use rose sharply in the decades since 1995, 27 of the 41 nations consistently showed decreased teen suicide for both genders and all ages. Four nations consistently showed increased teen suicide for both genders and all ages, and 10 showed increases or decreases depending on gender and age.

Would a different data set produce different results? I substituted the World Health Organization’s less complete Demographic Statistics Database for 1995 forward.* The results were nearly identical to those using IHME data. In the United Kingdom’s case, I substituted national data indicating a small increase in teen suicide for my previous analysis using IHME and WHO data that had indicated slightly decreased suicide.**

The “problem nations”

The United States, Japan, Mexico, United Kingdom, and Argentina, which are among the minority of nations that generally showed increased suicide among teens, include 4 of the most populous Western nations. So, if you lump teen suicides into one gross total, these countries have disproportionate impact.***

What factors might have caused teen suicide to increase in these 5 nations? Grownups’ rising opiate deaths and more abusive families are certainly suspect causal factors in the US and UK. However, drug abuse does not appear to be the problem in Japan, Mexico, or Argentina, so I will leave explanations to those more familiar with their sociologies.

Meanwhile, other countries led by France, Italy, Spain, Chile, and Romania (and Russia, though I’m more skeptical of its statistics and the efficacy of its social media suppressions) showed post-1995 teen suicide declines, as did the large majority of medium-sized and small countries. So, if increased social media is the problem, why did a large majority of nations show decreased teen suicide as social media use rose?

Contrary to demagogues’ alarmism, teen girls have LOW suicide rates. Why?

In practical terms (ones never mentioned), the alarming “260% increase” in suicide by U.S. girls ages 10-14 represents a change in behavior by 1 in 105,000 younger girls over the 28-year 1995-2023 period.

Suicide by girls that age remains rarer by far than for any older age or gender. Here are US Centers for Disease Control-reported odds of an individual dying in 2023 from self-inflicted causes (suicides or overdoses):

· Girl age 10-14: 1 in 43,000

· Woman age 40-44 (her mother, caretaker, relative, teacher, coach, etc.): 1 in 2,300

· Man age 40-44 (her father, stepfather, relative, teacher, coach, etc.): 1 in 850.

Yes, you read that right: supposedly mentally unstable, social-media tortured, peer-bullied middle-school girls are 20 times more likely to lose a mother or other close female grownup and 50 times more likely to lose a father or other close male grownup to suicide and suicidal causes than the other way around.

Why might that be? Let’s first look at how girls use self-harm to gain attention for personal distress.

Table 2. Does self-harm work to reduce actual suicide in girls?

Sources: CDC mortality and ER visits, 2023.

Ban-teens crusaders like Haidt, Rausch, and Twenge hype girls’ ER visits as somehow proving their social-media-driven mental troubles, alongside boasting about the healthy stability of us grownups as proven by our lower ER case toll. Strangely, they fail to mention older ages are far more likely to just commit suicide – to hell with trying to get help first.

Instead of using ER cases to stigmatize girls as mentally ill, Haidt, Twenge, Rausch et al might humbly consider how ER cases might help produce girls’ very low actual suicide rates compared to the older, vastly more suicidal grownups around them – grownups whose mental health problems girls cite on CDC surveys as major causes of their own depression and suicidal behaviors.

Does social media deter teen suicide?

In a society whose health professionals, psychologists, political leaders, institutions, media reporters, etc., actually cared about reducing tragedies like suicide and caring for young people – the U.S., UK, Australia, and growing numbers of Western countries emphatically are not those kinds of societies – we would be forcefully confronting uncomfortable but real factors.

We might re-evaluate our views of social media, particularly CDC surveys showing how large proportions of teenagers (particularly girls) who report abuses and household adversities act to prevent personal progression to worse tragedies like suicide and self-harm.

Were we to pursue a better-grounded correlation-equals-causation theory that social media actually REDUCES teen suicide, including for girls – one I’m not endorsing, since it requires more solid causal evidence – then we might start with completely ignored figures from the CDC’s most recent Youth Risk Behavior Survey:

Table 3. The more girls use social media, the less self-harming and suicidal they are

Source: CDC, 2023. Total n=7,588.

Notice how sharply risks of self-harm and suicide attempt increase with greater abuse and adult-inflicted adversity – a reality teens face but political leaders, professionals, health experts, and media reporters so concerned with teens’ mental health just can’t bring themselves to confront. Notice next how more social media use is associated with less of these serious issues.

In that context, we might also cite Pew Research Center surveys in 2022 and 2025 showing large majorities of teens report using social media to feel “more accepted” and to connect with “people who can help them get through tough times.”

The social-media/suicide panic is another destructive hoax

The willful ignorance and misinformation gushing forth on teenagers, girls, mental health, and suicide is grotesque. Authorities’ mindless stampede to ban youths from social media is not just unwarranted and repressive, but dangerous, and their crude exploitation of alarmism about girls’ suicide is reprehensible.

But the situation young people face – and the reason they should openly defy social media bans where imposed – may be more troubling still. I don’t think genuine advocates for young people and their media freedoms perceive how deeply threatening and enraging younger generations’ increasingly progressive views on issues like Israel, climate change, and racial/immigrant/LGBTQ justice are to established interests.

Authorities in many countries led by United States, United Kingdom, and Australia now display such high levels of hostility toward young people they may be indifferent to harming them. That is, ignoring the abuses teens face is not incompetence; it is deliberate, part of the larger plan now unfolding to shove young people out of society, both online and public.

——-

* What mortality data set is the best to use? None are perfect, suffering inconsistencies, incompleteness, and gaps (national health agency tabulations, World Health Association databases); or using formulas to adjust for data gaps and produce estimates of the true numbers of suicides (Institute for Health Metrics and Evaluation tabulations). These issues are not biasing; whatever errors they contain appear to balance out across populations and over time, so that their trends turn out to be similar no matter what database is used. I use the IHME numbers because they have no data gaps, are produced under a consistent estimation technique informed by multiple agencies, and incorporate population changes. As a result, IHME estimates tend to be slightly higher than other databases’ unadjusted numbers. I looked at the same trends using WHO numbers and found almost identical results.

If you’re suspicious that data from developing countries like Mexico might show artificial increases in suicides as death registration became more complete, or that authoritarian states like Russia might mess with numbers, I share your skepticism. I present their numbers to be as complete as possible for those countries that evaluations suggest have tabulated 90% or more of their deaths since the 1990s.

** Both the IHME and WHO databases indicate United Kingdom girls’ and boys’ suicide rates fell during the post-1995 social media era, but their numbers are considerably lower than those presented by the UK’s Office for National Statistics. The problem may be differing definitions. In any case, I substituted ONS suicide numbers. The result is that instead of showing a small decrease, the ONS-based analysis shows slightly increased UK teen suicide from 1995 forward (ONS numbers also show teens’ suicide rates increasing long before the social media era). It’s not a huge deal. My substack posting last week showing the UK had an 8% increase in teen suicide after 1995 should be revised to 10%.

*** If you aggregate all nations reporting to the IHME’s “global” database – which the IHME adjusts to compensate for incomplete death registration over time to estimate some 63,000 teen suicides in a worldwide population of 1.35 billion ages 10-19 – the result is major declines in teen suicides from 1995 to 2023, especially for younger (down 44%) and older (down 47%) teen girls, much larger than for boys (down 25% to 30%). This is counter-intuitive, because one would think that as nations made death registration more complete and suicide became more acceptable to diagnose over time, numbers would rise, not fall. I do not include these numbers because their national derivations and consistency are not easily determined.