247: The Anxious Generation Review (Part 1): Is There Really a Mental Health Crisis in the U.S.?

Hi, I'm Adrian in suburban Chicagoland, and this is Your Parenting Mojo with Jen Lumanlan. Jen is working on a series of episodes based on the challenges you are having with your child. From tooth brushing to sibling fighting to the endless resistance to wHaidtver you ask, Jen will look across all the evidence from thousands of scientific papers across a whole range of topics related to parenting and child development to help you see solutions to the issue you're facing that hadn't seen possible before. If you'd like a personalized answer to your challenge, just make a video if possible, or an audio clip if not. That's less than one minute long that describes what's happening, and email it to support@yourparentingmojo.com and listen out for your episode soon.

Hello and welcome to the Your Parenting Mojo podcast. I've been getting emails about the book “The Anxious Generation” for months now. People saying, “Have you read it?”,” What do you think?”, “Our book club is discussing it next week”, parents in a lot of different places are asking variations of the same question. Are smartphones in general, and is social media in particular, really destroying our kids' mental health? If you're wondering what The Anxious Generation is all about, the central thesis of the book is that we're facing a true mental health crisis in the United States and in other places as well, and that smartphones, generally, and social media specifically, are rewiring our kids' brains. The decline of what Dr. Jonathan Haidt calls the play-based childhood and the rise of the phone-based childhood that occurred between twenty-ten and twenty-fifteen, are, he says, responsible for this big change in Gen Z's mental health. But as I dug into the research behind these claims, I got more and more uncomfortable, because what if we're panicking over the wrong things? What if those dramatic increases in teen depression aren't quite what they seem? What if we're looking at data in ways that make normal fluctuations look kind of like a crisis? Don't get me wrong. I do think our kids are struggling. We're seeing real anxiety in teenagers and concerning youth mental health trends, but before we decide how to help them, we need to make sure we understand what's actually happening, because here's what's at stake, right? If we're fighting the wrong battle, we're not just wasting time and energy, we might actually be missing our chance to address the real problems our kids are facing. This topic turned out to be so big that I've split it into three episodes.

In this first episode, we're going to examine whether there really is a mental health crisis among teens. We'll look at those scary graphs in the book, dig into what's actually driving the numbers, and discover how changes in the way we diagnose and track mental health might be inflating what we perceive as a crisis. We'll also spend time in a wealthy California community that had everything Dr. Jonathan Haidt says should protect kids, tight social bonds involve parents shared values, and yet, the community's teen suicide rate was four to five times the national average. What went wrong there? In part two, we'll examine the studies that claim to prove social media causes mental health problems. And just a little bit of a spoiler alert, when researchers tell kids they're studying digital wellness, the kids figure out pretty quickly what answers the adults want to hear. We'll also look at what actually does explain teen mental health, and it may not be what most people think. And in part three, we'll look at the solutions that everyone's talking about. Should we ban phones in schools, as many school districts and even states are now doing. Should we forbid children from having smartphones at all until they're 16? Will these actions really make a difference in kids' wellbeing, or will they end up creating bigger problems in our relationship than they address for our kids? All right, so let's dig into our first topic for today, and I want to orient you a little bit to where I'm coming from in this episode. I read The Anxious Generation when it was first published, and I knew it was something I wanted to share some thoughts on. But I'm not a quick thinker, and I knew it would take me a long time to go through the book and a large number of its references as well and digest it and create something for you that is really meaningful and useful. And I finally had sometime this summer to be able to do the kind of in-depth analysis that I wanted to do. So here it is. I do want to be clear before we dive in, this is not a takedown of Dr. Haidt or The Anxious Generation. I've actually been incredibly impressed by his and his lead researcher, Zach Roush, his commitment to really understanding what's happening with kids’ mental health.

They've published a whole host of Google Docs where they share the sources of their research and invite comment from researchers. They publish things like an extended back and forth between Dr. Haidt and Dr. Chris Ferguson at Stetson University, where Dr. Ferguson says that he would fail a student for submitting a senior project using the statistical evidence that Dr. Haidt presents in the Google Doc and that appear throughout The Anxious Generation. We'll go into much more detail on that kind of evidence in a bit. But here, I just want to point out that I find Haidt and Rausch commitment to transparency and ongoing collaboration to truly understand what's going on for kids to be really admirable. They also make it clear that we all have biases. In fact, Dr. Haidt's previous book, The Righteous Mind, talks about how we all have a six-ton elephant representing our emotional side that's driven by instinct, and a little rider perched on the back of the elephant that represents our rational mind, which is a whole lot smaller than the elephant. Haidt and his collaborator Dr. Jean Twenge have both taken public positions that there is a causal relationship between teen social media usage and their declining mental health. They acknowledge their confirmation bias. They knew what they were going to look for, and they found it, and they still invite “help from critics to improve our thinking and get closer to the truth”. In the spirit of this, I want to acknowledge my own biases going into this episode as well, I wasn't convinced that social media usage can be responsible for as much of a change as Haidt says we're seeing, or even that kids are experiencing as much of a change as Haidt says that we're experiencing. Before I started working on this episode, I had seen Dr. Candice Odgers’ criticism of The Anxious Generation in the journal Nature, where she asserted that Haidt had mistaken correlation for causation in his work. I'd also seen Dr. Peter Gray, whose work I generally respect, and who collaborates with Haidt on other projects, who was critical of some of the studies that Haidt draws on. So I definitely came in as a skeptic, but I hadn't actually looked at any of the data, the skeptics were drawing on, so I wasn't willing to state any strong conclusions to the parents who were asking my opinion. I don't love social media myself and Carys is now 11, and she does not have a phone or use social media accounts of her own. There are a few kids at her not school program, who have phones, but she doesn't play with those kids as much as she plays with the younger kids who don't have phones. We actually went for a meet and greet with a new pet-sitting client last week for Carys’s pet sitting business, and the client made some kind of offhand comment about how kids these days are all stuck in their phones.

And Carys spontaneously said, I don't want a phone. And I talked with her about that afterwards, and she said she likes playing with kids, and she wouldn't want to do things like sit in a room with her friends, where they all text each other. But I assume there may come a time when she is interested in getting one, and I want to be ready for that. So that's kind of how I'm now approaching this topic. So across the episodes in this series, I want to answer a couple of main questions. In this episode. We're going to look at a question that you probably think is pretty easy to answer, but it turns out that it might not be that easy after all, and that is: Is there a mental health crisis among teens in the US specifically and internationally as well? In our next episode, we'll look at whether this mental health crisis, if we can call a crisis, has much to do with screen time, in particular, and especially teen girls mental health. And then finally, we'll look at whether we should and can move away from what Haidt called a phone-based childhood and towards a phone-free childhood by taking actions like banning kids access to social media or phones more broadly, and encouraging them to play more. So let's begin by looking at whether there really is a mental health crisis among kids.

I want to mention that we're going to talk quite a bit about self-harm and suicide in this episode. So if that would be difficult for you, please do consider coming back to it another time and take care of yourself today. We're going to look at the evidence for a sharp decline in children and teens mental health between two thousand ten and two thousand fifteen, which Dr. Haidt says are the critical dates but Dr. Jean Twenge, one of his collaborators, has said the first date should be two thousand seven when the iPhone was released. Haidt shades the period between two thousand ten and two thousand fifteen on all of his graphs in the book, because the iPhone four was released in June two thousand ten it was the first iPhone with a front facing camera, which made it easier to take photos and pictures of yourself. And Samsung offered one on its Galaxy S the same month. Instagram was created that year and could only be used on smartphones at that point. Instagram had 10 million users near the end of two thousand eleven and that was up to 90 million by early twenty thirteen. Haidt says, “we might therefore say that the smartphone and selfie based social media ecosystem we knew today emerged in two thousand twelve with Facebook's purchase of Instagram following the introduction of the front facing camera. By two thousand twelve many teen girls would have felt that everyone was getting a smartphone and an Instagram account and everyone was comparing themselves with everyone else.”, and this does not make it at all clear that the front facing camera and Instagram created a mental health crisis that began in two thousand ten, and especially not in two thousand seven.

The Anxious Generation has quite a lot of graphs in it, and especially in its chapter called The Surge of Suffering and no melodrama was intended in the chapter name, I'm sure. A lot of the lines on those graphs are shaped like hockey sticks, and it reminded me of a time when I was at Yale, and I was lucky enough to attend a lecture by the environmental scientist, Dr. Robert Sokolo, it was open to the public as well. And so my husband, Alvin, came into town, and Dr Sokolo put up a slide of the hockey stick graph showing average global temperatures over the last thousand years, with that jolt upward at the end. And even my husband, who worked in advertising, looked at that graph and said, “wow, climate change is real. How can anyone doubt that?” And it's almost like Haidt thought. How can we convince people that screen time is bad? Hockey stick graphs. So there are graphs showing dramatic increases in emergency visits for self-harm, up a hundred eighty-eight percent from two thousand ten to two thousand twenty. Major depression among girls is up a hundred forty five percent from two thousand ten to two thousand twenty. Self-harm episodes among UK teens are up seventy-eight percent for girls from two thousand ten to two thousand eighteen. And there's been an eighty one percent increase in girls’ mental health hospitalizations in Australia from two thousand ten to twenty twenty, the list goes on. Sometimes the increases for boys are higher from a lower baseline, sometimes the increases for boys are lower from a much higher baseline, most notably for suicide. But there are hockey sticks, and there are lots of them, and sometimes the explanations for them don't really seem to fit the data, like when Haidt says that the climate crisis can't be causing the drop in mental health because,” impending threats to a nation or generation do not historically cause rates of mental illness to rise. When countries are attacked, either by military force or by terrorism, citizens usually rally around the flag and each other. They're infused with a strong sense of purpose. Suicide rates drop, and researchers find that decades later, people who were teens during the start of the war show higher levels of trust and cooperation in lab experiments. When young people rally together around a political cause, from opposing the Vietnam War in the nineteen sixties through peak periods of earlier climate activism in the nineteen seventies and nineties, they became energized, not dispirited or depressed.”

So Haidt’s data often only goes back to two thousand eight, but the US has data on depression going back to nineteen fifty and the depression rate during the twenty years of the Vietnam War climbed pretty steadily up one hundred ninety five percent for teen boys, whom we have to imagine were most affected by the draft. We would expect to see a massive drop in the male teen suicide rate after the World Trade Center was attacked in two thousand one and then maybe in two thousand two as well. The rate was already declining by then, and if anything, the decline flattened slightly in two thousand one to two thousand two. I wanted to see the suicide data for myself, so I collected up as much of it as I could find for the US, the UK, Canada, Australia, New Zealand and Europe. In the book Haidt shows data like self-harm episodes for UK teens, but mental health hospitalizations for Australia, depression in the US. Seeing all of those different trends really makes it seem like the same thing is happening all over the Anglosphere, as Haidt puts it. But when you see just the suicide data graphed together, it's quite a different picture. Male teen suicides in the US, which drive the overall rate, sort of look like a sine wave. They peaked in about nineteen-ninety and then dropped and are more recently rising again to their nineteen-ninety levels. There was a nineteen-ninety peak in New Zealand, Australia and Canada as well, but we don't see any of that when the data set starts in two thousand eight as most of Haidt's graphs do. UK boy’s suicides have hovered around a pretty low mean, but New Zealand has been awful. Its nineteen-ninety peak was almost double the rate in the US, and it's been declining ever since female teen suicides are up in all of those countries, and there really is a hockey stick in the US, but they're down in New Zealand, and Haidt argues that smartphone usage and its mental health effects have penetrated the entire Anglosphere.

Both male and female depression rates are up, but female depression was already slowly rising before two thousand seven when the first smartphone was released, and two thousand ten which Haidt says, is the start of these trends and seemed to initially rise faster in Australia, New Zealand and the UK, when the US had a much faster penetration of smartphones. Yes, the percentage of teens reporting good or excellent mental health is declining, especially in the US and the UK, but the graph looks really horrific when your Y axis starts at 50% and you're showing a twenty nine percent decline. It doesn't look great when you look at the full data set on a Y axis from zero to a hundred percent but it doesn't look like quite the same crisis as it does when you're focusing only on the top half of the graph so the drop seems so much bigger. Dr. Twenge wrote a post for a Dr. Haidt’s blog, which is called after babble, in which she listed out 13 other explanations for what she calls an adolescent mental health crisis that do not involve smartphones, and then she proceeded to refute each one of them. The first two items on the list are that people are now more willing to seek help, so there's not really an increase in depression, which is ruled out because surveys showing an increase in teen depression rely on cross sections of the whole population, not just those who seek treatment. The determination of depression is done through people reporting their symptoms on a survey, not by therapists or doctors, and the second linked reason is that more teens are okay with saying they aren't okay, so they're more likely to say they're depressed in a national survey. Twenge says that that can't be the reason that rates have increased either because hospital admissions for self-harm have nearly doubled for fifteen to nineteen year-old girls, and the rate of suicide for this age group has also doubled or more. So, it can't just be that more people are willing to say they're depressed. I know from the work that I've done over the last years on the show that self-report data is notoriously unreliable. People forget, they mis estimate how long they felt depressed. They don't think they should complain about how they feel. Men tend to minimize their depression symptoms more than women, and are much more likely to report that they feel depressed when you ask them questions about life stressors than if you ask them directly about depression. That's why I was especially interested in the data on hospital admissions, because that seems much more cut and dried. Until I started digging into it, I found a fascinating working paper. I think it was actually somewhere in one of Haidt's Google docs called, To what extent are trends in teen mental health driven by changes in reporting? the example of suicide related hospital visits, which is a working paper by Adriana Corridor-Waldron and Janet Currie.

This paper traces changes in the way that depression is diagnosed and suicide is reported that have very real implications on the graphs that are scattered throughout The Anxious Generation, showing an uptick in all kinds of problems teens are having, especially between the years of two thousand ten and two thousand fifteen. In two thousand nine the United States Preventive Task Force recommended screening of adolescents aged twelve to eighteen for major depressive disorder, but only when systems are in place to ensure accurate diagnosis, psychotherapy and follow up, which, of course, is not the case everywhere. In two thousand eleven the Affordable Care Act covered evidence-based services recommended by the US Preventive Task Force on August first two thousand twelve, health insurance plans were required to cover an annual depression screening for women and girls age twelve and older. Several things happened in two thousand fifteen firstly, the American Academy of Pediatrics recommended depression screening from age 11 as a preventive pediatric service. Then there were some really big changes, the 10th edition of the International Classification of Diseases was released, and hospital administrators were required by the government to update their computer systems to use codes from the 10th edition. In the ninth edition, a clinician had to enter two separate codes to indicate self-harm, the first one, recording the type of injury, and the second indicating it had an external cause. The external cause codes were often missing, so researchers couldn't tell if the harm was intentionally inflicted or not. In the updated version, there was one code for intentional drug overdose and another for unintentional drug overdose, and this immediately increased the number of injury and poisoning visits coded as self-harm, and decreased the number of visits with undetermined intent. Finally, in two thousand fifteen the Diagnostic and Statistical Manual was updated to its fifth edition, which involved updated definitions of depression and anxiety. It recognized premenstrual dysphoric disorder for the first time, which involved severe mood swings, anger, anxiety and, you guessed it, depression. The working paper looked at suicide related hospital visits in New Jersey, which increased between two thousand eight and two thousand eighteen with two small but pretty clear inflection points in two thousand eleven and two thousand sixteen and suicide related deaths, which increased two but not nearly as much. Teen suicide related hospital visits by condition showed the same small infection in two thousand eleven and a much bigger one in two thousand sixteen which is very late compared with Haidt’s data sets. And finally, it looked at teens primary diagnosis when suicidal ideation is their secondary diagnosis with mood disorder as the most common primary diagnosis, again, showing an upward inflection in two thousand eleven and a much bigger upward inflection in two thousand sixteen mood disorders include bipolar spectrum disorders and the three main types of depressive disorders. So the point that I want to make here is that there was a really huge shift in our support for depression in these years. 46 million people, almost 18% of the US population did not have health insurance in two thousand ten.

Due to the rollout of the Affordable Care Act, that number was down to 16.8% by two thousand thirteen by two thousand sixteen it was down to 10% and has continued to drop since then. It seems possible that at least some of the rise in depression rates was a result of people who didn't realize they were depressed, going to the doctor, getting a diagnosis, and realizing they're depressed for the first time in this window, which they then tell the surveyors when they call. We can also see that the way data on suicide was collected and coded was attributed by these researchers as responsible for the increase in the suicide rate in New Jersey, where the working paper specifically looked at the data, but the same changes happened on a national level across all 50 states. Haidt's rebuttal at this point is likely to be, well, that doesn't explain the international data. He often cites countries that are far apart, like Iceland and New Zealand, and says that it's not possible that in so many countries, primarily pretty well-off Anglophone countries “adolescent mental health collapsed in similar ways around the same time”, in countries like Canada, the UK, Australia and New Zealand. But did it really? Let's start in the UK, where the two thousand seventeen study by Sarginson and colleagues called Temporal Trends in Antidepressant Prescribing to children in UK Primary Care, two thousand to two thousand fifteen. It included all children and adolescents aged three to 17 years who were registered with a general practitioner who contributed to the clinical practice research data link. About 7% of the UK population belongs to practices who participate and is broadly representative of the UK population, this study found that there were about three prescriptions per one thousand per person years among all children in the study, in the year two thousand with two incidents per one thousand person years for boys and four for girls, the prescriptions rise for all age groups in two thousand one and continue rising for girls after two thousand one. Then there was a sharp decrease in prescriptions through two thousand five which the authors correlated with a TV program doing an investigation on the use of SSRI paroxetine in children. The FDA requested that GlaxoSmithKline provide clinical trial data in two thousand three as prescriptions are already falling. And then the US and UK safety agencies completed investigations into antidepressant safety and efficacy in children in two thousand four. And the FDA required safety warnings be added to antidepressants as usage continued to drop to a low of about 1.6 incidences per one-thousand-person years overall, a little over two for girls, a little over one for boys. A UK agency recommended limitations on antidepressant use in two thousand five although prescriptions only stayed flat for a year for girls

They were already on the rise again by two thousand seven again, before Haidt’s two thousand ten window opens, and by two thousand ten prescriptions were already on the fourth year of an increasing trend. We can see similar ideas in the Canadian data, whereas Statistics Canada data, cited by Haidt, shows a very jagged graph of the suicide rate of teens aged 10 to 14, from two thousand four to two thousand twenty. The rate was one per one hundred thousand when the data set starts in two thousand four jumps to about 2.4 per one hundred thousand in two thousand five then it drops back down to one by two thousand eight and then there's a mostly steady increase up to about 3.3 per a hundred thousand in two thousand sixteen. So the increase had already started before the two thousand ten to two thousand fifteen windows. And after two thousand sixteen it dropped again, dramatically, back down to two per one hundred thousand in two thousand twenty. Yes overall, there's a 43% increase in the suicide rate for girls between two thousand ten and two thousand twenty but I don't believe that girls in Canada suddenly stopped using social media in two thousand sixteen so why the decline? Looking at the same type of data for boys, we see an increase of 25% from two thousand ten to two thousand twenty which is an increase from five per a hundred thousand to six per a hundred thousand Haidt reports that 25% increase, but you have to make sure you're looking really carefully at that Y axis on the graph to see that the actual raw increase is tiny. It reminds me of the 25% increase in deaths from breast cancer among women who were using hormone replacement therapy that made the news in the nineteen-nineties which the news media was in an absolute frenzy over, and nobody stopped to consider that the 25% additional risk was on a comparable scale as the one that we're discussing here. The number of people taking HRT dropped dramatically overnight, and only now, two decades later are those numbers finally recovering, with millions of people having suffered from its absence in the meantime. We could go further into the tiny raw data increases that are reported as huge percentage increases, but let's save that for just a minute to wrap up with the Canadian data. I assume that Haidt reports Canadian data on young women's declining mental health in the book, because it's the most dramatic trend suicide rates. Among 10 to 19 year old girls in Canada did increase very slightly between two thousand to two thousand eighteen but it was a pretty steady increase over those whole 18 years.

Rather than showing a large inflection in two thousand nine or two thousand seven or two thousand ten. There was a small inflection in two thousand nine but another one just as big in two thousand six before phones were coming. There are three other lines on that graph, the one above it is running on almost the same slope. It's a slight rise in suicide rates for females aged 20 to 29 a little higher than that is the rate for males aged 10 to 19, and much higher than that is males aged 20 to 29. Both of those suicide rates are still declining, but are obviously way too high. Haidt focuses on the impact that social media is having on girls and women when suicide rates are far higher than that for boys and men who don't use social media as much. And this is where we remind ourselves of Dr. Chris Ferguson's critique that he would fail a senior student research project for trying to make the inferential leaps that Haidt is trying to reach. In one of his Google Docs, Haidt observes two big jumps in suicides of 10 to 14 year old females in the US, from 66 to 88 in two thousand nine and from 85 to 141 in two thousand thirteen. He says that the rate for the last five years of data is nearly triple the rate for the first five years. Ferguson's counter argument is the raw increase in the number of suicides among 45 to 49 year old men is a thousand deaths, which is a nine hundred percent increase among comparably sized populations of about 10 million each. I do want to be clear, I believe that any suicide death is one too many for the families who are left behind. I can't even imagine the pain and the suffering that each of the families who have lost a child in this way are going through, and I'm so sorry that they have to experience that. But if you're looking at raw numbers rather than an increase in rates, you'd do a lot more to prevent deaths and maybe to prevent hardship among children as well by focusing on older men than on teenage girls.

We can also look at teen girls through a different lens. Transgender kid suicides are difficult to track because they're often lumped in with whatever sex they were assigned at birth, which muddies the water a lot, both for understanding what's going on with trans kids as well as what's happening with cisgender kids. The National Transgender Discrimination survey in the US found that 38% of those assigned male at birth reported a lifetime suicide attempt, and that rate was 44% for those assigned female at birth and identifying as trans. Among actual completed suicides, one study found a disparity in the rate at which LGBTQ plus and non-LGBTQ plus youth widened between two thousand fourteen and two thousand eighteen. So we're looking at four thousand eighty-six total suicides in the US among 11 to 29 year olds between two thousand fourteen and two thousand and nineteen of which six hundred seventy-three are identified as LGBTQ plus cases. When we know that the suicide rate for LGBTQ people who were assigned female at birth is 14% higher than those assigned male at birth, that's pushing up the suicide rate for girls. So we could do a lot to reduce girls suicide rate by focusing on LGBTQ suicides in a much more targeted way, rather than this broad reducing everyone's access to social media approach that Haidt recommends, especially since social media can be a source of support for LGBTQ kids, although, of course, it can also be a source of hurtful comments as well. Haidt focuses on self-harm episodes among teens in the UK with another of those worrying hockey stick graphs that shows a 78% increase in self-harm episodes among girls since two thousand ten and a 134% increase among boys from a much lower baseline. I imagine he doesn't report the suicide data for the UK, because that trend does not show the hockey stick curve. Suicides among 15 to 24 year olds in the UK were around 10 to 12 per a hundred thousand between nineteen ninety-two and two thousand and then fell to about six and a half to seven per one hundred thousand from two thousand six to two thousand thirteen with only a very small uptick since then to about eight and a half per one hundred thousand in two thousand eighteen. And as we've come to expect, boys and men's suicide deaths massively outweigh girls. The decline in suicides among men aged 20 to 24 from five hundred to five hundred fifty a year between nineteen ninety-eight to nineteen ninety-four and to 300 to 350 per year, between two thousand two and two thousand seventeen has driven the overall suicide rate down.

The numbers have ticked up a bit since two thousand eighteen probably because that's the year that England and Wales changed the standard of proof used by coroners from beyond reasonable doubt that the death was a suicide to a balance of probabilities that it was. This will drive up overall rates in the UK, because England's population is so much bigger than Wales, Scotland's and Northern Ireland's. When I looked into data on self-harm in the UK, I found one study of five thousand six hundred patients who were seen in emergency departments in three cities in England, which send data to researchers looking at self-harm. By far and away, the most prevalent problem 11 to 15 year olds reported facing at their first assessed episode for self-harm was their relationship with family members at 64%. The researchers didn't ask specifically about social media, but we can imagine that some of the pressure from social media shows up in teens reporting their relationship with friends and others is a prevalent life problem, which was the case for 35% of girls and 19% of boys. Employment and study challenges were next, then relationship with boyfriend girlfriend, then consequences of abuse. And mental health appears only in the fifth position, reported by 14% of boys and 12% of girls. Mental Health takes up a bigger share of presenting problems for 16 to 18 year olds in the fourth spot, with an average of 16 and a half percent of issues. Again, far more important were the relationship with family members, boyfriends, girlfriends, and employment and study. So the idea that teen girls are committing suicide at record levels driven by their poor mental health, which is driven by their social media habits, is kind of hard to see in this data. I also wanted to know more about the experience of self-harm to see what individuals who do it say about why they do it. I looked for some ethnographic research, which is sort of the opposite of the large-scale data sets that Haidt looks at. PhD student Lisa Quigley and Dr. Amy Chandler, did a meta-analysis of ethnographic studies on self-harm for the Scottish government, and found what they called “common narratives around attention seeking the role of social media influencers and genuineness of suicide attempts as causing particular distress to participants.” So in other words, the people engaging in self-harm not only don't perceive social media as a major driver of their behavior, but they see the narrative that social media was a primary driver of their self-harm as wrong. The study authors “suggest that attempts to understand the reasons for or functions of self-harm may be counterproductive, and many study participants reflect a sense of frustration at repeated attempts by professionals, by parents, by themselves, to understand why someone is self-harming. For every study participant who shared a more typical experience of self-harm as a coping mechanism, as a way to deal with tough emotions as a response to anger, there were many others for whom these characteristics not only did not apply, but contributed to feelings of isolation, otherness and shame.” Many people who self-harm are not teenagers, and they often report feeling frustrated that resources to help people with self-harming behavior are focused on the kinds of challenges that teens struggle with. Looking at the studies of people who have considered or attempted suicide, we find a range of reasons for their actions. One study looked particularly at the differences between Asian Americans, Latinx Americans, and Caucasian Americans who had attempted suicide.

Latinx Americans with a suicide history were the least likely to have experienced thoughts of hopelessness, low self-esteem and meaninglessness. They were more likely to conceptualize suicide as a means to escape suffering caused by social status, lack of income and experiences of abuse and discrimination. Other research has found that stress related to the immigration experience and poverty, low social status, education or income, has been linked to increased suicide risk. Among Asian Americans, suicide risk is strongest when individuals have problems between themselves and others, rather than within themselves. The perception of depression leading to self-harm and suicide risk may be much more of a Eurocentric way of viewing the experience, rather than a universal one, and we're obviously lumping a lot of cultures together, in Asian American and Latinx American. This matters because Haidt's entire framework assumes that measuring depression captures what's wrong with teens. But if different communities conceptualize and experience distress differently, then our measurements and our solutions might be missing huge pieces of the puzzle. We're trying to solve a problem that we haven't even properly defined. School problems are more than twice as likely to be a contributing factor to suicide for Asian and Pacific Islander Americans than for their white counterparts, perhaps because many Asian Americans and their parents put a lot of pressure on themselves to do well in school. But this pressure is not unique to Asian Americans. And ethnography of Latina teens who had attempted suicide “one girl” named Lola as saying “I guess I started thinking about, like, my life, like, about school, I'm not doing so good. I'm like, what am I doing with my future? And I guess it made me kind of sad. My mom screams at me. She's like, why don't you do. Do better. Why don't you try? I do try.” And then the study Lola's mom as saying, “I don't remember these issues growing up, you just did what you had to do, and that was pretty much the end of it. You just do it. You don't get a gold star.” Four of the 10 attempters in that study had experienced sexual or physical abuse. One was terrified to go home each day because she didn't know if that was the day her abusive father would have killed her mother, which created a feeling of hopelessness about the future. But sometimes the situations are much less dramatic. Participant, Sophia described frequent arguments with her mother about chores. When Sophia did her chores, she believed her mother didn't notice. Sophia's Mom told her during a fight, I don't care what you do no more, I don't care. Sophia thought her participation in the family was inconsequential, and concluded, so you don't care if I die. And then she took pills. Her mother interpreted Sofia's behavior in terms of resistance. She just doesn't want to listen. I hope it's a phase. I just don't think it's a phase. I want to know what it is with her, because what happens is her anger comes to, I don't have to do this, that attitude. It's like a disrespect. I'm not your child. I'm your mother. Several of the Latina girls felt isolated, even if they were physically surrounded by friends and family, when others were being showered with praise and attention.

One girl, Daniela, had parents who wanted her to be friends with the kids who were studying a lot, but Daniela had failing grades and was not going to get into college. She believed she couldn't find her place in the world, and said, I just wanted to end everything because I don't feel like there was any solution so I just overdosed. This matters, because Haidt's entire framework assumes that screen time is causing mental health issues which is causing suicide, but if different communities think about and experience distress differently, then our measurements and our solutions might be missing huge pieces of the puzzle. We're trying to solve a problem that we haven't properly defined. It's also important to look at places where teen suicides do not seem to have been driven by social media. I received a book last year from the publisher called Life Under Pressure, which was an about an anonymous town called Poplar Grove. If you'd read Hannah Roisin's article on a suicide epidemic involving Caltrain Tracks in The Atlantic, it takes precisely no detective work at all to identify the town in the book as Palo Alto, which is one of the richest towns in the country, and which experienced a suicide cluster among teens several years ago. The authors, Drs. Anna Mueller and Seth Abrutyn describe Poplar Grove as “an idyllic American town, lovely houses, dot tree lined streets, jogging trails cut through well maintained parks, and the community nestles between two scenic rivers.” The community is tight knit. Parent Victoria moved there with her young daughter and knocked on the door of every house that had a stroller on their front porch. She had developed a crew of 17 moms who would walk for four miles a day, and their kids would play on their driveways for two hours a night. 20 years later, the moms still get together almost every weekend. People in Poplar Grove actually know each other, giving the neighborhoods a real community feel. The authors said that their field notes were filled with parents showing up for each other's events and giving practical and emotional support after tragedies. “Young adults reported knowing that when they were children, if they ever got hurt on the street, they could go into any house and get help. Many families moved to the town and financially supported the public school because it was massively cheaper than sending their kids to private school. Mueller and Abrutyn met Carrie, who graduated from Poplar Grove High School. Carrie said all the kids at school wanted to be successful and internalized the pressure to keep up with and preferably outdo everyone else. She said, I wanted to do good in school, and I wanted to do great in La Crosse. I wanted to be student government president. I just constantly pushed myself. The authors go on to say, in fact, Carrie had achieved all of her high school goals. She was student government president and lacrosse team captain, and she was great at school. She also ended up in the hospital, what her doctors later identified as depression, escalated into a suicidal crisis, but Carrie did not reach out to any of the loving and supportive people in her social networks feeling utterly isolated. This high school superstar attempted to die by suicide.”

The two main types of kids in Poplar Grove who attempted and or completed suicide seem to be the ones who had it all and still didn't think they were good enough, and the ones who weren't academically gifted and knew they could never measure up. One boy, Brian graduated high school but committed suicide during college, you can't help but hear the echoes of Lola's mother in the study on Latina teens in Brian's father Bruce, who says, I grew up in a small rural town in Wisconsin, we had what we needed, but we didn't have any extra luxuries. We tried to give our kids all we think they need, and they're smart kids, they're in college, they're all straight A's. I mean, what is it? What's the problem? Bruce can't conceive that pushing a child to fit into a particular mold of what success looks like might not work well for them. Bruce is a self-employed entrepreneur who admits being demanding with his kids. He said, I'm a driver, and I would drive the kids to be productive citizens in society. But Brian was a laid-back jokester, not an academic star. Years later, Bruce said, I feel bad sometimes that maybe I drove him too hard. The phrase social media does not appear in the index of life under pressure, nor does Facebook or Instagram. Social media only appears once in Hannah Roisin's Atlantic article in the context of kids getting updates on Instagram and Twitter after a peer had committed suicide. I put money on the vast majority of students in Palo Alto having smartphones pretty early on, given their parent’s affluence and presence in Silicon Valley, yet it was their real-life community that pressured these teens to commit suicide, not an online community. Jonathan Haidt uncritically quotes the French sociologist Emily Durkheim, who wrote in eighteen ninety-seven that “in Europe, the general rule was the more tightly people are bound into a community that has the moral authority to restrain their desires, the less likely they are to kill themselves.” Haidt himself writes “Gen Z is less able than any generation in history to put down roots in real world communities populated by known individuals who will still be there a year later. Communities are the social environments in which humans and human children evolved. In contrast, children growing up after the great rewiring, skip through multiple networks whose nodes are a mix of known and unknown people, some using aliases and avatars, many of whom will have vanished by next year or perhaps by tomorrow. Life in these networks is often a daily tornado of memes, fads and ephemeral micro dramas played out among a rotating cast of millions of bit players. They have no roots to anchor them or nourish them. They have no clear set of norms to constrain them and guide their path to adulthood.”

But Poplar Grove had what Mueller and Abrutyn just called, “a dense webbing of community held knowledge that protected the children within it.” Poplar Grove had exactly the kind of binding social order that Durkheim and Haidt argue is crucial to children's wellbeing, and yet, combined with affluent parents who wanted to create affluent children in a school system that pushed them to succeed, this community ended up creating a suicide rate among teens that is four to five times the national average. Visits to the emergency room in Palo Alto for suicidal ideation were highest during the school year and lowest during the summer months. Scientists with the Centers for Disease Control surveyed Palo Alto residents for an epidemiological report and found that residents believed depression and mental health issues academic distress or pressure, disconnection and social isolation, family and cultural pressure and life challenges were the biggest contributors to youth suicide in the town. Students were most likely to report academic distress or pressure and life challenges as being the biggest contributors toward youth suicide in Palo Alto, experiencing cyber bullying on the internet in the last 12 months, made the list of statistically significant risk factors for suicidal ideation in the last year among Palo Alto High School students, but it was ninth of 10 items, far below the top items of ever having drunk alcohol, used illicit drugs, pain medication, smoking cigarettes and being female. Teenager Natalie admitted to Miller and Abrutyn, I had four B’s on my report card, and I was mortified, and I did not want to tell my friends about my grades, because everyone is a straight A student. The class valedictorian is spoken of with awe because they took 19 advanced placement courses. Mueller and Abrutyn say that “of course, the grades and AP classes and varsity championships were not the end goal for poplar groves kids just steps along the way to the long-term goal of adult success and the replication of their own privileged childhoods. For another generation in that progression, going to college, especially a top tier one, was a taken for granted next step.” While these parents could afford college, scholarships were a deep source of pride and external validation for Poplar Grove’s parents, “undeniable evidence that their child really was exceptionally excellent.” But college isn't the end goal either, one Dad Harrison described the cultural narrative in the town, which says “Your kids will have multiple advanced placement classes. Have a high-grade point average, excel in a sport, excel in drama, get a scholarship to Princeton, marry their college sweetheart, end up having a house on the water, two SUVs and a jet ski and a beach condo” And in many cases, the parents had moved to Poplar Grove precisely to expose kids to the kinds of ideals and community that would create this narrative for kids. Remember what Durkheim says about how critical a binding social order is for our wellbeing, as I quote Mueller and Abrutyn one last time “this place exemplifies what sociologists call a tight culture, or a culture where expectations are clear and narrowly defined without a lot of variation or flexibility. The ties that generate the strong sense of weeness of social support and pride in the place also generate this tight, homogenous local culture in which everyone knows exactly who they are supposed to be, how they are supposed to behave and what they should be doing.” In two thousand ten the City of Palo Alto developed a suicide prevention policy to train residents on how to identify and report people at risk of suicide to appropriate authorities.

The policy also set out an intention to support young people's mental health, primarily through “problem solving skills, coping skills and resilience.” I couldn't find any data on whether these effects have been successful, because there's no city specific data after the CDC report, although suicides in Santa Clara County as a whole and across all age groups are generally declining, and the county had the lowest rate of any county in California, my biggest concern is that problem solving skills, coping skills and resilience. Sounds an awful lot like we're going to teach kids how to cope better with the huge expectations we put on them, rather than we're going to see if our culture and values are really serving our kids as much as we had thought they were. This morning, I recorded this episode, I saw a post on Facebook from a parent whose 22-year-old daughter works as a barista at Peet's Coffee near where she grew up, which is only notable because she grew up in Palo Alto. The parents were highly educated, but the daughter had struggled with depression, anxiety and learning differences. A customer came into Pete's coffee sized her up and said, why aren't you in school, meaning University. The irony was that the parent herself who was writing the Facebook post was a former Stanford admissions officer who became a private college admissions consultant, working to get kids who were privileged enough to be able to afford an admissions consultant an advantage in the elite college applications process. It turns out her daughter is in school, in Community College, which is an affordable, not very competitive option for vocational training or to complete the first two years of introductory courses before transferring to a four-year college, which is actually how I went to Berkeley. The Facebook poster said “many don't consider Community College as an option. They should.

They do not want to consider community college or a gap year for many reasons, but often because their parents, who, like my husband and me, moved to Palo Alto because the schools were good, do not want to consider these possibilities. What will people think? What will they post or not post on social media? What will they write on their holiday card affix to their car bumper? The shame of it all, the thinly veiled looks of pity from other parents in the cereal aisle at Trader Joe's. And because the idea of missing a stop on the success train good school to good college to a good job is too gruesome for words. No one likes to talk about it, and so we don't.” But whether or not we live in Palo Alto, chances are we're probably hoping our child would go to college, preferably a good one with a scholarship. Community College would be a last resort a meta-analysis of studies related to perfectionist pressure that parents put on kids indicates that perfectionism has increased among college students between nineteen eighty-nine and two thousand nineteen and that this has a strong association with parental expectations. One pathway that I haven't seen Haidt explore or anyone else actually, is parents got cell phones too between two thousand nine and two thousand twelve he does concede that point, but for the first time, it became much clearer what everyone else's kids are doing in terms of sports and college admissions and other achievements. It's possible this caused parents to put extra pressure on their kids to succeed, and that created at least some of the stress that we're seeing in kids. If this were true, getting our kids off their phones wouldn't make as much of a difference to our kids as we would want, because we aren't talking about cutting off our social media access. We're still going to be seeing all the posts from our peers about their kids' success, so our pressure on them is likely to continue.

The Poplar Grove Palo Alto case study reveals something crucial that Haidt’s framework struggles to explain here we have a community with exactly the kind of tight knit, binding social order that both Durkheim and Haidt argue protects against suicide, and yet it experienced suicide rates four to five times the national average. What made the difference was not social media penetration, but the relentless academic pressure these teens faced, which was largely created by their parents and teachers, not by social media. It's easy enough to say, well, I don't live in Palo Alto, so this is not relevant to me, but if we're putting pressure on our kids to always get good grades, to excel in their extracurriculars, in the hopes of getting a scholarship to do some sort of unique volunteering, because it will look good on their application essays, and if the other parents in our child's school are doing this kind of thing too, then we may unwittingly be creating mini Palo altos in other places as well. If you feel worried that your child may be experiencing depression or anxiety, I do want to give you some key indicators to look out for. Real signs of youth depression in teenagers include persistent sadness lasting more than two weeks, loss of interest in activities that they used to enjoy, changes in sleep or appetite, difficulty concentrating, feelings of worthlessness, and in serious cases, thoughts of self-harm. Anxiety in teenagers often shows up as excessive worry, avoiding social situations, physical symptoms like headaches or stomach aches, perfectionism or panic attacks, the key is persistent patterns that interfere with daily life, not just occasional bad days that all teens experience. So those are the classic signs of depression and anxiety. But I think what makes the Palo Alto story unsettling for parents is that many of the teens who attempted or committed suicide, looked like they were thriving. They had good grades, they played sports, they had friends and romantic relationships. So what are the signs of youth depression and anxiety in teenagers who seem to have it all together? You can look for a perfectionism that goes beyond healthy goal setting, so kids who can't handle any mistake or see anything less than perfect as a failure. Watch out for emotional exhaustion disguised as busyness, where they're constantly achieving but they seem joyless about their accomplishments. Pay attention to anxiety in teenagers that shows up as needing constant validation about their performance, or an inability to relax even during downtime. Other concerning signs include social comparison that never stops. So constantly measuring themselves against peers, even when they're doing well, kids who seem to derive their entire sense of worth from external achievements rather than internal satisfaction, and perhaps most importantly, teens who seem afraid to disappoint you, even when you think you've never put that pressure on them.

The Palo Alto teens teach us that sometimes the kids who look most successful on paper are carrying the heaviest emotional loads, and they may be the least likely to ask for help because they think they should be able to handle it all. Returning to the last part of Haidt's chapter, where he lays out the evidence for the surge of suffering. He presents data on alienation in school, worldwide, it looks pretty bad, but here's why these matters for a podcast where you can't see the visuals. When Haidt describes surges and dramatic increases in alienation, he's painting a mental picture based on the visuals in the hockey stick graphs. You'd be forgiven for thinking that something dramatic is happening. What if I told you that on a scale of one to four, where one is low alienation and four is high, the average moved from about 1.7 to 1.9. That's the actual change about naught point two points on a four-point scale. That looks a whole lot bigger when you don't show the full one to four scale of the scores, and instead only show the area between 1.6 and 2.2 then that change looked much more dramatic. It looks like a hockey stick. In reality, it's more like the tide creeping in a couple of inches than a tsunami. This matters because the language of crisis, things like mental health, emergency, anxious generation, surge of suffering, shapes how we think about solutions. If we believe there's a tsunami, we reach for emergency measures like blanket bans on phones, if we recognize it's a modest tide, we might consider more measured responses. And that's assuming the data on school alienation exists, because I actually couldn't find it. Haidt and Twenge say this data comes from the Program for International Student Assessment, or PISA test, which is given to students in many countries every three years. But they don't specify the exact source.

And I looked pretty hard and couldn't find it. PISA asks students to rate their sense of belonging, not their sense of alienation. The twenty, twenty-two PISA had a separate report on managing screen time, how to protect and equip students against distraction. This would probably have been available during The Anxious Generations editing phase, and it shows two very interesting graphs linking screen time with school belonging. Firstly, students who don't use devices for either learning or leisure outside of school on weekdays or weekends have a negative score on the sense of belonging index. This relationship becomes neutral at one hour per day of using devices for learning and also for leisure, belonging continues to rise for both categories, peaking for device usage for both learning and leisure between one to three hours. Only drops into negative territory for devices used for learning at three to five hours, it's still positive for using devices for leisure at three to five hours, and becomes negative at five hours, plus. For the relationship between the sense of belonging and device use at school, not using devices for leisure at all at school has a small negative relationship with belonging, and what that means is, when you don't use screens, you don't feel like you belong, it becomes slightly positive at up to an hour's usage for leisure at school, then it becomes negative again for time spent over an hour. Never using devices for learning at school, is also negatively associated with belonging, and the relationship hovers around neutral for any usage between none and more than seven hours. So while there might be an increase in alienation in school between two thousand ten and two thousand fifteen it is not at all clear the device use in general and social media use in particular, either inside or outside of school, is responsible for this change. So when Haidt points to multiple countries experiencing similar trends, we need to ask, are they really that similar?

And if the trends aren't consistent across countries with similar smartphone adoption rates. What else might explain the patterns we do see? One possibility is countries in the Anglosphere share more than just smartphones. They also share an economic inequality that's been rising since the two thousand eight financial crisis, and cultural narratives about individual achievement that may create the same kinds of pressures we saw in Palo Alto. So where does all this leave us at the end of part one of our mini-series on The Anxious Generation? We've seen that while teen mental health challenges are real, the crisis narrative might be overblown. Those hockey stick graphs that look so alarming are often measuring changes of naught point two points on a four-point scale, the surge in suicide rates, much of it coincides with changes in how we screen for diagnose and code mental health conditions, and the rate for girls are still so much lower than the rate for boys, who use social media so much less. And when teens actually show up in emergency rooms for self-harm, what's their biggest problem? Not social media, 64% say it's their relationship with family members, while mental health issues, which are supposedly driven by social media use, rank fifth on the list.

And then there's Palo Alto. This community had everything that Durkheim and Haidt said should protect children. Tight social bonds, shared values involve parents, a strong sense of belonging, and yet their teen suicide rate was four to five times the national average. The culprit was not Instagram, it was not Tiktok. It was the relentless pressure to achieve, to get into the right college and to live up to impossible expectations. But even if we accept that there are more challenges among teens today, and for some groups, there clearly are, we still need to ask what's causing them. Is social media really rewiring our children's brains? As Haidt claims that's what we'll explore in part two, we'll examine the experimental studies that Haidt says prove social media causes depression and anxiety. We'll look at what happens to teens mental health when countries roll out high speed internet so they can access social media more easily, and we'll discover why some of the world's leading researchers think that the evidence is far weaker than The Anxious Generation suggests it is. Most importantly, we'll ask if social media is not the main problem, what is and what does that mean for how we should help our kids. Join me next time, as we dig into the research that's shaping how millions of parents think about their children's social media use and why we might be looking entirely in the wrong direction.

If you'd like Jen to address the challenge you're having in parenting, just email your one minute video or audio clip to support@yourparentingmojo.com and listen out for your episode soon.