Adolescent mood disorders since 2010: A collaborative review
An ongoing open-source literature review posted and curated by Jonathan Haidt (NYU-Stern) and Jean Twenge (San Diego State U), and Zach Rausch (NYU-Stern). You can cite this document as:
Haidt, J., Rausch, Z., & Twenge, J. (ongoing). Adolescent mood disorders since 2010: A collaborative review. Unpublished manuscript, New York University.
See also our companion review: Social Media and Mental Health, which examines the evidence that social media use is a substantial contributor to these recent increases. Also see our reviews of trends in Australia, Canada, New Zealand, and other countries here.
You can always find this doc linked from jonathanhaidt.com/reviews
You can also access this doc at this link: https://tinyurl.com/TeenMentalHealthReview
We thank those researchers who put in the time to read this document and raise additional points and counterpoints, including Chris Ferguson (Stetson U), Sally Satel (AEI); Friedrich Johenning (Neuroscience Research Center, Charité University Medicine Berlin); others to come]…… We also thank Cameron How and Chris Vaccaro for their earlier work on this document.
First posted: Feb 18, 2019. Last updated Aug 16, 2024.
If you are a researcher and would like to notify us about other studies, or add comments or counterpoints to this document, please request commenting access to the Google Doc, or contact Haidt directly, and he will set your permissions to add comments to the Google doc, or will paste in your comments. This document is evolving based on feedback.]
====================================================
CLICKABLE TABLE OF CONTENTS:
SECTION 1: DEPRESSION, ANXIETY, AND SUICIDAL IDEATION 6
1.2 USA: COUNTERPOINTS AND CRITICISMS 49
1.4 UK: COUNTERPOINTS AND CRITICISMS 81
1.5 WHAT IS HAPPENING IN OTHER ENGLISH-SPEAKING COUNTRIES? 81
1.5.4 Other Anglosphere Countries 90
1.6 WHAT IS HAPPENING OUTSIDE THE ANGLOSPHERE? 94
SECTION 2: SELF HARM AND SUICIDE ATTEMPTS 115
2.2 USA: COUNTERPOINTS AND CRITICISMS 136
2.4 UK: COUNTERPOINTS AND CRITICISMS 143
2.5 WHAT IS HAPPENING IN OTHER ENGLISH-SPEAKING COUNTRIES? 143
3.1 USA: STUDIES AND DATASETS SHOWING A RISE IN RECENT YEARS IN TEEN SUICIDE 149
3.2 USA: COUNTERPOINTS AND CRITICISMS 159
3.3 UK: STUDIES AND DATASETS SHOWING A RISE IN RECENT YEARS IN TEEN SUICIDE 169
3.4 UK: COUNTERPOINTS AND CRITICISMS 174
3.5 WHAT IS HAPPENING IN OTHER ENGLISH-SPEAKING COUNTRIES? 175
3.6 WHAT IS HAPPENING OUTSIDE THE ANGLOSPHERE? 180
SECTION 4: SUMMARY OF FINDINGS 187
APPENDIX A: VARIATION BY SEX 189
APPENDIX B: VARIATION BY AGE (INCLUDING ADULTS) 189
APPENDIX C: VARIATION BY RACE/ETHNICITY 201
APPENDIX D: VARIATION BY SEXUALITY 208
APPENDIX E: VARIATION BY IDEOLOGY 221
APPENDIX F: VARIATION URBAN VS. RURAL AND U.S. REGION 226
APPENDIX G: VARIATION BY SOCIOECONOMIC STATUS 231
APPENDIX H: VARIATION BY RELIGIOSITY 235
APPENDIX I: WHAT HAPPENED DURING COVID? 236
APPENDIX J: STUDIES ON PERFECTIONISM 246
APPENDIX K: OTHER RELEVANT STUDIES 252
APPENDIX L: COLLABORATIVE REVIEW DOCS 254
APPENDIX M: IS THE RISE JUST BECAUSE OF CHANGES IN REPORTING? 255
INTRODUCTION
Are rates of teen depression and anxiety--and their behavioral correlates such as self-harm and suicide--really rising in the US and UK? If so, for whom, and when did the rise start? Or is this another moral panic, an overreaction to small fluctuations, or to increases that are really due to changes in diagnostic criteria, rather than to real increases in suffering? Most importantly, is there a cohort effect? Is Gen Z (also known as “iGen”; those born after 1995) really different from the Millennials and previous generations? This Google doc collects studies and datasets published/released in 2015 or later (on the assumption that the rise seems to begin around 2012, so would not be clearly detectable until 2014, so could not be published until 2015.)
* * * * * * * * * * * * * * *
MAIN THEORIES THAT ARE WIDELY DISCUSSED AS CAUSES OF THE RISE:
1) The rapid adoption of smartphones and rapid spread of social media among teens in the years 2009-2011. (This argument is highlighted in iGen, by Jean Twenge. See a summary of the argument here.)
2) Changes to parenting practices -- particularly the loss of autonomy and unsupervised play beginning in the 1990s. (This argument is highlighted in The Coddling of the American Mind. Lukianoff and Haidt point to multiple factors, but they believe this is the largest single contributor, interacting with the spread of social media. See a summary of the argument here, and then see our other lit review, on social media and mental health.)
3) Economic factors, including the global financial crisis and the rise of inequality which has fostered more competitive parenting in countries that have high inequality. (This argument is made in Love, Money, & Parenting, by Doepke & Zilibotti. See a summary of the argument here. See also Hogberg 2021)
[Haidt assumes that all three causal processes are at work, and are complementary]
Chris Ferguson comment: “There are probably a great number of historical factors happening during this time frame. Left out, for instance, are everything from more wounded young parent war veterans from 2 wars, to increased political polarization to the opioid epidemic. I'm reading a fascinating book now "The Book of Woe" that suggests around this time that increased diagnosis of bipolar disorder in children may have exposed them to more pharmaceuticals including some with suicide as a risk factor. Point being, it's difficult to isolate a historical cause.” [Haidt’s response: I agree with you, and I am particularly interested in learning more about new chemicals and pharmaceuticals that became common in the late 1990s, when Gen Z was in utero or early childhood, or changes in the use pharmaceuticals during the later childhood of Gen Z. It is possible that we are replaying the catastrophe of leaded gas, which I believe was a major contributor to the American crime wave that began in the late 1960s and ended, mysteriously, in the 1990s. More on that when we discuss suicide rates below.]
* * * * * * * * * * * * * * *
CAUTIONS AND CAVEATS
1) The numbers and patterns almost always vary by gender, so we always show data for male and female separately where available.
2) Young people today, known as Gen Z (born 1996 or later) are probably more willing than previous generations to admit to symptoms and to seek out and accept diagnoses. This may contribute to rising trends in some datasets, particularly for depression and anxiety, without necessarily indicating a real change in underlying rates. See this Guardian article for the claim that this is all that is happening. But this factor is much less relevant for measures of behaviors, such as self-harm and suicide.
3) There is a risk of moral panic. The popular press loves to report on problems with “kids today,” especially if it can be linked to a new technology that kids today use, such as smartphones and social media. So we must all be cautious about believing the steady stream of reports in the American and British press about the mental health crisis. We must look at the original studies and the raw datasets, and we must invite researchers to offer conflicting interpretations.
[Chris Ferguson put the counterpoint this way: “As one thought, the argument that a rise in suicide could be linked to technology seems eerily reminiscent of similar arguments in the 1970s and 80s, that a rise in violent crime could be linked to the introduction of television. Then, when crime began to plummet again in the 1990s, but television violence continued to increase in popularity, those same scholars began to argue that population data wasn't important after all. Are you willing to certify, now and forever more, that should at some point in the future, 10, 20, 30 years even, that suicide rates fall, but social media and screen use continues to remain high or increase, that this pattern would falsify your hypotheses linking screens/social media to suicide?”
Haidt’s response: “No, because crime rates and suicide rates respond to many causal factors. Such a drop would only prove that high rates of social media are not sufficient, on their own, to drive up suicide rates and hold them there. But you are right that many researchers saw two correlated social trends and jumped to the conclusion that one caused the other. You are right that past claims of links between technology use and bad outcomes for kids have mostly been shown to be false or inflated, so the onus is on me to show that “this time is different.”]
* * * * * * * * * * * * * * * * * * * * * * * * *
SECTION 1: DEPRESSION, ANXIETY, AND SUICIDAL IDEATION
1.1 USA: STUDIES AND DATASETS SHOWING A RISE IN RECENT YEARS IN TEEN DEPRESSION, ANXIETY, AND SUICIDAL IDEATION
1.1.1 Twenge, Cooper, Joiner, Duffy, & Binau (2019). Age, period, and cohort trends in mood disorder indicators and suicide-related outcomes in a nationally representative dataset, 2005-2017. Journal of Abnormal Psychology.
ABSTRACT: Drawing from the National Survey on Drug Use and Health (NSDUH; N = 611,880), a nationally representative survey of U.S. adolescents and adults, we assess age, period, and cohort trends in mood disorders and suicide-related outcomes since the mid-2000s. Rates of major depressive episode in the last year increased 52% 2005–2017 (from 8.7% to 13.2%) among adolescents aged 12 to 17 and 63% 2009–2017 (from 8.1% to 13.2%) among young adults 18–25. Serious psychological distress in the last month and suicide-related outcomes (suicidal ideation, plans, attempts, and deaths by suicide) in the last year also increased among young adults 18–25 from 2008–2017 (with a 71% increase in serious psychological distress), with less consistent and weaker increases among adults ages 26 and over. Hierarchical linear modeling analyses separating the effects of age, period, and birth cohort suggest the trends among adults are primarily due to cohort, with a steady rise in mood disorder and suicide-related outcomes between cohorts born from the early 1980s (Millennials) to the late 1990s (iGen). Cultural trends contributing to an increase in mood disorders and suicidal thoughts and behaviors since the mid-2000s, including the rise of electronic communication and digital media and declines in sleep duration, may have had a larger impact on younger people, creating a cohort effect.
FIGURES:
Figure 1: Major depressive episode in the last 12 months, by age group, 2005-2017, NS-DUH. [Note that the rise happens only for Gen Z and younger millennials.]
Figure 2: Major depressive episode in the last 12 months, by age group and sex, 2005-2017, NS-DUH
[Note that 2011 is the last year of normal rates. Beginning around 2012, rates begin to rise.]
Figure 3: Serious psychological distress in the past month, by age group, 2008-2017, NS-DUH
Figure 4: Suicidal thoughts in the last 12 months, by age group, 2008-2017, NS-DUH
NOTES: 1. Teen depression was essentially unchanged before, during, and immediately after the Great Recession (which officially lasted 2007-2009, with unemployment peaking in 2010). Depression among teens began to rise around 2012. The rise came a little later for young adults, around 2014 (see Figure 1).
2. Serious psychological distress in the last month and suicidal thoughts in the last 12 months also rose among young adults (only measured in those 18+; see Figures 3 and 4).
3. The increases were confined to teens and young adults and were weaker or non-existent among those ages 26 and over (see Figures 1, 3, and 4). HLM-based APC analysis confirmed that most of the increase was due to birth cohort (generation) rather than to time period.
% of U.S. teens (12-17 years) experiencing a major depressive episode in the last 12 months. Source: NS-DUH
Thus: The number of teens with depression doubled between 2011 and 2019. More than 23% of girls ages 12-17 experienced a major depressive episode during 2019.
Here is the same data from NSDUH, updated to include up to 2022 data and with a neater format from Zach and Jon:
NOTE: There were methodology changes beginning in 2020: Researchers incorporated web-based interviewing (instead of ONLY in-person interviews), which may have impacted scores, although on these variables we see no sign of a sharp discontinuity. For more on the method change, see here.
1.1.2 Mojtabai, R., Olfson, M., & Han, B. (2016). National trends in the prevalence and treatment of depression in adolescents and young adults. Pediatrics.
ABSTRACT: OBJECTIVES: This study examined national trends in 12-month prevalence of major depressive episodes (MDEs) in adolescents and young adults overall and in different sociodemographic groups, as well as trends in depression treatment between 2005 and 2014.
METHODS: Data were drawn from the National Surveys on Drug Use and Health for 2005 to 2014, which are annual cross-sectional surveys of the US general population. Participants included 172 495 adolescents aged 12 to 17 and 178 755 adults aged 18 to 25. Time trends in 12-month prevalence of MDEs were examined overall and in different subgroups, as were time trends in the use of treatment services.
RESULTS: The 12-month prevalence of MDEs increased from 8.7% in 2005 to 11.3% in 2014 in adolescents and from 8.8% to 9.6% in young adults (both P < .001). The increase was larger and statistically significant only in the age range of 12 to 20 years. The trends remained significant after adjustment for substance use disorders and sociodemographic factors. Mental health care contacts overall did not change over time; however, the use of specialty mental health providers increased in adolescents and young adults, and the use of prescription medications and inpatient hospitalizations increased in adolescents.
CONCLUSIONS: The prevalence of depression in adolescents and young adults has increased in recent years. In the context of little change in mental health treatments, trends in prevalence translate into a growing number of young people with untreated depression. The findings call for renewed efforts to expand service capacity to best meet the mental health care needs of this age group.
FIGURE:
1.1.3 Twenge, Joiner, Rogers, & Martin (2018). Increases in depressive symptoms, suicide-related outcomes, and suicide rates among U.S. adolescents after 2010 and links to increased new media screen time. Clinical Psychological Science.
ABSTRACT: In two nationally representative surveys of U.S. adolescents in grades 8 through 12 (N = 506,820) and national statistics on suicide deaths for those ages 13 to 18, adolescents’ depressive symptoms, suicide-related outcomes, and suicide rates increased between 2010 and 2015, especially among females. Adolescents who spent more time on new media (including social media and electronic devices such as smartphones) were more likely to report mental health issues, and adolescents who spent more time on nonscreen activities (in-person social interaction, sports/exercise, homework, print media, and attending religious services) were less likely. Since 2010, iGen adolescents have spent more time on new media screen activities and less time on nonscreen activities, which may account for the increases in depression and suicide. In contrast, cyclical economic factors such as unemployment and the Dow Jones Index were not linked to depressive symptoms or suicide rates when matched by year.
FIGURE:
Below, with data updated to 2017, and instead showing the percentage with high depressive symptoms (item mean of 3 or over):
Source: Monitoring the Future, 8th, 10th, and 12th graders combined, by sex, 1991-2017
[Note that 2012 is the last year of normal data. Surge begins in 2013, for girls only.]
1.1.4 Keyes, K. M., Gary, D., O’Malley, P. M., Hamilton, A., & Schulenberg, J. (2019). Recent increases in depressive symptoms among US adolescents: Trends from 1991 to 2018. Social Psychiatry and Psychiatric Epidemiology.
ABSTRACT: Mental health problems and mental health related mortality have increased among adolescents, particularly girls. These trends have implications for etiology and prevention and suggest new and emerging risk factors in need of attention. The present study estimated age, period, and cohort effects in depressive symptoms among US nationally representative samples of school attending adolescents from 1991 to 2018. Data are drawn from 1991 to 2018 Monitoring the Future yearly cross-sectional surveys of 8th, 10th, and 12th grade students (N = 1,260,159). Depressive symptoms measured with four questions that had consistent wording and data collection procedures across all 28 years. Age–period–cohort effects estimated using the hierarchical age–period–cohort models. Among girls, depressive symptoms decreased from 1991 to 2011, then reversed course, peaking in 2018; these increases reflected primarily period effects, which compared to the mean of all periods showed a gradual increase starting in 2012 and peaked in 2018 (estimate = 1.15, p < 0.01). Cohort effects were minimal, indicating that increases are observed across all age groups. Among boys, trends were similar although the extent of the increase is less marked compared to girls; there was a declining cohort effect among recently born cohorts, suggesting that increases in depressive symptoms among boys are slower for younger boys compared to older boys in recent years. Trends were generally similar by race/ethnicity and parental education, with a positive cohort effect for Hispanic girls born 1999–2004. Depressive symptoms are increasing among teens, especially among girls, consistent with increases in depression and suicide. Population variation in psychiatric disorder symptoms highlight the importance of current environmental determinants of psychiatric disorder risk, and provide evidence of emerging risk factors that may be shaping a new and concerning trend in adolescent mental health.
FIGURE:
NOTES: This paper uses the same dataset (Monitoring the Future) used in 1.1.3, replicating the rise in depressive symptoms among teens since 2012 with a slightly different scale (using 4 items instead of 6) and including data up to 2018. Increases in depressive symptoms begin after 2012 and are largest for girls, though increases for boys became more pronounced after 2016. Increases for girls are d = .50 2012-2018 and d = .33 for boys 2012-2018. This paper also adds an age-period-cohort analysis, which finds that depressive symptoms showed a similar increase across age groups (13 to 18 year olds), suggesting a period effect among teens. [This means that the pattern suggests that something happened to all teens, as they entered a new period of time, rather than it being a cohort -- such as Gen Z -- entering the dataset gradually. But since all high school students in the study during the sharp rise after 2013 ARE Gen Z, this “period effect” could explain when and why Gen Z became so depressed and anxious. We would want to see if the same thing happened to young adults in their 20s at the same time -- the millennial generation.]
1.1.5 Center for Collegiate Mental Health (2018) Penn State University
ABSTRACT: [None. This is not a representative survey of American college students. It is a compilation of data provided by 152 college and university counseling centers, about 1.3 million appointments made by 179,964 college students, showing the changing nature of student concerns that motivate an increasing number of students to come in for help. The increases here probably reflect, in part, the increasing willingness of Gen Z to self-diagnose. [Sally Satel adds: It may also reflect the increased willingness of universities in recent years to accept students with a history of mental illness] I (Jon Haidt) am including this study here for two reasons: 1) It confirms the commonly stated concern of mental health counselors on campus that volume is increasing during the years in which Gen Z replaced the Millennials, and 2) it shows us that the increase is entirely localized in mood disorders, it is not an across-the-board increase in Gen Z’s willingness to self-diagnose in other categories.]
[Note: the 2021 report is here]
FIGURES:
[The same study shows a rise in self-harm and in suicidal ideation, and in self-reports of having ever made a suicide attempt, as a percentage of all those who come to the counseling center.]
2022 Figures:
1.1.6 Duffy, Twenge, & Joiner (2019). Trends in mood and anxiety symptoms and suicide-related outcomes among U.S. undergraduates, 2007-2018: Evidence from two national surveys. Journal of Adolescent Health.
ABSTRACT: Past work has evidenced increased utilization of mental health services on college campuses, as well as rising rates of mood and suicide-related pathology in adolescents and young adults in recent years. We examined whether such findings are reflective of large-scale, nationwide trends in college student mental health in the past decade.
METHODS: We examined trends in mood, anxiety, and suicide-related outcomes among U.S. college students from 2007 to 2018 across two large national datasets: (1) the National College Health Assessment (n = 610,543; mean age = 21.25 years; 67.7% female; and 72.0% white) and (2) the Healthy Minds Study (n = 177,692; 86% students aged 18–22 years; 57% female; and 74% white). Participants, randomly selected by their educational institution, completed self-report measures of past-year mood, anxiety, nonsuicidal self-injury, and suicidal thoughts and behaviors.
RESULTS: In both samples, rates of depression, anxiety, nonsuicidal self-injury, suicidal ideation, and suicide attempts markedly increased over the assessed years, with rates doubling over the period in many cases. Anger, low flourishing, and suicide plans, each assessed in only one dataset, also exhibited upward trends.
FIGURES:
Figure 1: Mood disorder indicators, intentional self-injury, and suicide-related outcomes in the last year among U.S. undergraduates, National College Health Assessment, 2011–2012 to 2017–2018 academic years.
Figure 2: Mood disorder indicators, intentional self-injury, and suicide-related outcomes in the past year among U.S. undergraduates, Healthy Minds Study, 2007–2018. Note: Self-injury, suicide plans, and suicide attempts were not assessed in the 2015–2016 academic year, and anxiety was not assessed until 2013.
NOTES: 1. Figure 1 shows that reports of overwhelming anxiety and depression rose markedly over the years, showing relative percentage increases of 24% and 34%, respectively. The largest increases were in intentional self-injury (47%), suicidal ideation (76%), and suicide attempts (58%). Feelings of overwhelming anger increased less (13%). All variables exhibited relatively steadily rising trends across academic years.
2. Figure 2 cells include means or percentages, SDs in parentheses, and total n completing the item. The survey was not administered in 2008; dashes indicate years when items were not queried. Comparisons in the last column are 2013–2017 for anxiety and 2012–2017 for all other items. d = difference in terms of standard deviations. PD = percentage difference (B − A/A) representing the percentage difference from the first year to last year in the number of individuals fitting the criteria. T-tests indicated all comparisons were significant at the p < .001 level.
1.1.7 Pew Research (2019). Most U.S. Teens See Anxiety and Depression as a Major Problem Among Their Peers. As reported and graphed by The Economist.
Note that for anxiety and depression, there is no SES difference in the graph on the right. Teens in all three SES categories rate “anxiety and depression” as the most common problem facing people their age, in their community.
1.1.8 YRBS dataset, up to 2021 data [more work needed here] see this overview article from 2019
Regraphed, broken by sex [see Zach’s spreadsheet].
1.1.9 Daly (2021). Prevalence of Depression Among Adolescents in the U.S. From 2009 to 2019: Analysis of Trends by Sex, Race/Ethnicity, and Income. Journal of Adolescent Health.
Abstract: Methods: This study drew on 11 years of the National Survey on Drug Use and Health (N = 167,783), a nationally representative survey of adolescents aged 12–17 years conducted between 2009 and 2019.
Results: The prevalence of past-year major depressive episode (MDE) increased by 7.7 percentage points from 8.1% to 15.8% between 2009 and 2019. MDE increased by 12 percentage points from 11.4% to 23.4% among girls. The gender difference in the prevalence of MDE increased from 6.4% to 14.8% between 2009 and 2019. Black participants experienced a comparatively small increase in depression (4.1%).
Figures from Daly (2021):
1.1.10 The U.S. Surgeon General’s Advisory (2021). Protecting youth mental health.
EXCERPT: Since the pandemic began, rates of psychological distress among young people, including symptoms of anxiety, depression, and other mental health disorders, have increased. Recent research covering 80,000 youth globally found that depressive and anxiety symptoms doubled during the pandemic, with 25% of youth experiencing depressive symptoms and 20% experiencing anxiety symptoms. Negative emotions or behaviors such as impulsivity and irritability—associated with conditions such as ADHD— appear to have moderately increased. Early clinical data are also concerning: In early 2021, emergency department visits in the United States for suspected suicide attempts were 51% higher for adolescent girls and 4% higher for adolescent boys compared to the same time period in early 2019. Moreover, pandemic-related measures reduced in-person interactions among children, friends, social supports, and professionals such as teachers, school counselors, pediatricians, and child welfare workers. This made it harder to recognize signs of child abuse, mental health concerns, and other challenges
ADDITIONAL EXCERPT: Unfortunately, in recent years, national surveys of youth have shown major increases in certain mental health symptoms, including depressive symptoms and suicidal ideation. From 2009 to 2019, the proportion of high school students reporting persistent feelings of sadness or hopelessness increased by 40%; the share seriously considering attempting suicide increased by 36%; and the share creating a suicide plan increased by 44%. Between 2011 and 2015, youth psychiatric visits to emergency departments for depression, anxiety, and behavioral challenges increased by 28%. Between 2007 and 2018, suicide rates among youth ages 10-24 in the US increased by 57%. Early estimates from the National Center for Health Statistics suggest there were tragically more than 6,600 deaths by suicide among the 10-24 age group in 2020.
1.1.11 Radhakrishnan (2022). Pediatric emergency department visits associated with mental health conditions before and during the COVID-19 pandemic—United States, January 2019–January 2022. MMWR.
SUMMARY: WHAT IS ALREADY KNOWN ABOUT THIS TOPIC? The proportion of pediatric emergency department (ED) visits for mental health conditions (MHCs) increased during 2020.
WHAT IS ADDED BY THIS REPORT? Weekly ED visits among adolescent females (aged 12–17 years) increased for two MHCs (eating and tic disorders) during 2020, four (depression, eating, tic, and obsessive-compulsive disorders) during 2021, and five (anxiety; trauma and stressor-related; eating; tic; and obsessive-compulsive disorders) and overall MHC visits during January 2022, compared with 2019. The proportion of ED visits with eating disorders doubled among adolescent females; those for tic disorders approximately tripled during the pandemic.
WHAT ARE THE IMPLICATIONS FOR PUBLIC HEALTH PRACTICE? Early identification and expanded evidence-based prevention and intervention strategies are critical to improving pediatric mental health, especially among adolescent females, who might have increased need.
1.1.12 Bitsko… & Ghandour (2022). Mental Health Surveillance Among Children — United States, 2013–2019. Morbidity and Mortality Weekly Report.
SUMMARY: Mental health encompasses a range of mental, emotional, social, and behavioral functioning and occurs along a continuum from good to poor. Previous research has documented that mental health among children and adolescents is associated with immediate and long-term physical health and chronic disease, health risk behaviors, social relationships, education, and employment. Public health surveillance of children’s mental health can be used to monitor trends in prevalence across populations, increase knowledge about demographic and geographic differences, and support decision-making about prevention and intervention. Numerous federal data systems collect data on various indicators of children’s mental health, particularly mental disorders. The 2013–2019 data from these data systems show that mental disorders begin in early childhood and affect children with a range of sociodemographic characteristics. During this period, the most prevalent disorders diagnosed among U.S. children and adolescents aged 3–17 years were attention-deficit/hyperactivity disorder and anxiety, each affecting approximately one in 11 (9.4%–9.8%) children. Among children and adolescents aged 12–17 years, one fifth (20.9%) had ever experienced a major depressive episode. Among high school students in 2019, 36.7% reported persistently feeling sad or hopeless in the past year, and 18.8% had seriously considered attempting suicide. Approximately seven in 100,000 persons aged 10–19 years died by suicide in 2018 and 2019. Among children and adolescents aged 3–17 years, 9.6%–10.1% had received mental health services, and 7.8% of all children and adolescents aged 3–17 years had taken medication for mental health problems during the past year, based on parent report. Approximately one in four children and adolescents aged 12–17 years reported having received mental health services during the past year. In federal data systems, data on positive indicators of mental health (e.g., resilience) are limited. Although no comprehensive surveillance system for children’s mental health exists and no single indicator can be used to define the mental health of children or to identify the overall number of children with mental disorders, these data confirm that mental disorders among children continue to be a substantial public health concern. These findings can be used by public health professionals, health care providers, state health officials, policymakers, and educators to understand the prevalence of specific mental disorders and other indicators of mental health and the challenges related to mental health surveillance.
1.1.13 Kreski… & Keyes (2022). Time use and associations with internalizing symptoms from 1991 to 2019 among US adolescents. SSM - Population Health.
ABSTRACT: INTRODUCTION: Adolescent time use in recent cohorts is distinguished by large-scale changes, including shifts in parental monitoring, supervision, and adolescent activity patterns, that together may provide a more complete perspective on changing patterns of mental health than can be captured by single risk factors.
METHODS: To determine whether patterns of adolescent time use explain recent increases in depressive and other internalizing symptoms, we first conducted latent profile analyses of 465,839 adolescents, grades 8/10, from annual, cross-sectional Monitoring the Future surveys, years:1991–2019, using twenty-one variables (e.g., frequency of attending parties) to identify groups based on patterns of time use. Most of the sample was female (51.0%), non-Hispanic white (58.8%), and in grade 8 (52.2%); mean age: 14.60 years (95% CI: 14.57, 14.64). We subsequently examined differences in depressive and other internalizing symptoms between these time use groups over time with survey-weighted logistic regressions producing odds ratios.
RESULTS: Analyses derived six groups: part time workers, full time workers, and four groups based on levels of social activities (Low, medium, and high levels, with “High Social” split between those engaged in sports, academics, and community service and those who were not). Internalizing symptoms were predicted by lower socialization, low engagement in activities like sports, academics, and community service, and time spent at a paid job. Adolescents decreasingly engaged in social activities over time, though shifts in time use patterns did not account for much of the overall increase in depressive symptoms.
CONCLUSION: Shifts in adolescent time use do not explain increases in depressive and other internalizing symptoms, which increased across different patterns of time use. Levels of internalizing symptoms were highest among those with low socialization, low recreational engagement, and those working substantial hours. Encouraging socialization, engagement in recreational activities, and providing mental health resources for isolated adolescents may reduce internalizing symptom trends.
FIGURE 3:
1.1.14 Pontes, Ayres, & Pontes (2020). Trends in depressive symptoms and suicidality: Youth risk behavior survey 2009–2017. Nursing Research.
ABSTRACT: BACKGROUND: Death by suicide is the second leading cause of death among adolescents globally. Healthy People 2020 set a goal to reduce by 10% the rate of suicide attempts that required treatment and designated the Youth Risk Behavior Survey (YRBS) to measure this objective.
OBJECTIVES: This study used cross-sectional YRBS data (2009–2017) to (a) investigate whether gender moderates the linear time trend for the rate of depressive symptoms, suicidal ideation, and suicide attempts among high school students within the United States; (b) estimate these same measures stratified by gender; and (c) investigate whether the Healthy People 2020 goal for reduction in suicide attempts was met in 2017 for males and females.
METHODS: Secondary analysis of nationally representative high school students using YRBS data 2009–2017 were analyzed using R and the R survey package. Based on Strengthening the Reporting of Observational Studies in Epidemiology guidelines (STROBE), additive interactions using rate difference were compared to multiplicative interactions using odds ratios.
RESULTS: Additive interactions were identified between female gender and linear trends for depressive symptoms, suicidal ideation, and suicide attempts requiring treatment. Females, but not males, had positive linear trends for depressive symptoms, suicidal ideation, and suicide attempts requiring treatment. Suicide attempts requiring treatment among females was 36% higher in 2017 than in 2009 but decreased 13% in males during the same period.
DISCUSSION: The Healthy People 2020 goal to reduce suicide attempts requiring treatment by 10% has not been met among females. The divergent trends by gender highlight the importance of surveillance measures by gender. Future research is needed to identify better suicide prevention strategies that address underlying factors and are gender specific.
FIGURE:
1.1.15 Mojtabai & Olfson (2020). National Trends in Mental Health Care for US Adolescents. JAMA Psychiatry.
ABSTRACT: IMPORTANCE: The prevalence of adolescent depression and other internalizing mental health problems has increased in recent years, whereas the prevalence of externalizing behaviors has decreased. The association of these changes with the use of mental health services has not been previously examined.
OBJECTIVE: To examine national trends in the care of different mental health problems and in different treatment settings among adolescents.
DESIGN, SETTINGS, AND PARTICIPANTS: Data for this survey study were drawn from the National Survey on Drug Use and Health, an annual cross-sectional survey of the US general population. This study focused on adolescent participants aged 12 to 17 years interviewed from January 1, 2005, to December 31, 2018. Data were reported as weighted percentages and adjusted odds ratios (aORs) and analyzed from July 20 to December 1, 2019.
MAIN OUTCOMES: Time trends in 12-month prevalence of any mental health treatment or counseling in a wide range of settings were examined overall and for different sociodemographic groups, types of mental health problems (internalizing, externalizing, relationship, and school related), and treatment settings (inpatient mental health, outpatient mental health, general medical, and school counseling). Trends in the number of visits and nights in inpatient settings were also examined.
RESULTS: A total of 47,090 of the 230,070 adolescents across survey years (19.7%) received mental health care. Of these, 57.5% were female; 31.3%, aged 12 to 13 years; 35.8%, aged 14 to 15 years; and 32.9%, aged 16 to 17 years. The overall prevalence of mental health care did not change appreciably over time. However, mental health care increased among girls (from 22.8% in 2005-2006 to 25.4% in 2017-2018; aOR, 1.11; 95% CI, 1.04-1.19; P = .001), non-Hispanic white adolescents (from 20.4% in 2005-2006 to 22.7% in 2017-2018; aOR, 1.08; 95% CI, 1.03-1.14; P = .004), and those with private insurance (from 19.4% in 2005-2006 to 21.2% in 2017-2018; aOR, 1.11; 95% CI, 1.04-1.18; P = .002). Internalizing problems, including suicidal ideation and depressive symptoms, accounted for an increasing proportion of care (from 48.3% in 2005-2006 to 57.8% in 2017-2018; aOR, 1.52; 95% CI, 1.39-1.66; P < .001), whereas externalizing problems (from 31.9% in 2005-2006 to 23.7% in 2017-2018; aOR, 0.67; 95% CI, 0.62-0.73; P < .001) and relationship problems (from 30.4% in 2005-2006 to 26.9% in 2017-2018; aOR, 0.75; 95% CI, 0.69-0.82; P < .001) accounted for decreasing proportions. During this period, use of outpatient mental health services increased from 58.1% in 2005-2006 to 67.3% in 2017-2018 (aOR, 1.47; 95% CI, 1.35-1.59; P < .001), although use of school counseling decreased from 49.1% in 2005-2006 to 45.4% in 2017-2018 (aOR, 0.86; 95% CI, 0.79-0.93; P < .001). Outpatient mental health visits (eg, private mental health clinicians, from 7.2 in 2005-2006 to 9.0 in 2017-2018; incidence rate ratio, 1.30; 95% CI, 1.23-1.37; P < .001) and overnight stays in inpatient mental health settings (from 4.0 nights in 2005-2006 to 5.4 nights in 2017-2018; incidence rate ratio, 1.18; 95% CI, 1.02-1.37; P = .03) increased.
Conclusions and Relevance This study’s findings suggest that the growing number of adolescents who receive care for internalizing mental health problems and the increasing share who receive care in specialty outpatient settings are placing new demands on specialty adolescent mental health treatment resources.
FIGURE 1:
FIGURE 2:
1.1.16 Rapaport & Silver (2022). National Survey Findings Shed Light on Dimensions of Teen Mental Health Concerns. Evidence Base.
EXCERPT: Perhaps surprisingly, the overall mental health “difficulties” score (higher scores represent more negative experiences) in 2022 was only a bit higher than it was in 2013 (8.2 compared to 7.9). However, this subtle change at the overall level masks opposing patterns in particular areas, or domains, of mental health. First, there was a statistically significant increase in the Hyperactivity domain (e.g., restless, overactive, easily distracted, fidgeting/squirming), accompanied by a statistically significant decrease in the Conduct Problems domain (e.g., often loses temper, fights with other youth, lies or cheats). In other words, when reporting on the mental health of a child, parents in 2022 reported more restlessness, distraction, fidgeting-type behaviors than a group of parents reported on their children in 2013; but they also reported fewer fights, lying, cheating and losing temper-like behaviors.
…When looking at results by gender of the child, we see similar drops in the Conduct Problems domain compared to the 2013 snapshot, and similar increases in the Hyperactivity domain for both boys and girls (Figure 2). Notably, though, the magnitude of change for girls was always worse than for boys – in all domains that compose the scale. Difficulty scores were larger for girls than boys in the Emotional Symptoms, Hyperactivity, and Peer Relationships domains, and the relative positive lower score in the Conduct Problems domain was smaller for girls. Further exacerbating concerns about girls’ mental health, the relative protective increase in the Prosocial Behaviors domain (the “Strengths” of the SDQ) was smaller for girls than for boys. Though these patterns were not always statistically significant, they align with recent findings that girls’ mental health is in a heightened state of crisis.
…Scores for girls in the difficulty domains were higher (worse) in 2022 than in 2013. When looking only at 2022 data (Table below), girls’ Emotional Symptoms score was significantly higher (worse) than boys’. This domain includes items such as “often seems worried”, “has many fears”, and “often unhappy, depressed or tearful”. In 2022 girls scored almost one full point higher (worse) on this domain scale compared to boys. Boys are struggling more than girls in the area of Hyperactivity (i.e., Hyperactivity scores were higher (worse) for boys), but the difference between boys and girls on that domain is almost half what it is for the Emotional Symptoms domain. Importantly, girls score significantly higher (better) than boys on the potentially protective strength domain of prosocial behaviors (7.83 for girls compared to 7.22 for boys), though as shown above, this domain score increased more (improved more) for boys than it did for girls since 2013.
FIGURE:
1.1.17 American College Health Association (2019), National College Health Assessment. Zach graphed the numbers across datasets in this spreadsheet to produce this graph:
NOTE: The ACHA-NCHA was revised in fall 2019, which prevents further data from being shown. From the American College Health Association’s website: “The ACHA-NCHA III is a new instrument. Results from the ACHA-NCHA III should not be compared with earlier versions of the survey.”
1.1.18 Askari, Rutherford, Mauro, Kreski, & Keyes (2022). Structure and trends of externalizing and internalizing psychiatric symptoms and gender differences among adolescents in the US from 1991 to 2018. Social Psychiatry and Psychiatric Epidemiology.
ABSTRACT: PURPOSE: We aimed to estimate the structure of internalizing and externalizing symptoms and potential time dynamics in their association. This is understudied among adolescents, despite increasing internalizing and decreasing externalizing symptoms in recent years.
METHODS: We analyzed data from US Monitoring the Future cross-sectional surveys (1991-2018) representative of school-attending adolescents (N = 304,542). Exploratory factor analysis using maximum likelihood estimation method and promax rotation resulted in a two-factor solution (factor correlation r = 0.24) that differentiated eight internalizing and seven conduct-related externalizing symptoms. Time-varying effect modification linear regression models estimated the association between standardized internalizing and externalizing symptoms factor scores over time overall and by gender.
RESULTS: In 2012, trends in average factor scores diverged for internalizing and externalizing factors. The average standardized internalizing factor score increased from - 0.03 in 2012 to 0.06 in 2013 and the average externalizing factor score decreased from - 0.06 in 2011 to - 0.13 in 2012. We found that for every one-unit increase in standardized internalizing factor score, standardized externalizing factor score increased by 0.224 units in 2010 (95% CI: 0.215, 0.233); the magnitude of this increase was 22.3% lower in 2018 (i.e., 0.174 units; 95% CI: 0.160, 0.188). Decoupling of internalizing and externalizing symptoms began earlier among boys (~ 1995) than among girls (~ 2010).
CONCLUSIONS: The decoupling of internalizing and externalizing symptoms among adolescents suggests that changes in the prevalence of shared risk factors for adolescent psychiatric symptoms affect these dimensions in opposing directions, raising the importance of considering symptoms and their risk factors together in prevention and intervention efforts.
EXCERPT: Adolescents with high internalizing symptoms, both with and without high externalizing symptoms, are increasing in recent decades, especially in the 2010s, while those with high externalizing symptoms only are decreasing. Associations between demographic and substance use indicators and internalizing and externalizing symptom severity were strengthening in recent decades for high internalizing symptoms, both with and without high externalizing symptoms… When accounting for confounding factors, adjusted time-varying effect modification models indicated that the magnitude of the relationship between internalizing and externalizing symptoms decreased by 22.3% from 2010 to 2018. Recent decreases in the association suggest a potential weakening relationship between internalizing and externalizing symptoms over time. This could mean that the underlying mechanisms that led to an increase in adolescent internalizing symptoms and a decrease in externalizing symptoms in the 2010s are distinct.
FIGURE:
Figure. Average standardized externalizing and internalizing factor scores and the correlation between factor scores over time, 1991–2018
1.1.19 Lipson, Lattie, & Eisenberg (2019). Increased Rates of Mental Health Service Utilization by U.S. College Students: 10-Year Population-Level Trends (2007–2017). Psychiatric Services.
ABSTRACT: OBJECTIVE: This study aimed to document population-level trends in mental health service utilization by college students.
METHODS: The study drew on 10 years of data from the Healthy Minds Study, an annual Web-based survey, with a sample comprising 155,026 students from 196 campuses. Analyses focused on past-year mental health treatment and lifetime diagnoses of a mental health condition. Changes in symptoms of depression and suicidal ideation and levels of stigma were hypothesized as potential explanatory factors.
RESULTS: Rates of treatment and diagnosis increased significantly. The rate of treatment increased from 19% in 2007 to 34% by 2017, while the percentage of students with lifetime diagnoses increased from 22% to 36%. The prevalence of depression and suicidality also increased, while stigma decreased.
CONCLUSIONS: This study provides the most comprehensive evidence to date regarding upward trends in mental health service utilization on U.S. campuses over the past 10 years. Increasing prevalence of mental health problems and decreasing stigma help to explain this trend.
EXCERPT: Rates of past-year treatment for the full sample increased from 18.7% in 2007 to 33.8% in 2016–2017, with similar patterns for both therapy/counseling and medication use. The proportion of students with a diagnosed mental health condition increased from 21.9% in 2007 to 35.5% in 2016–2017. In later years (2014–2017), rates of both treatment and diagnosis were at or above 30% each year. Among students with depression (N=41,299), rates of past-year treatment went from 42.4% in 2009 to 53.3% in 2016–2017 and diagnosis from 43.3% to 55.8%, with rates of nearly 50% or more for both treatment and diagnosis in later years.
…The most common location for receiving services was on campus, with rates increasing from 6.6% in 2007 to 11.8% in 2016–2017. Rates of psychiatric emergency service utilization showed general upward trends, with 0.3% seeking such services in 2007 and 1.0% in 2016–2017. Similarly, service use from other locations increased from 5.2% in 2007 to 8.7% in 2016–2017.
Rates of both perceived and personal stigma decreased over time, from 64.2% to 46.0% and from 11.4% to 5.7%, respectively. Among students with depression, there was no noticeable change in levels of perceived stigma over time, but personal stigma decreased from 8.2% to 5.1%.
FIGURE:
Figure. Mental health service utilization, in percentages, by the full population of college students (top) and students with depression (bottom), by year. aSource: Healthy Minds Study (full population, N=155,026 students at 196 campuses; students with depression, N=41,299 students at 183 campuses). Past-year treatment is any past-year use of therapy/counseling and/or psychotropic medication. Depression is defined as a score of ≥3 on the adapted Patient Health Questionnaire–2 (PHQ-2); PHQ-2 data unavailable for 2007.
1.1.20 Youth Risk Behavior Survey Data Summary & Trends Report: 2013-2023. CDC.
NOTE: See Jon’s commentary on the 2021 YRBS survey. He argues that the new CDC report shows that Covid added little to teen mental health trends.
FIGURES:
1.1.21 Pontes, Ayres, & Pontes (2020). Trends in Depressive Symptoms and Suicidality: Youth Risk Behavior Survey 2009–2017. Nursing Research.
ABSTRACT: BACKGROUND: Death by suicide is the second leading cause of death among adolescents globally. Healthy People 2020 set a goal to reduce by 10% the rate of suicide attempts that required treatment and designated the Youth Risk Behavior Survey (YRBS) to measure this objective.
OBJECTIVE: This study used cross-sectional YRBS data (2009–2017) to (a) investigate whether gender moderates the linear time trend for the rate of depressive symptoms, suicidal ideation, and suicide attempts among high school students within the United States; (b) estimate these same measures stratified by gender; and (c) investigate whether the Healthy People 2020 goal for reduction in suicide attempts was met in 2017 for males and females.
METHODS: Secondary analysis of nationally representative high school students using YRBS data 2009–2017 were analyzed using R and the R survey package. Based on Strengthening the Reporting of Observational Studies in Epidemiology guidelines (STROBE), additive interactions using rate difference were compared to multiplicative interactions using odds ratios.
RESULTS: Additive interactions were identified between female gender and linear trends for depressive symptoms, suicidal ideation, and suicide attempts requiring treatment. Females, but not males, had positive linear trends for depressive symptoms, suicidal ideation, and suicide attempts requiring treatment. Suicide attempts requiring treatment among females was 36% higher in 2017 than in 2009 but decreased 13% in males during the same period.
DISCUSSION: The Healthy People 2020 goal to reduce suicide attempts requiring treatment by 10% has not been met among females. The divergent trends by gender highlight the importance of surveillance measures by gender. Future research is needed to identify better suicide prevention strategies that address underlying factors and are gender specific.
1.1.22 Mojtabai (2023). Problematic Social Media Use and Internalizing Symptoms in Adolescents [Preprint]. In Review.
ABSTRACT: PURPOSE: This study examined time trends in significant child and adolescent internalizing symptoms and explored the association of excessive and problematic social media use with these symptoms.
METHODS Time trends in internalizing symptoms were assessed using data from five waves of the international survey of Health Behavior in School-aged Children (HBSC), conducted between 2001 and 2018 (N=1,036,869). The associations of frequent and problematic social media use with significant internalizing symptoms were assessed by hierarchical multinomial logistic regression using data from 2001-2002 and the 2017-2018 survey waves. Causal direction between social media use and internalizing symptoms was assessed using linear non-gaussian acyclic models (LiNGAM).
RESULTS: Prevalence of more severe internalizing symptoms increased from 6.7% in 2001-2002 to 10.4% in the 2017-2018 survey waves. The increase was especially large among 15-year old and older girls: from 10.9% to 19.1%. The difference in prevalence of more severe internalizing symptoms across survey waves was fully explained by problematic social media use. LiNGAM analysis confirmed the causal direction of social media use variables with internalizing symptoms.
CONCLUSIONS: The study findings suggest that widespread use of social media may explain the increased prevalence of internalizing symptoms in adolescents in recent years.
FIGURES:
1.1.23 Parodi, Holt, Green, Porche, Koenig, & Xuan (2022). Time trends and disparities in anxiety among adolescents, 2012-2018. Social Psychiatry and Psychiatric Epidemiology.
ABSTRACT: PIRPOSE: Prior studies have been inconclusive in documenting whether the prevalence of adolescent anxiety is increasing, given sampling and measurement limitations. This study adds new information on recent time trends in anxiety prevalence, specifically investigating trends among previously unexamined sociodemographic subgroups.
METHODS: Weighted data of 37,360 youth respondents (51.1% female, 71.8% White, 91.3% heterosexual, 99.2% cisgender) from the 2012-2018 Dane County Youth Assessment, a county-wide survey administered to youth in participating school districts, were analyzed to estimate time trends in anxiety prevalence among the whole sample and by sociodemographic subgroups.
RESULTS: The prevalence of youth meeting anxiety-screening criteria increased from 34.1% (95% CI 33.4-34.9) in 2012 to 44% (95% CI 43.2-44.7) in 2018 (OR for trend = 1.07, P for trend < 0.001). The trend remained significant after adjusting for known confounds (AOR for trend = 1.07, P for trend < 0.001). Anxiety increased significantly for several subgroups and widening disparities were documented among females relative to males (P < 0.001), and sexual minority youth relative to heterosexual youth (P = 0.003). In addition, Black youth did not increase at the same swift rate as White youth over the study period (P < 0.001).
CONCLUSION: This study reports recent data on anxiety time trends and finds that among a geographically representative sample of adolescents, anxiety prevalence is rising. Findings provide new evidence documenting increased anxiety prevalence among sexual minority youth relative to their peers. Results highlight the need to bolster public health interventions focused on adolescent mental health, with tailored interventions for vulnerable groups.
1.1.24 Nash, … & Venkatesh (2021). Prolonged Emergency Department Length of Stay for Pediatric Mental Health Visits (2005-2015) Pediatrics.
ABSTRACT: BACKGROUND: Children seeking care in the emergency department (ED) for mental health conditions are at risk for prolonged length of stay (LOS). A more contemporary description of trends and visit characteristics associated with prolonged ED LOS at the national level is lacking in the literature. Our objectives were to (1) compare LOS trends for pediatric mental health versus non–mental health ED visits and (2) explore patient-level characteristics associated with prolonged LOS for mental health ED visits.
METHODS: We conducted an observational analysis of ED visits among children 6 to 17 years of age using the National Hospital Ambulatory Medical Care Survey (2005–2015). We assessed trends in rates of prolonged LOS and the association between prolonged LOS and demographic and clinical characteristics (race and ethnicity, payer type, and presence of a concurrent physical health diagnosis) using descriptive statistics and survey-weighted logistic regression.
RESULTS: From 2005 to 2015, rates of prolonged [Length of Stay] for pediatric mental health ED visits increased over time from 16.3% to 24.6% (LOS >6 hours) and 5.3% to 12.7% (LOS >12 hours), in contrast to non–mental health visits for which LOS remained stable. For mental health visits, Hispanic ethnicity was associated with an almost threefold odds of LOS >12 hours (odds ratio 2.74; 95% confidence interval 1.69–4.44); there was no difference in LOS by payer type.
CONCLUSIONS: The substantial rise in prolonged LOS for mental health ED visits and disparity for Hispanic children suggest worsening and inequitable access to definitive pediatric mental health care. Policy makers and health systems should work to provide equitable and timely access to pediatric mental health care.
1.1.25 Chen, Cowden, Fulks, Plake, & VanderWeele (2022). National Data on Age Gradients in Well-being Among US Adults. JAMA Psychiatry.
ABSTRACT: METHODS: The sample for this survey study was drawn from the NORC AmeriSpeak panel, a probability-based panel designed to be representative of the US household population.3 A stratified sample was selected based on age, race and ethnicity, gender, and education. Surveys were conducted online and via telephone using a 15-minute questionnaire. Data were collected from January 10 to 28, 2022. Based on US census data, post hoc weighting was performed to ensure the sample was representative of US adults 18 years or older within all 50 states and the District of Columbia. Participants responded to a validated 12-item measure of flourishing4,5 that assesses 6 domains of well-being, including happiness, health, meaning, character, relationships, and financial stability. Items were self-reported on a scale of 0 to 10. Domain scores were derived by calculating the mean of the 2 indicators under each domain; the overall well-being score was the mean indicator responses across domains. Ethical approval was granted by the NORC institutional review board. Participants provided written informed consent. This study followed AAPOR and STROBE reporting guidelines. Analysis of variance and the t test (2-sided) were used to examine group differences by age, race and ethnicity, and gender. Analyses were conducted using SAS, version 9.4. The eMethods in the Supplement gives more details.
RESULTS: Of 8618 individuals contacted, 2598 (30.1%) provided complete responses (1338 [51.50%] female; mean [SD] age, 47.92 [17.94] years). Well-being increased monotonically cross-sectionally with age for overall well-being, the domains, and across all individual items (Table and Figure). Differences between the youngest (age 18-25 years) and oldest (age ≥77 years) age groups were present for meaning (2.08 points; 95% CI, 1.63-2.53), happiness (1.99 points; 95% CI, 1.60-2.38), health (1.93 points; 95% CI, 1.56-2.31), relationships (1.91 points; 95% CI, 1.46-2.36), financial stability (1.83 points; 95% CI, 1.23-2.43), and character (1.28 points; 95% CI, 0.93-1.62). The maximum difference in overall well-being among age groups was between the youngest and oldest (1.84 points; 95% CI, 1.52-2.15) and was larger than the largest differences among Asian (mean [SD], 6.50 [2.27]), Black (mean [SD], 6.76 [1.78]), Hispanic (mean [SD], 6.67 [1.54]), White (mean [SD], 7.00 [1.50]), and other (≥2 races, non-Hispanic; other, non-Hispanic) (mean [SD], 6.53 [1.21]) race and ethnicity and between men (mean [SD], 6.87 [1.53]) and women (mean [SD], 6.88 [1.59]).
DISCUSSION: This study found that mean well-being scores across multiple domains increased cross-sectionally with age, with a substantial age gradient. This finding contrasts with evidence from the early 2000s that showed U-shaped curves for some well-being domains (eg, happiness, life satisfaction), with well-being scores being higher in earlier adulthood and older age than in midlife.6 Our findings support evidence of a mental health crisis and increase in loneliness in the US that has disproportionally affected young adults1,2 and extend evidence of age gradients to multiple additional facets of well-being beyond mental health. Younger adults reported lower well-being even on the self-rated physical health item.
A limitation is that we could not disentangle age and cohort effects. Further studies are needed to replicate the findings. Although indicators in the study were obtained by self-report, this was also the case with earlier research that documented U-shaped patterns with age. These findings suggest that the well-being of young people has declined compared with older age groups. Protecting the mental health of young people is regarded as a national emergency; this study suggests that other facets of their well-being also need attention. Comprehensive strategies for expanding mental health services, supporting education and meaningful employment, and strengthening social fabric for young people are needed to enhance the well-being of this population.
1.1.26 Kreski … & Keyes (2023). Changing adolescent activity patterns and the correlation of self-esteem and externalizing mental health symptoms across time: results from the USA from 1991 through 2020. Psychological Medicine.
ABSTRACT: BACKGROUND: Common adolescent psychiatric symptoms cluster into two dominant domains: internalizing and externalizing. Both domains are linked to self-esteem, which serves as a protective factor against a wide range of internalizing and externalizing problems. This study examined trends in US adolescents' self-esteem and externalizing symptoms, and their correlation, by sex and patterns of time use.
METHODS: Using Monitoring the Future data (N = 338 896 adolescents, grades:8/10/12, years:1991-2020), we generated six patterns of time use using latent profile analysis with 17 behavior items (e.g. sports participation, parties, paid work). Groups were differentiated by high/low engagement in sports and either paid work or high/low peer socialization. Within each group, we mapped annual, sex-stratified means of (and correlation between) self-esteem and externalizing factors. We also examined past-decade rates of change for factor means using linear regression and mapped proportions with top-quartile levels of poor self-esteem, externalizing symptoms, or both.
RESULTS: We found consistent increases in poor self-esteem, decreases in externalizing symptoms, and a positive correlation between the two across nearly all activity groups. We also identified a relatively constant proportion of those with high levels of both in every group. Increases in poor self-esteem were most pronounced for female adolescents with low levels of socializing, among whom externalizing symptoms also increased.
CONCLUSIONS: Rising trends in poor self-esteem are consistent across time use groups, as is the existence of a group facing poor self-esteem and externalizing symptoms. Effective interventions for adolescents' poor self-esteem/co-occurring symptoms are needed broadly, but especially among female adolescents with low peer socialization.
1.1.27 Blanchflower & Bryson (2024). The Consequences of Abuse, Neglect and Cyber-Bullying on the Wellbeing of the Young. National Bureau of Economic Research.
ABSTRACT: Using cross-sectional data files for the United States we show that difficulties experienced in childhood - so-called Adverse Child Experiences (ACE)s - are strongly and significantly associated with mental health in adulthood. Our evidence is taken from eight Behavioral Risk Factor Surveillance System (BRFSS) surveys from 2009-2023 which contain a special supplement asking respondents to recall abuse in their childhood. We find that poor mental health is on the rise in the United States, particularly among young women. This upward trend among the young is also apparent from the National Health Interview Surveys 1997-2021, the Healthy Minds surveys of 2007-2023 and the Youth Risk Behavior Surveillance System (YRBSS) for high school students. We find ACEs are strongly correlated with poor mental health among both the young and older people and the effect is additive. The impact of living with a household member with poor mental health is large relative to other ACEs and is particularly pronounced among younger people. Being bullied, including electronically, is also strongly negatively associated with the wellbeing of high school students. Time spent in front of a screen has been rising over time for the young and has an independent negative impact on their mental health over and above bullying, one which is more pronounced for young women.
1.1.28 Mojtabai & Olfson (2024). Trends in Mental Disorders in Children and Adolescents Receiving Treatment in the State Mental Health System. Journal of the American Academy of Child & Adolescent Psychiatry.
ABSTRACT: OBJECTIVE: To examine recent trends in clinical diagnoses of children and adolescents receiving treatment in publicly funded mental health treatment services in the United States.
METHOD: Data on children and adolescents (≤17 years) receiving treatment from publicly funded mental health treatment services recorded in Mental Health Client Level Data (MH-CLD) 2013-2021 (total number of records=13,684,154) were used to examine temporal trends in the proportion of different child and adolescent psychiatric disorders. Trends were examined overall and in age, sex, racial/ethnic and service strata focusing on community-based programs.
RESULTS: The analyses revealed increases in the proportion with anxiety disorders from 9.6% in 2013 to 19.2% in 2021, AOR=2.17, 95% CI=1.85-2.55, p<0.001, trauma- and stressor-related disorders from 22.7% to 27.4%, AOR=1.31, 1.09-1.57, p=0.004, and depressive disorders from 13.4% to 17.0%, AOR=1.20, 1.03-1.41, p=0.04. During this same period, the proportion with bipolar disorders declined almost eight-fold from 10.0% to 1.3%; AOR=0.07, 0.06-0.09, p<0.001. The proportion with conduct disorders also declined from 9.7% to 4.4%; AOR=0.42, 0.32-0.55, p<0.001, and the proportion of oppositional-defiant disorder declined from 11.1% to 7.8%; AOR=0.79, 0.65-0.98, p=0.03. Trends varied across sex, age, and racial/ethnic strata.
CONCLUSION: The composition of childhood psychiatric diagnoses in patients within publicly funded mental health settings changed over the past decade. While some of the trends may reflect changes in diagnostic practices of clinicians, increases in anxiety and depressive disorders parallel trends in the prevalence of these conditions in the general population and highlight a growing need for identifying and treating these conditions in this age group.
1.1.29 Borg, Heffer, & Willoughby (2025). Generational Shifts in Adolescent Mental Health: A Longitudinal Time-Lag Study. Journal of Youth and Adolescence.
ABSTRACT: There is concern that adolescents today are experiencing a "mental health crisis" compared to previous generations. Research has lacked a longitudinal time-lag design to directly compare depressive symptoms and social anxiety of adolescents in two generations. The current study surveyed 1081 adolescents in the current generation (Mage = 14.60, SD = 0.31, 49% female) and 1211 adolescents in a previous generation (Mage = 14.40, SD = 0.51, 51% female) across the high school years (grades 9-12), 20 years apart. Mixed-effects analysis revealed that the Current-Sample reported higher and increasing mental health problems over time compared to the Past-Sample. Although most adolescents reported consistently low mental health problems, the Current-Sample had a higher proportion of adolescents who were consistently at risk across the high school years compared to the Past-Sample. These findings highlight while most adolescents in both generations do not report elevated mental health problems, there may be a small, yet growing, group of adolescents today at risk for experiencing a "mental health crisis".
[What are we missing?]
1.2 USA: COUNTERPOINTS AND CRITICISMS
1.2.1 Friedman (2018). The Big Myth About Teenage Anxiety. New York Times.
ABSTRACT/EXCERPT: “Despite news reports to the contrary, there is little evidence of an epidemic of anxiety disorders in teenagers… There are a few surveys reporting increased anxiety in adolescents, but these are based on self-reported measures — from kids or their parents — which tend to overestimate the rates of disorders because they detect mild symptoms, not clinically significant syndromes.”
COMMENT from Haidt: Friedman is surely right that the bar drifts down as “concept creep” expands psychiatric diagnoses. But he does not address the rise in self harm and suicide, which are not based on self-report. He also says that there has not been a major study of psychiatric disorders among teens in a decade, but see the studies listed above in section 1.1, and the many studies below done in the UK.
[What are we missing? So far Friedman is the only skeptic we’ve found, and nobody has offered data that indicates that the rise is not real]
1.3 UK: STUDIES AND DATASETS SHOWING A RISE IN RECENT YEARS IN TEEN DEPRESSION, ANXIETY, AND SUICIDAL IDEATION
1.3.1 Fink, Patalay, Sharpe, Holley, Deighton, & Wolpert (2015). Mental health difficulties in early adolescence: A comparison of two cross-sectional studies in England from 2009 to 2014. Journal of Adolescent Health.
ABSTRACT: PURPOSE: To examine the changes in mental health difficulties in early adolescence between 2009 and 2014 in England.
METHODS: Analysis reports on data from two cross-sectional samples of adolescents (aged 11-13 years) collected 5 years apart in 2009 and 2014 in secondary schools across England. Samples were matched using propensity scoring, resulting in a total pooled sample of 3,366 adolescents. Mental health difficulties were reported by participants using the Strengths and Difficulties Questionnaire.
RESULTS: Overall, there were similar levels of mental health difficulties experienced by adolescents in 2009 and 2014. Notable exceptions were a significant increase in emotional problems in girls and a decrease in total difficulties in boys in 2014 compared to 2009.
CONCLUSIONS: The increased prevalence of emotional problems in girls mirrors a trend found in other similar studies, and the results are discussed in the context of recent economic and societal changes. The small decrease in total difficulties in boys, although promising, clearly warrants further research.
NOTE: Emotional problems among girls increased from 13.1% in 2009 to 20.3% in 2014, a 55% increase in the rate. Some other difficulties were down, indicating that the change is specific to emotional problems. For boys, there was a small decrease in emotional problems, along with decreases in most of the other problems.
1.3.2 Patalay, & Gage, (2019). Changes in millennial adolescent mental health and health-related behaviours over 10 years: A population cohort comparison study. International Journal of Epidemiology.
ABSTRACT: …. Prevalences in mental health (depressive symptoms, self-harm, anti-social behaviours, parent-reported difficulties) and health-related behaviours (substance use, weight, weight perception, sleep, sexual intercourse) were examined at age 14 in two UK birth cohorts; Avon Longitudinal Study of Parents and Children (ALSPAC, N = 5627, born 1991–92) and Millennium Cohort Study (MCS, N = 11,318, born 2000–02).... Depressive symptoms (9% to 14.8%) and self-harm (11.8% to 14.4%) were higher in 2015 compared with 2005. Parent-reported emotional difficulties, conduct problems, hyperactivity and peer problems were higher in 2015 compared with 2005 (5.7–8.9% to 9.7–17.7%). Conversely, substance use (tried smoking, 9.2% to 2.9%; tried alcohol, 52.1% to 43.5%, cannabis, 4.6% to 3.9%), sexual activity (2% to 0.9%) and anti-social behaviours (6.2–40.1% to 1.6–27.7%) were less common or no different. Adolescents in 2015 were spending less time sleeping (<8 h 5.7% to 11.5%), had higher body mass index (BMI) (obese, 3.8% to 7.3%) and a greater proportion perceived themselves as overweight (26.5% to 32.9%). … Substantial increases in mental health difficulties, BMI and poor sleep-related behaviours highlight an increasing public health challenge.
Figure 1: UK 13- to 15-year-olds with clinically significant depression or engaging in self-harm behaviors, 2005 vs. 2015, by sex [Graphed by Jean Twenge]
NOTE: This study found roughly equal cohort increases for girls and boys when compared using odds ratios.
1.3.3 Mental Health of Children and Young People in England (2018). NHS Digital.
ABSTRACT: This survey series provides England’s best source of data on trends in child mental health. Major surveys of the mental health of children and young people in England were carried out in 1999, 2004, and 2017. While many surveys use brief tools to screen for nonspecific psychiatric distress or dissatisfaction, this series applied rigorous, detailed and consistent methods to assess for a range of different types of disorder according to International Classification of Disease (ICD-10) diagnostic criteria. All cases were reviewed by clinically-trained raters.
FIGURE:
Figure 1: Trends among 11- to 15-year-old English boys and girls for anxiety disorders and depressive disorders, at 3 points in time (1999, 2004, 2017). (Graphed by Jean Twenge. Note that we can’t tell where the elbow is, since we only have data from 3 specific years.)
UPDATE: Mental Health of Children and Young People in England (2021). NHS Digital.
A number of participants (N = 3570 in 2020, N = 3667 in 2021) from the 2017 survey agreed to take part in two online follow up surveys.
MAJOR FINDINGS FROM FOLLOW UP SURVEYS: “Rates of probable mental disorder in 6 to 16 year olds increased from 11.6% in 2017 to 17.4% in 2021. The increase was statistically significant in both boys and girls. Amongst those aged 17 to 19 years, rates of probable mental disorder rose from 10.1% in 2017 to 17.4% in 2021. The increase was significant in young women of this age (from 13.4% in 2017 to 24.8% in 2021), but not in young men. Change between 2017 and 2021 in rates of probable mental disorder could not be examined for those aged 20 to 22 years, for whom there was no 2017 data. Rates of probable and possible mental disorder remained broadly similar in 2020 and 2021, with no statistically significant change identified in any age group or by sex.”
NOTE: Between 2017 and 2020, the percentage of girls with a probable mental disorder doubled.
UPDATED 2023 DATA: (Download data)
NOTE: The Strengths and Difficulties Questionnaire (SDQ) was used to assess different aspects of mental health, including problems with emotions, behaviour, relationships, hyperactivity and concentration. Please note that the mental disorder prevalence estimates in the initial MHCYP 2017 survey reported on a different and more detailed diagnostic assessment of mental disorder. Therefore, any comparisons between 2017 and the subsequent waves must draw on the results based on a comparable measure (the SDQ) using children that were aged between 8 to 19 years at the time of each survey.
Figure made by Zach, from NHS data (see spreadsheet):
1.3.4 The Good Childhood Report (2018).
[[From the summary:] For about a decade from the mid-1990s, the gender gaps in happiness with appearance and life as a whole narrowed substantially, but in recent years these have been widening, so that the gap is now of a similar size to 20 years ago.]
UPDATE: The Good Childhood Report (2021)
Trends in children’s happiness with different aspects of life by gender, UK, 2009-10 to 2018-19 (ages 10-15)
NOTE: There are a few differences between the 2018 and 2021 figures. The 2018 figure used data from the British Household Panel Survey (BHPS) for the years 1995-2008. Both figures used data from the Understanding Society survey for years after 2008. Additionally, the researchers used three-year moving averages in the 2018 figure to minimise short-term fluctuations. For this reason, the data for 2009–10 to 2015–16 are slightly different between the two charts (and is why the two figures are not merged together).
UPDATE: The Good Childhood Report (2022)
Trends in children’s happiness with different aspects of life by gender, UK, 2009-10 to 2019-20 (ages 10-15)
UPDATE: The Good Childhood Report (2023) (ages 10-15)
Trends in children’s happiness with different aspects of life by gender, UK, 2009-10 to 2019-21
ADDITIONAL FIGURES FROM 2023 REPORT:
Predicted probability of children (aged 10 to 15) being unhappy (scoring below the midpoint) with different aspects of life:
QUOTE: “As reported in previous years, appearance was the aspect of life for which the largest proportion of children scored below the midpoint.” (15.9%)
1.3.5 Girlguiding Report
Guardian article, UK survey finds sharp decline in happiness of young women and girls
1.3.6 Cybulski, Ashcroft, Carr, Garg, Chew-Graham, Kapur, & Webb (2021). Temporal trends in annual incidence rates for psychiatric disorders and self-harm among children and adolescents in the UK, 2003–2018. BMC Psychiatry.
ABSTRACT: BACKGROUND: There has been growing concern in the UK over recent years that a perceived mental health crisis is affecting children and adolescents, although published epidemiological evidence is limited.
METHODS: Two population-based UK primary care cohorts were delineated in the Aurum and GOLD datasets of the Clinical Practice Research Datalink (CPRD). We included data from 9,133,246 individuals aged 1–20 who contributed 117,682,651 person-years of observation time. Sex- and age-stratified annual incidence rates were estimated for attention-deficit/hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) (age groups: 1–5, 6–9, 10–12, 13–16, 17–19), depression, anxiety disorders (6–9, 10–12, 13–16, 17–19), eating disorders and self-harm (10–12, 13–16, 17–19) during 2003–2018. We fitted negative binomial regressions to estimate incidence rate ratios (IRRs) to examine change in incidence between the first (2003) and final year (2018) year of observation and to examine sex-specific incidence.
RESULTS: The results indicated that the overall incidence has increased substantially in both boys and girls in between 2003 and 2018 for anxiety disorders (IRR 3.51 95% CI 3.18–3.89), depression (2.37; 2.03–2.77), ASD (2.36; 1.72–3.26), ADHD (2.3; 1.73–3.25), and self-harm (2.25; 1.82–2.79). The incidence for eating disorders also increased (IRR 1.3 95% CI 1.06–1.61), but less sharply. The incidence of anxiety disorders, depression, self-harm and eating disorders was in absolute terms higher in girls, whereas the opposite was true for the incidence of ADHD and ASD, which were higher among boys. The largest relative increases in incidence were observed for neurodevelopmental disorders, particularly among girls diagnosed with ADHD or ASD. However, in absolute terms, the incidence was much higher for depression and anxiety disorders.
CONCLUSION: The number of young people seeking help for psychological distress appears to have increased in recent years. Changes to diagnostic criteria, reduced stigma, and increased awareness may partly explain our results, but we cannot rule out true increases in incidence occurring in the population. Whatever the explanation, the marked rise in demand for healthcare services means that it may be more challenging for affected young people to promptly access the care and support that they need.
[Note that all classifications are based on the researchers’ analysis of medical records for anonymized cases within the cohort. The records were created by doctors. None of the data in this study is based on self-reports or parent reports]
Regraphed by Zach Rausch:
NOTE ADDED ON MARCH 31, 2023:
In a Twitter thread, prof. Lewis Appleby accused me of manipulating the original graphs from Cybulski to make my case stronger and hide inconvenient facts. He noted that I added the vertical line at 2012 (yes, as in many of my graphs). More seriously, he charged that “4 of 6 age-sex groups are removed, leaving a greater focus on young teenage girls. In fact… the % rise was greater in boys and, among girls, was highest in a younger group, aged 10-12 years. But then these findings don’t support the narrative about young teenage girls and their smartphones.”
This is untrue. As I noted above, I asked Zach Rausch to regraph the Cybulski et al. data on self-harm to put it in the standard form that I do in most of my graphs: Boys and Girls on the same graph, so that we can see when and where they diverge. That’s the graph above, and it’s the graph I used in my Substack posts. Is it misleading? What would happen if I graphed the other four age-sex groups? I asked Zach to do so, and here they are:
1.3.7 Walker (2022). Girls’ mental health “at a precipice” and increasingly worse than boys’, data shows. Steer Education. (Summary Findings)
REPORT HIGHLIGHTS:
- To date, the pandemic has had a 25 % adverse effect on students’ ability to self-regulate in school
- Girls are particularly affected, with a 33% decline, whilst boys’ self regulation outside school has actually improved during the pandemic
- By the time girls reach 18, they now have more than twice the self regulatory risks as boys of the same age
- The pandemic public exam cancellation provided a unique opportunity to measure the social-emotional impacts of public exams vs not having public exams. Data suggests that public exams have an adverse effect on student wellbeing
- Relatively stable levels of visible disruptive adolescent behaviours during the pandemic may be explained by an increase of hidden, internalised risks which are not yet visible
- Strong signals indicate that the specific long-term psychological risk from the pandemic will be pathologies driven by internalised control, particularly in girls aged 14-18
- Independent day schools saw some similar increases to state schools, but from a lower pre-pandemic baseline and to a lower extent
FIGURE:
1.3.8 Equality, diversity, and students characteristics data (2022). Office for Students.
EXCERPT: The proportion of undergraduate entrants reporting a mental health condition has risen by 3.7 percentage points since 2010-11, from 0.7 per cent to 4.4 per cent in 2020-21. This rate of increase is more than any other disability type.
FIGURE:
Updated survey through 2022: (download data)
Student populations data is shown for a population of undergraduate entrants for the time series between 2010-11 and 2021-22.
1.3.9 Mental health of children and young people. NHS Government Statistical Service.
FIGURES:
Figure 1. Data from this survey series reveal a slight increase over time in the prevalence of mental disorder in 5 to 15 year olds (the agegroup covered on all surveys in this series). Rising from 9.7% in 1999 and 10.1% in 2004, to 11.2% in 2017.
Figure 2. Emotional disorders have become more common in 5 to 15 year olds: up from 4.3% in 1999 and 3.9% in 2004, to 5.8% by 2017. The increase since 2004 in emotional disorders is evident in both boys and girls.
Figure 3. Young women have been identified as a high risk group in relation to mental health. This survey also found rates of emotional mental disorder and self-harm were higher in this group than other demographic groups. Nearly one in four (23.9%) 17 to 19 girls had a mental disorder. And 22.4% had an emotional disorder. One in eighteen (5.6%) young women were identified with body dysmorphic disorder (BDD). BDD is an anxiety disorder characterised by the obsessive idea that some aspect of one's body part or appearance is severely flawed and warrants exceptional measures to hie or fix.
Figure 4.
1.3.10 McElroy, Tibber, Fearon, Patalay, & Ploubidis (2023). Socioeconomic and sex inequalities in parent-reported adolescent mental ill-health: time trends in four British birth cohorts. J Child Psychol Psychiatry.
ABSTRACT PURPOSE: To investigate whether and how developmental trajectories of emotional problems have changed across generations.
METHODS: Used data from two UK prospective cohorts assessed 10 years apart: the Avon Longitudinal Study of Parents and Children (ALSPAC) including individuals born in 1991–92, and the Millennium Cohort Study (MCS) with individuals born in 2000–02. Our outcome was emotional problems, assessed using the parent-rated emotional subscale of the Strengths and Difficulties Questionnaire (SDQ-E). Data were available for 19 418 participants, of whom 9678 (49·8%) were female and 9740 (50·2%) were male.
RESULTS: Individuals born between 2000 and 2002 had higher emotional problem scores from around 9 years (intercept statistic β 1·75, 95% CI 1·71–1·79) than did individuals born in 1991–92 (1·55, 1·51–1·59). Differences between cohorts peaked overall at age 14 years. (...) There was little difference between the two cohorts for emotional problems in childhood at age 8 years but substantial and increasing differences from age 11 onwards.
CONCLUSIONS: Our comparison of two cohorts of young people provides evidence that compared with a cohort assessed 10 years prior, emotional problems emerge earlier in development in the more recent cohort, and these are especially pronounced for females during mid-adolescence. This study provides novel evidence that the developmental course of parent-rated emotional problems has changed between generations
1.3.11 McElroy, Tibber, Fearon, Patalay and Ploubidis, (2023), Socioeconomic and sex inequalities in parent-reported adolescent mental ill-health: time trends in four British birth cohorts. Journal of Child Psychol Psychiatry.
ABSTRACT: BACKGROUND: Studies using symptom-based screeners have suggested that mental ill-health has increased in adolescents in recent decades, however, few studies have tested the equivalence of their instruments, which is critical for inferring changes in prevalence.
METHODS: Using structural equation modeling, we explored SEP and sex differences in harmonized parent reports of emotional and behavioral problems, using data from four UK birth cohorts: the 1958 National Child Development Study (NCDS'58; n = 10,868), the 1970 British Cohort Study (BCS'70; n = 8,242), the 1991–92 Avon Longitudinal Study of Parents and Children (ALSPAC'91; n = 5,389), and the 2000–01 Millennium Cohort Study (MCS'01; n = 9,338).
RESULTS: Compared with the two earliest cohorts, members of MCS '01 [2000-01 cohort] had higher latent mean scores on emotional problems (both sexes), and lower scores on behavioral problems (females only). The associations between four indicators of SEP and emotional problems were strongest in MCS'01, with housing tenure having the strongest association. All four SEP indicators were associated with behavioral problems in each cohort, with housing tenure again more strongly associated with problems in the MCS'01. Mediation analyses suggested that the increase in emotional problems occurred despite broadly improving socioeconomic conditions.
CONCLUSIONS: Our findings suggest that parent reports of adolescent emotional problems, but not behavioral problems, have risen in recent generations and this trend is not solely due to reporting styles.
1.3.12 Dykxhoorn, Osborn, Walters, Kirkbride, Gnani, & Lazzarino (2023). Temporal patterns in the recorded annual incidence of common mental disorders over two decades in the United Kingdom: a primary care cohort study. Psychological Medicine.
ABSTRACT: BACKGROUND: Common mental disorders (CMDs) including depression, anxiety, and stress are very common, but it is unclear whether the last decades of social, economic, and political change have impacted incidence of CMD. This study explored temporal trends in the recorded incidence of CMD in the United Kingdom.
METHODS: We created a longitudinal cohort of over 29 million UK adults aged ≥16 years registered to a general practice using de-identified individual electronic health records (EHRs) from the Clinical Research Practice Datalink (CPRD) Aurum and Gold databases. We used data from general practices in the United Kingdom to estimate the annual recorded incidence of CMD for 2000–2020, including symptoms, diagnosis, or pharmaceutical treatment. Trends were explored by sex, age, ethnicity, region, deprivation, and comorbidity.
RESULTS: The recorded incidence of CMD episodes was 55.9 per 1000 person-years in 2000 [95% confidence interval (CI) 55.8–56.1], increasing to 79.6 per 1000 person-years in 2019 (95% CI 79.5–79.8). Females had higher recorded incidence rates, as did those living in more deprived areas. We observed striking patterns by age over time, with rates in ages 16–24 increasing from 40.2 per 1000 in 2000 to 107.8 per 1000 in 2019. In contrast, the rates in those aged ≥55 years decreased since 2014.
DISCUSSION: We observed an increasing incidence of recorded CMDs in the United Kingdom over the past 20 years, with important differences by population group. Much of this increase was observed before 2005, with stable rates from 2006 to 2019 and a decrease in 2020. Contrary to our hypothesis, we did not find clear evidence linking the temporal trends of recorded CMD incidence to the 2007–08 financial crisis and subsequent austerity measures.
[NOTE: From 2000 to 2019, rates of common mental disorders increased by 168.2% for 16-24 year olds in the UK.]
. Figure 1: Recorded annual incidence rate of common mental disorders (per 1000 person-years), by age.
Figure 2. Recorded annual incidence rate of common mental disorders (per 1000 person-years), by sex.
1.3.11 Collishaw (2015), Annual Research Review: Secular trends in child and adolescent mental health. J Child Psychol Psychiatry.
ABSTRACT: BACKGROUND: It is [] important to understand the extent to which the prevalence of mental health problems has changed over time, and to identify reasons behind any trends in mental health.
METHODS: This review evaluates evidence on whether the population prevalence of child and adolescent mental health problems has changed. The primary focus of the review is on epidemiological cross-cohort comparisons identified by a systematic search of the literature (using the Web of Knowledge database).
NOTE: There are several reasons for thinking that time trends in symptom reports do in fact reflect real changes in population prevalence of mental health problems. The first is that there is a convergence of evidence across multiple study methods and multiple informants. For example, police-recorded and victim-reported crime, as well as parent-reported conduct problems, have followed broadly similar trends. Second, it seems unlikely that there has been a general shift in openness about reporting symptoms, because there is considerable specificity in observed trends. For example, cross-cohort comparisons have demonstrated marked variations in rates of adolescent emotional and conduct problems, but much less so in symptoms of hyperactivity. This demonstrates that informants have not simply become more open about reporting difficult behaviour, or that parents have become more sensitive to troublesome behaviour in their children per se. The strongest evidence for real changes in mental health comes from studies that provide independent validation of informants’ reports and that directly compare the validity of reports over time. Such studies are exceedingly rare, but are in principle possible. As discussed above, parent-reported conduct problems increased substantially over time, but equally strongly predicted a broad range of independently assessed adult difficulties in successive longitudinal cohorts in the United Kingdom.
1.4 UK: COUNTERPOINTS AND CRITICISMS
1.4.1 [none found so far]
[What are we missing?]
1.5 WHAT IS HAPPENING IN OTHER ENGLISH-SPEAKING COUNTRIES?
1.5.1 Canada
See our separate Google doc that examines a variety of trends: The Coddling of the Canadian Mind. The short answer is that the same pattern is occurring among Canadian teens. Here are two figures from that document:
SOURCE: Ontario Student Drug Use and Health Survey (2019).
[NOTE: Self-rated mental health item: “How would you rate your emotional or mental health?” Response options were: Poor, Fair, Good, Very good, or Excellent. Figure 3.4.3 shows the percentage of students who rated their mental health as fair or poor.]
SOURCE: Garriguet (2021). Portrait of youth in Canada: Data report. Statistics Canada.
ABSTRACT: Using Statistics Canada data, this chapter goes beyond the measure of life expectancy to describe the health of Canadian youth, defined as those aged 15 to 30 years old. Indicators of physical and mental health and substance use are presented to highlight the health of young Canadians. Comparisons with older Canadians and trends mostly over the last 20 years are provided. When possible, indicators are also examined by sex, ethnicity and household income. The health of Indigenous youth is also important. However, to put results for this population in the proper context, a full chapter will be dedicated to Indigenous youth.
FIGURE:
SOURCE: Canadian Community Health Survey (CCHS)
[NOTE: Excellent or very good self-perceived general mental health: Respondents answered excellent or very good to the question “In general, would you say your mental health is... ?” The question was first asked in 2003.]
Regraphed by Zach Rausch:
1.5.2 Australia
See our separate Google doc that examines a variety of trends: The Coddling of the Australian Mind. The short answer is that there is clearly a rise in mood disorders for teen girls, but it’s not yet so clear for teen boys. Here is one figure from that document:
SOURCE: Psychological Distress in Young People in Australia Fifth Biennial Youth Mental Health Report: 2012-2020 (See Zach’s spreadsheet)
NOTES: 1. “Psychological distress” is based on a widely accepted measure of non-specific psychological distress, the Kessler 6.
1.3.4 Australia's Health Snapshots 2022: Mental Health of Young Australians (2022). AIHW. (Chapter 8 in Australia’s Health 2022). Regraphed by Zach Rausch (See spreadsheet)
% change since 2010 | |||
2010 | 2020 | % Change | |
Male (12-24) | 411 | 619 | 51% |
Female (12-24) | 558 | 1012 | 81% |
% change since 2010 | |||
2010 | 2020 | % Change | |
Male (16-24) | 11 | 17.3 | 57% |
Female (16-24) | 20 | 35.2 | 76% |
1.3.5 Botha, Morris, Butterworth, & Glozier (2023). Generational differences in mental health trends in the twenty-first century. Proceedings of the National Academy of Sciences.
ABSTRACT: Given the observed deterioration in mental health among Australians over the past decade, this study investigates to what extent this differs in people born in different decades—i.e., possible birth cohort differences in the mental health of Australians. Using 20 y of data from a large, nationally representative panel survey (N = 27,572), we find strong evidence that cohort effects are driving the increase in population-level mental ill-health. Deteriorating mental health is particularly pronounced among people born in the 1990s and seen to a lesser extent among the 1980s cohort. There is little evidence that mental health is worsening with age for people born prior to the 1980s. The findings from this study highlight that it is the poorer mental health of Millennials that is driving the apparent deterioration in population-level mental health. Understanding the context and changes in society that have differentially affected younger people may inform efforts to ameliorate this trend and prevent it continuing for emerging cohorts.
1.5.3 New Zealand
See our separate Google doc that examines a variety of trends: The Coddling of the Kiwi Mind. The short answer is that there is clear evidence of a rise in depression and anxiety, for both sexes, but the rise is larger for girls. Here are two figures from that document:
Source: New Zealand Ministry of Health, New Zealand Health Survey 2020, datasets can be downloaded here. Cameron How and Zach Rausch went through the reports for each year and extracted the relevant data points for the youngest age group (15-24) by sex to create these graphs. You can see their spreadsheets, with the relevant cells highlighted, here for anxiety, and here for depression.
Graph shows the percent who answered “yes” to the question: “Have you ever been told by a doctor that you have anxiety disorder? This includes panic attacks, phobia, post-traumatic stress disorder, and obsessive compulsive disorder?” [Note that this is a self-report of a medical diagnosis; it could reflect either changes in willingness to self-report, or willingness of doctors to make this diagnosis. But it is the only dataset I have been able to find that shows data from the key period in question, after the arrival of social media]
Same question, different age group:
Here are changes in anxiety by age group:
Graph shows the percent who answered “yes” to the question: “Have you ever been told by a doctor that you have depression?”
[What are we missing? ]
1.5.4 Other Anglosphere Countries
1.5.4.1 Scottish Government (2019). Exploring the reported worsening of mental wellbeing among adolescent girls in Scotland.
ABSTRACT: Adolescent girls in Scotland report lower levels of wellbeing than boys and this appears to have been worsening in recent years, particularly around the mid-teens. This has been found across surveys and using different indicators of wellbeing.
For example, the Scottish Adolescent Lifestyle and Substance Use Survey (SALSUS) found an increase in the proportion of 15 year old girls reporting high levels of emotional and behavioural difficulties in the last decade, especially since 2010 (Scottish Government, 2017b). In 2015, nearly 4 out of 10 (39%) of 15 year old girls had a borderline or abnormal total difficulties score, compared to 31% in 2006. Among 15 year old boys, 22% had a borderline or abnormal total difficulties score in 2006, compared to 28% in 2015.
…Using a different measure, SALSUS also found a reduction in 15 year old girls’ average levels of mental wellbeing between 2010 and 2015, and to a lesser extent among 13 year old girls (Scottish Government, 2017b). No such changes were observed in either 13 or 15 year old boys in this time period.
The Health Behaviours in School-Aged Children (HBSC) survey found that nearly a quarter of Scottish adolescents (23%) experienced two or more psychological complaints within the past week, with difficulty sleeping being particularly common (Cosma et al., 2016). Girls (30%) reported more frequent psychological health complaints than boys (17%), with the gender difference widening with age. The proportion of 13 year olds and particularly 15 year old girls reporting psychological health complaints has increased substantially since 2006.
Although the Scottish Health Survey has not reported on changes over time, data from 2012-2015 found that girls aged 13-15 had worse mental wellbeing than boys of the same age (Scottish Government, 2016). It also found that wellbeing decreased between age 13 and 15 for all children.
FIGURES:
1.5.4.2 O’Connor, Fitzgerald, & O’Reilly (2019). My World in Survey 2. National Study of Youth Mental Health in Ireland.
EXCERPT: MWS-2 (published in 2019) is the second wave of a national study of youth mental health in Ireland. My World Survey 1 (MWS-1) was published in 2012 and it offered an insight into the positive and negative factors related to the wellbeing of young people aged 12-25 years in Ireland then. Before MWS-1, no previously published study had comprehensively profiled youth mental health functioning in the community, exploring both protective and risk factors.
KEY FINDINGS:
- Levels of depression and anxiety in adolescents and young adults increased from MWS-1 to MWS-2. Adolescents and young adults in MWS-2 were much less likely to be in the normal range for depression and anxiety and much more likely to be in the moderate, severe or very severe ranges for depression and anxiety than adolescents and young adults in MWS-1.
- Levels of protective factors related to mental health such as self-esteem, optimism and resilience have decreased.
- Females, in particular, indicated increased levels of anxiety and decreased levels of self-esteem, body esteem, resilience and other protective factors than males of the same age.
- Young people from seldom heard groups showed a particular vulnerability with heightened anxiety and suicide attempts than their age-matched peers.
- Factors such as sleep, physical activity, social media use and pornography use were strongly associated with depression and anxiety.
- Fewer adolescents and young adults reported drinking alcohol in MWS-2 than MWS-1. However, adolescents in MWS-2 who reported more drinking engaged in more problematic drinking than adolescents in MWS-1. Alcohol use was strongly associated
STATS: The proportion of adolescents (aged 12 to 19) reporting severe anxiety doubled from 11% to 22% since 2012.
Levels of reported severe anxiety in young adults (aged 18 to 25) increased 11%, from 15% in 2012 to 26% in 2019.
1.5.4.3 Griffin, McMahon, McNicholas, Corcoran, Perry, & Arensman, E. (2018). Increasing rates of self-harm among children, adolescents and young adults: A 10-year national registry study 2007-2016. Social Psychiatry and Psychiatric Epidemiology. [Ireland]
ABSTRACT: PURPOSE: Rates of hospital-treated self-harm are highest among young people. The current study examined trends in rates of self-harm among young people in Ireland over a 10-year period, as well as trends in self-harm methods.
METHODS: Data from the National Self-Harm Registry Ireland on presentations to hospital emergency departments (EDs) following self-harm by those aged 10-24 years during the period 2007-2016 were included. We calculated annual self-harm rates per 100,000 by age, gender and method of self-harm. Poisson regression models were used to examine trends in rates of self-harm.
RESULTS: The average person-based rate of self-harm among 10-24-year-olds was 318 per 100,000. Peak rates were observed among 15-19-year-old females (564 per 100,000) and 20-24-year-old males (448 per 100,000). Between 2007 and 2016, rates of self-harm increased by 22%, with increases most pronounced for females and those aged 10-14 years. There were marked increases in specific methods of self-harm, including those associated with high lethality.
CONCLUSIONS: The findings indicate that the age of onset of self-harm is decreasing. Increasing rates of self-harm, along with increases in highly lethal methods, indicate that targeted interventions in key transition stages for young people are warranted.
1.5.4.4 Grimm et al. (2022). Improving children and young people’s mental health services: Local data insights from England, Scotland and Wales. The Health Foundation.
ABSTRACT: Mental health services are expanding, but not fast enough to meet rising needs, leaving many children and young people with limited or no support. Too little is known about who receives care and crucially, who doesn’t.
This briefing presents analysis from the Health Foundation’s Networked Data Lab (NDL) about children and young people’s mental health. The analysis from local teams across England, Scotland and Wales has highlighted three key areas for urgent investigation, to help ensure children and young people get the care they need. These are:
- rapid increases in mental health prescribing and support provided by GPs
- the prevalence of mental health problems among adolescent girls and young women
- stark socioeconomic inequalities across the UK.
To inform national policy decisions and local service planning and delivery, the quality of data collection, analysis and the linkage of datasets across services and sectors need to be improved and used more effectively.
[What are we missing? ]
1.6 WHAT IS HAPPENING OUTSIDE THE ANGLOSPHERE?
For country-specific data, see our additional collaboration docs: The European Adolescent Mood Disorders since 2010; The Coddling of the Spanish Mind; and many more. The studies included below only include cross-national studies. Also see Zach’s series on the youth International mental health crisis on After Babel.
1.6.1 Mental Health Million Project (2021). Mental State of the World Report 2021. Sapiens Lab.
SUMMARY: The Mental State of the World report provides insight into the mental wellbeing of Internet Enabled populations around the globe. This report for 2021 covers 223,087 respondents across 34 countries in the Core Anglosphere, Spanish-speaking Latin America, the Arab world, Spanish & French speaking continental Europe and Africa sampled in 2021 based on responses to the MHQ assessment in English, Spanish, French and Arabic. The MHQ provides an aggregate metric of mental wellbeing (the MHQ) as well as multiple dimensional views.
KEY FINDINGS:
- A smaller decline in mental wellbeing relative to 2020: Across the 8 English speaking countries measured from 2019 to 2021, there was a decline in mental wellbeing of 3% in 2021, smaller than the 8% decline in 2020. The decline across countries from 2019 to 2021 was significantly correlated to the stringency of Government Covid-19 measures and directionally correlated to the cases and deaths per million.
- Latin America and continental Europe reported the highest mental wellbeing: 8 out of 10 countries with the highest MHQ scores were located in Spanish-speaking Latin America and continental Spanish and French-speaking Europe. In contrast, 8 out of 10 countries with the lowest MHQ scores were from the English-speaking world. The relative MHQ scores of countries were significantly negatively correlated with cultural indicators such as Performance Orientation and Individualism as well as with core economic indicators such as GDP per capita and GNI per capita, correlations that were particularly strong for males aged 18-64.
- The alarming decline in mental wellbeing of younger generations was a global phenomenon: Whereas only 7% (6% to 9% across different regions) of those aged 65+ were Distressed or Struggling with their mental wellbeing at a level that impacts functioning and would be considered clinical, 44% (38% to 50% across different regions) of 18-24 year olds had mental wellbeing scores within this Distressed or Struggling range. This represents a growing gap between generations that, while present prior to the Covid-19 pandemic, has since been exacerbated. It also stands in stark contrast to the happiness and wellbeing patterns documented prior to 2010 across several regions of the world, where young adults 18-24 typically had the highest wellbeing.
- Higher education and employment were associated with higher MHQ scores: Across all countries, higher education was associated with higher mental wellbeing with those with Masters and PhDs scoring highest. In addition, being employed or self-employed was associated with substantially higher mental wellbeing compared to being unemployed, most significantly in the Core Anglosphere
FIGURES:
NOTE ON FIGURE: Lower MHQ scores are significantly correlated with national rates of suicide and sexual violence, and directionally with rates of violent assault of countries, as reported by United Nations Office on Drugs and Crime (UNODC). The correlations are particularly significant for young men 18-34 and with the dimension of Social Self. Left unmanaged, the present trend suggests that there could be substantial increases in these statistics."
1.6.2 Twenge, Haidt, Blake, McAllister, Lemon & Le Roy (2021). Worldwide increases in adolescent loneliness. Journal of Adolescence.
ABSTRACT: INTRODUCTION: Several studies have documented increases in adolescent loneliness and depression in the U.S., UK, and Canada after 2012, but it is unknown whether these trends appear worldwide or whether they are linked to factors such as economic conditions, technology use, or changes in family size.
METHODS: The Programme for International Student Assessment (PISA) survey of 15- and 16-year-old students around the world included a 6-item measure of school loneliness in 2000, 2003, 2012, 2015, and 2018 (n = 1,049,784, 51% female) across 37 countries.
RESULTS: School loneliness increased 2012–2018 in 36 out of 37 countries. Worldwide, nearly twice as many adolescents in 2018 (vs. 2012) had elevated levels of school loneliness. Increases in loneliness were larger among girls than among boys and in countries with full measurement invariance. In multi-level modeling analyses, school loneliness was high when smartphone access and internet use were high. In contrast, higher unemployment rates predicted lower school loneliness. Income inequality, GDP, and total fertility rate (family size) were not significantly related to school loneliness when matched by year. School loneliness was positively correlated with negative affect and negatively correlated with positive affect and life satisfaction, suggesting the measure has broad implications for adolescent well-being.
CONCLUSIONS: The psychological well-being of adolescents around the world began to decline after 2012, in conjunction with the rise of smartphone access and increased internet use, though causation cannot be proven and more years of data will provide a more complete picture.
See summary of the study in this New York Times op ed, which includes this graph:
1.6.3 Cosma… & de Looze, M. (2020). Cross-National Time Trends in Adolescent Mental Well-Being From 2002 to 2018 and the Explanatory Role of Schoolwork Pressure. Journal of Adolescent Health.
ABSTRACT: PURPOSE: Previous research has shown inconsistent time trends in adolescent mental well-being, but potential underlying mechanisms for such trends are yet to be examined. This study investigates cross-national time trends in adolescent mental well-being (psychosomatic health complaints and life satisfaction) in mainly European countries and the extent to which time trends in schoolwork pressure explain these trends.
METHODS: Data from 915,054 adolescents from 36 countries (50.8% girls; meanage = 13.54; standard deviationage = 1.63) across five Health Behaviour in School-aged Children surveys (2002, 2006, 2010, 2014, and 2018) were included in the analyses. Hierarchical multilevel models estimated cross-national trends in adolescent mental well-being and schoolwork pressure. We also tested whether schoolwork pressure could explain these trends in mental well-being.
RESULTS: A small linear increase over time in psychosomatic complaints and schoolwork pressure was found. No change in life satisfaction emerged. Furthermore, there was large cross-country variation in the prevalence of, and trends over time in, adolescent mental well-being and schoolwork pressure. Overall, declines in well-being and increases in schoolwork pressure were apparent in the higher income countries. Across countries, the small increase in schoolwork pressure over time partly explained the increase in psychosomatic health complaints.
CONCLUSIONS:Our findings do not provide evidence for substantial declines in mental well-being among adolescents. Yet, the small increase in mental well-being and increases in schoolwork pressure appear to be quite consistent across high-income countries. This calls for the attention of public health professionals and policy-makers. Country differences in trends in both adolescent mental well-being outcomes and schoolwork pressure were considerable, which requires caution regarding the cross-national generalization of national trends.
1.6.4 Rausch, Potrebny, & Haidt (2024). The Youth Mental Health Crisis is International Part 4: Europe. After Babel.
EXCERPT: We began our exploration of European mental health trends using the Health Behavior in School-Age Children Study (HBSC), a multinational survey that has tracked the physical and mental well-being of thousands of 11-, 13-, and 15-year-old adolescents since 2002 from 51 countries. We used a measure of high psychological distress from the HBSC that includes four items that tap different symptoms of psychological distress: feeling low, feeling nervous, feeling irritable, and having sleep difficulties. Respondents rated how often they felt these four symptoms over the last six months (1: Every day, 2: More than once/week; 3, About every week; 4, Every month; 5, Rarely or never). We operationalized high psychological distress as having three or more of the four psychological ailments every day or more than once a week over the last six months. Using this measure, we computed the average psychological distress score in each country, for girls and for boys separately, in each of the 5 survey years. This gave us two new datasets (one for girls, one for boys), each with 33 rows (one for each country), which facilitated the various cross-country comparisons that we report below. We weight each country equally, despite their population differences, because we see each country as a separate case study, each able to contribute information.
Figure. Percent of European students who experienced three or more psychological distress symptoms in the last week for at least six months. The percent change compares the average 2002-2010 scores with 2018 scores. Source: Health Behavior in School-Age Children Survey, 2002-2018. (See Zach’s spreadsheet for data points). Countries include: Austria, Belgium (Flemish Region), Belgium (Walloon Region), Italy, Malta, Poland, Portugal, Spain, Hungary, Croatia, Slovakia, Lithuania, Slovania, Luxembourg, France, Finland, Germany, Denmark, Sweden, United Kingdom (Wales), Ireland, Netherlands, United Kingdom (Scotland), Switzerland, Iceland, Estonia, United Kingdom (England), Norway, Latvia, Bulgaria, Russia, Romania, Ukraine, Greece, and North Macedonia.
Figure 2. Percent of students who experienced three or more psychological distress symptoms in the last week for at least six months by Eastern vs. Western Europe, split by sex. The percent change compares average 2002-2010 scores with 2018 scores. Source: Health Behavior in School-Age Children Survey, 2002-2018. (See Zach’s spreadsheet for data points).
1.6.5 Rausch & Haidt (2023). The Youth Mental Health Crisis is International Part 2: The Nordic Nations. After Babel.
EXCERPT: The Health Behaviour in School Age Children Survey includes four items that map onto different dimensions of psychological distress: feeling low, feeling nervous, feeling irritable, and having sleep difficulties. Respondents rated how often they felt these four symptoms over the last six months (1: Every day, 2: More once/week; 3, About every week; 4, Every month; 5, Rarely or never). With Thomas’s support, we organized the HBSC data to observe trends in high psychological distress (which we operationalized as having three or more of the four psychological ailments at least once a week over the last six months) among 11-15-year-old Nordic boys and girls since 2002. Countries include: Norway, Finland, Iceland, Sweden, and Denmark.
Figure. Percent of Nordic Teens with High Psychological Distress. Data from the Health Behavior in School-Age Children Survey (2002-2018). Graphs and data were organized, analyzed, and created by Thomas Potrebny and Zach Rausch. See 1.1.1 of Nordic Adolescent Mood Disorders since 2010.
1.6.6 Potrebny, Wiium, & Lundegård (2017). Temporal trends in adolescents’ self-reported psychosomatic health complaints from 1980-2016: A systematic review and meta-analysis. PLOS ONE.
ABSTRACT: OBJECTIVE: There is increasing concern that mental health may be deteriorating in recent generations of adolescents. It is unclear whether this is the case for self-reported psychosomatic health complaints (PSHC).
METHOD: We conducted a systematic review and meta-analysis of published primary studies on PSHC in the general adolescent population over time. The primary databases were MEDLINE, Embase and PsycINFO, which were searched from inception to November 2016. Studies were included if they involved an observational design, presented self-reported data from participants aged 10–19 years and included data from at least two time points, five years apart. Inclusion and study quality were assessed by two independent reviewers.
RESULTS: Twenty-one studies were included; 18 reported trends on the prevalence of PSHC in a single country, while three studies reported on multiple countries. In total, over seven million adolescents from 36 countries in Europe, North America, Israel and New Zealand were represented, covering the period 1982–2013. In the descriptive analysis, 10 studies indicated a trend of increasing PSHC, eight showed a stable trend and three showed a decreasing trend at certain points in time. The results from the meta-analysis showed a mean odds ratio (OR) of 1.04 (K = 139, 95% CI 1.01–1.08) for PSHC from 1982 to 2013, thus indicating a minor increase in general. In the subgroup analysis, this minor increase was observed mainly between the 1980s and 2000s, while the trend appeared to be more stable between the 2000s and 2010s. Some differences were also found between multinational subregions. Findings from subgroup analysis, however, only supported a significant increasing trend in Northern Europe.
CONCLUSIONS: There may have been a minor increasing trend in adolescent self-rated PSHC between the 1980 and 2000s, but has become more stable since the 2010s, from a multinational perspective. Northern Europe was the only region to show a clearly significant minor increasing trend, without being the region with the highest total prevalence of PSHC at the present time. The discrepant trends regarding PSHC between regions and the reliance on self-reported data may reflect true changes in the occurrence of PSHC in the adolescent population. However, they may also reflect changes in how adolescents perceive and report health complaints.
EXCERPT: Northern Europe showed a different trend than other multinational subregions. Indeed, it was the only region that showed a clearly significant minor increasing trend in adolescent PSHC between 1982 and 2013 (K = 46, OR 1.13, 95% CI 1.06–1.20). This increasing trend is also apparently similar in all nine Northern European countries included in this review across the Nordic countries, including Greenland and the Baltic states.
…The results from the meta-analysis showed a mean OR of 1.04 for PSHC between 1982 and 2013, thereby indicating a minor increasing trend in general. In subgroup analysis, this minor increasing trend was observed mainly from the 1980s to the 2000s, while the trend appeared to be more stable during the 2000s to the 2010s. Some differences were also found between multinational subregions, where some regions showed an increasing trend regarding PSHC (e.g., Western Europe, Northern Europe and “other regions”, that is, New Zealand and Israel), while others showed a more stable or decreasing trend (e.g., Eastern Europe, Southern Europe and North America). However, findings from the subgroup analysis only support a significant increasing trend in Northern Europe.
In the same time frame as this review, there have also been some notable indications of congruent trends among general adolescent populations in terms of increasing symptoms of mental health disorders, musculoskeletal pains and the increasing use of health services, for the diagnosis and treatment of both psychosomatic health complaints and mental health disorders, in high-income countries”
Fig 3. Forest plot for meta-analysed studies on adolescents’ PSHC, grouped by decade.
Table 4. Meta-analytic findings for studies on adolescents’ PSHC, by multi-national sub-region (Random effects model).
1.6.7 Ghai, Fassi, Awadh, & Orben (2023). Lack of Sample Diversity in Research on Adolescent Depression and Social Media Use: A Scoping Review and Meta-Analysis. Clinical Psychological Science.
ABSTRACT: Research on whether social media use relates to adolescent depression is rapidly increasing. However, is it adequately representing the diversity of global adolescent populations? We conducted a preregistered scoping review (research published between 2018 and 2020; 34 articles) to investigate the proportion of studies recruiting samples from the Global North versus Global South and assess whether the association between social media and depression varies depending on the population being studied. Sample diversity was lacking between regions: More than 70% of studies examined Global North populations. The link between social media and depression was positive and significant in the Global North but null and nonsignificant in the Global South. There was also little evidence of diversity within regions in both sampling choices and reporting of participants’ demographics. Given that most adolescents live in the Global South and sample diversity is crucial for the generalizability of research findings, urgent action is needed to address these oversights.
EXCERPT: It is remarkable that such little research comes from the Global South even though this area is home to the majority of the world’s adolescents and social media users (Statista, 2021; UNICEF Data, 2019). Hence, the lack of sample diversity in existing research can strongly bias the understanding of how social media use is associated with adolescents’ depression. This is especially the case given that attitudes toward depression and technology use will likely vary widely between the Global North and South (Altweck et al., 2015; Livingstone et al., 2019). Our meta-analysis, investigating whether the association of social media use and adolescents’ depression varied between the Global North and Global South, found—on average across all samples—a positive yet small correlation, a result aligned to previous literature (Orben, 2020). Among the factors that may explain such variability (e.g., gender-specific effects or methodological choices), our focus was on sample diversity, an often neglected yet crucial source of heterogeneous outcomes in clinical psychology. We found a positive and significant correlation between social media and depression in the Global North; in contrast, in the Global South, the correlation was null and nonsignificant. We ran moderation analysis to further explore the role of global area as a potential source of heterogeneity among the examined effect sizes. In this case, our results showed that global area (Global North vs. Global South) was not a significant moderator of the metacorrelation coefficient.
Together, these findings provide mixed evidence regarding whether differences exist in the association of depression and social media use between the Global North and the Global South. However, we want to highlight that the discrepancies emerging from these results might stem from the limited and highly heterogeneous correlation coefficients extracted from studies based in the Global South. Specifically, only seven out of 40 effect sizes were from this global area. Moreover, the metacorrelation coefficient for the Global South presented a broad confidence interval, spanning from a lower bound of –.14 to an upper bound of .29. These considerations, in turn, highlight the existing lack of literature on globally diverse samples. Therefore, more studies with samples from the Global South will be essential in allowing researchers to reach adequately powered conclusions about potential differences in the association between social media use and adolescent depression worldwide.
1.6.8 Bor, Dean, Najman, & Hayatbakhsh (2014). Are child and adolescent mental health problems increasing in the 21st century? A systematic review. The Australian and New Zealand Journal of Psychiatry.
ABSTRACT: OBJECTIVE: Up to one in five children experience mental health problems. Social and cultural factors may influence emergence of mental health problems. The 21st century has led to changes in many of these factors, but it is unclear whether rates of internalizing and externalizing problems have also changed in recent cohorts of young people.
METHODS: A comprehensive literature search was undertaken to locate cohort or population studies that examined changes in mental health of children over time, where participants were aged 18 years and under, and the time frame for change was at least 10 years, with data for at least one time point in the 21st century being statistically compared to at least one time point in the 20th century. Studies were reviewed for quality and outcome.
RESULTS: Nineteen studies met criteria for review. These included studies of toddlers, children, and adolescents. Seventeen studies examined internalizing problems, and 11 studies examined externalizing problems. For both children and toddlers, recent cohorts did not exhibit worsening of mental health symptoms. In adolescents, the burden of externalizing problems appear to be stable. However, the majority of studies report an increase in internalizing problems in adolescent girls. The findings for internalizing problems in boys were mixed.
CONCLUSIONS: These findings suggest that recent cohorts of adolescent girls are experiencing increases in internalizing symptoms compared to previous cohorts. Approaches for prevention and early intervention should be explored.
TABLE
STUDIES FROM TABLE 2:
- Henriksen, Nielsen, & Bilenberg (2012). New Danish standardization of the Child Behaviour Checklist. Danish Medical Journal.
- Tick, van der Ende, & Verhulst (2008) Ten-year trends in self reported emotional and behavioral problems of Dutch adolescents. Social Psychiatry and Psychiatric Epidemiology.
- Sweeting, Young, & West (2009) GHQ increases among Scottish 15 year olds 1987–2006. Social Psychiatry and Psychiatric Epidemiology.
- Collishaw, Maughan, Natarajan. & Pickles (2010). Trends in adolescent emotional problems in england: a comparison of two national cohorts twenty years apart. Journal of Child Psychology and Psychiatry.
- Hagquist (2009). Psychosomatic health problems among adolescents in Sweden – are the time trends gender related? European Journal of Public Health/
- Sourander…, & Lindroos (2012). Changes in adolescents mental health and use of alcohol and tobacco: a 10-year timetrend study of Finnish adolescents. European Child and Adolescent Psychiatry.
1.6.9 Uddin, Lee, Khan, Tremblay, & Khan (2020). Clustering of lifestyle risk factors for non-communicable diseases in 304,779 adolescents from 89 countries: A global perspective. Preventive Medicine.
ABSTRACT: The precursors of non-communicable diseases (NCDs) are often manifested during childhood and adolescence with little knowledge about co-occurrence of their related lifestyle risk factors. To address this deficit, we estimated the prevalence and clustering of six major NCD-risk factors in adolescents around the world. Data from the Global School-based Student Health Survey, collected between 2007 and 2016, were analysed in 304,779 adolescents aged 11–17 years (52.2% females) from 89 countries. We compared the observed (O) to expected (E) prevalence ratios of 64 possible combinations of six risk factors to determine their clustering patterns. Overall, 82.4% (95% CI 82.1–82.7) of adolescents had ≥2 risk factors, while 34.9% (34.6–35.3) had ≥3. Adolescents aged 16–17 years, compared to those aged 11–13 years, had higher odds (OR 1.33; 95% CI 1.31–1.36) of reporting ≥3 risk factors. Risk factors clustered in multiple combinations and differed by sex. The clustering of physical inactivity and low fruit and vegetable intake was evident in both males (O/E 1.10; 95% CI 1.07–1.12) and females (1.08; 1.06–1.10). The co-occurrence of cigarette smoking, alcohol drinking, physical inactivity, and low fruit and vegetable intake was 165% greater in females (2.65; 2.28–3.07) and 110% greater in males (2.10; 1.90–2.32) than expected. Globally, adolescents exhibit multiple modifiable risk factors for future development of NCDs. Early gender-specific prevention strategies targeting clusters of lifestyle risk factors should be prioritised to help mitigate future burden of NCDs globally. Periodical collection of behavioural risk factor data should be encouraged to facilitate a sustainable global surveillance.
EXCERPT: In regional analyses (Table 1), adolescents of the Americas reported the highest prevalence of ≥3 risk factors ( 5 6 . 2 % [ 5 5 . 5 – 5 6 . 9 ]), followed by adolescents of Western Pacific region (44.5% [ 4 3 . 7 – 4 5 . 4 ]). Three in 10 adolescents in Eastern Mediterranean and Southeast Asia reported having had ≥3 risk factors. Males were more likely to report ≥3 risk factors in Western Pacific and Southeast Asia regions. With older age, the prevalence of ≥3 risk factors became higher in all region s .
… The two most prominent lifestyle risk factors for NCDs were low fruit /vegetable intake and physical inactivity (86 % and 85% , respectively), similar to what has been reported elsewhere (Caleyachetty et al., 2015 ) . Prevalence of ≥3 lifestyle risk factors was more common in males than females. The strongest clustering of pairwise risk factors was cigarette smoking and alcohol consumption in both sexes. When multiple risk factors were considered simultaneously, different patterns of clusters were observed. Globally, physical inactivity and low fruit/vegetable consumption were clustered together as the most common risk factors in both sexes. In addition, clustering of all six risk factors were more than expected in both male s and females, indicating the need to develop global strategies for primary prevention of multiple risk factors.
1.6.10 Biswas… & Mamun (2022). Prevalence of multiple non-communicable diseases risk factors among adolescents in 140 countries: A population-based study. EClinicalMedicine.
ABSTRACT: BACKGROUND: Modifiable non-communicable disease (NCD) risk factors are becoming increasingly common among adolescents, with clustering of these risk factors in individuals of particular concern. The aim of this study was to assess global status of clustering of common modifiable NCD risk factors among adolescents.
METHODS: We used latest available data from nationally representative survey for 140 countries, namely the Global School-based Student Health Survey, the Health Behaviour in School-Aged Children and the longitudinal study of Australian Children. Weighted mean estimates of prevalence with corresponding 95% confidence intervals of nine NCD risk factors - physical inactivity, sedentary behaviour, insufficient fruits and vegetable consumption, carbonated soft drink consumption, fast food consumption, tobacco use, alcohol consumption and overweight/obesity - were calculated by country, region and sex.
FINDINGS: Over 487,565 adolescents, aged 11–17 years, were included in this study. According to trend analysis, prevalence of four or more NCD risk factors increased gradually over time. Prevalence of four or more NCD risk factors was 14.8% in 2003–2007 and increased to 44% in 2013–2017, an approximately three-fold increase (44.0%). Similar trends were also observed for three and two risk factors. Large variation between countries in the prevalence of adolescents with four or more risk factors was found in all regions. The country level range was higher in the South-East Asia Region (minimum Sri Lanka = 8%, maximum Myanmar = 84%) than Western Pacific Region (minimum China = 3%, maximum Niue = 72%), European Region (minimum Sweden = 13.9%, maximum Ireland = 66.0%), African Region (minimum Senegal = 0.8%, maximum Uganda = 82.1%) and Eastern Mediterranean Region (minimum Libya = 0.2%, maximum Lebanon = 80.2%). Insufficient vegetable consumption, insufficient fruit consumption and physically inactivity were three of the four most prevalent risk factors in all regions.
INTERPRETATION: Our results suggest a high prevalence of four or more NCD risk factors in adolescents globally, although variation was found between countries. Results from our study indicate that efforts to reduce adolescent NCD risk factors and the associated health burden need to be improved. These findings can assist policy makers to target the rollout of country- specific interventions.
EXCERPT: We found that insufficient fruit and vegetable consumption, and lack of physical activity were the most prevalent risk factors across all regions. In almost all countries, more than 40% of adolescents were physically inactive, and had insufficient fruit and vegetable consumption. Appropriate nutrition during adolescence is critical for current, future as well as intergenerational health… In this study, we found high prevalence of overweight/obesity in all regions. Data demonstrated that prevalence of overweight and obesity rises during mid-adolescence, which continues into early adulthood. This is an impending public health problem requiring further action as adolescent obesity strongly predicts adult obesity and associated morbidities. Addressing this issue at a population level will critically avert potential long-term impacts of adolescent obesity.
FIGURE:
1.6.11 Piao, Huang, Han, Li, Xu, Liu, & He (2022). Alarming changes in the global burden of mental disorders in children and adolescents from 1990 to 2019: A systematic analysis for the Global Burden of Disease study. European Child & Adolescent Psychiatry.
ABSTRACT: Mental disorders account for a large and increasing health burden worldwide, as shown in the Global Burden of Diseases (GBD) Study 2010. Unpacking how this burden in children and adolescents varies with sex, geographical regions, and ethnicities and how it has changed in the last 3 decades are important to improve the existing public health policies and prevention strategies. The study was conducted using GBD 2019 database. The burden of children and adolescents’ (< 20 years old) mental disorders was displayed as prevalence, incidence, disability-adjusted life-years (DALYs), years of life lost, and years lived with disability globally between 1990 and 2019. The number of DALYs in children and adolescents diagnosed with mental disorders was 21.5 million (95% CI: 15.2–29.6 million) in 2019. From 1990 to 2019, the age-standardized rates of DALYs of mental disorders increased from 803.8 per 100,000 (95% CI: 567.7–1104.3 per 100,000) to 833.2 per 100,000 (95% CI: 589.0–1146.1 per 100,000) population. Over the past 30 years, there had been a huge increase in the number of individuals suffering from anxiety disorders, major depressive disorders, and conduct disorders including an alarming increase in the rate of eating disorders such as 24.3% in bulimia nervosa and 17.0% in anorexia nervosa. Globally, 8.8% of children and adolescents have been diagnosed with varieties of mental illnesses, accounting for a heavy disease burden on public health. Besides, the worldwide increasing rates of anxiety disorders, major depressive disorders, and eating disorders have brought considerable challenges to public health undertakings, for which further prevention and treatment countermeasures are urgently needed.
1.6.12 Rudolf & Bethmann (2023). The Paradox of Wealthy Nations’ Low Adolescent Life Satisfaction. Journal of Happiness Studies.
ABSTRACT: Using PISA 2018 data from nearly half a million 15-year-olds across 72 middle- and high-income countries, this study investigates the relationship between economic development and adolescent subjective well-being. Findings indicate a negative log-linear relationship between per-capita GDP and adolescent life satisfaction. The negative nexus stands in stark contrast to the otherwise positive relationship found between GDP per capita and adult life satisfaction for the same countries. Results are robust to various model specifications and both macro and micro approaches. Moreover, our analysis suggests that this apparent paradox can largely be attributed to higher learning intensity in advanced countries. Effects are found to be more pronounced for girls than for boys.
EXCERPT: Table 2 presents the results of estimating Equation (1) over a sample of up to 72 countries. Column (1) shows a negative and statistically significant relationship between GDP per capita and adolescent life satisfaction. In particular, the coefficient of -0.425 (p=0.000) suggests that a doubling of per-capita GDP is associated with a 0.295-points (48 percent of a SD) lower level of adolescent life satisfaction. Per-capita GDP alone is able to explain 21.1 percent in the variation in adolescent life satisfaction across countries. Column (2) excludes Vietnam since no test scores have been published for Vietnam. In column (3), education variables were added to the model, and several observations can be made. First, including education variables reduces the effect of per capita GDP on adolescent life satisfaction by more than half, suggesting that education appears to be a potential mechanism behind the paradox. Second, PISA test scores are negatively and statistically significantly related to adolescent life satisfaction in a country. Moreover, higher levels of competition among students are found to be negatively related to adolescent life satisfaction, while co-operation among students has the opposite effect. Adding education variables raises the R2 by 20 percentage points
1.6.13 Marquez, & Long (2021). A Global Decline in Adolescents’ Subjective Well-Being: A Comparative Study Exploring Patterns of Change in the Life Satisfaction of 15-Year-Old Students in 46 Countries. Child Indicators Research.
ABSTRACT: There is a growing body of research that demonstrates declines in subjective well-being and increases in mental health problems among children and young people in recent decades. However, there is little comparative research examining changes in adolescents’ life satisfaction (LS) across a large number of countries, and critically, how this differs across sociodemographic groups. This study addresses this question by investigating changes in the LS of 15-year-old students between 2015 and 2018, with particular attention given to differences by gender, socio-economic status, immigrant background and urbanity. Data for this study come from the Programme for International Student Assessment (PISA). Due to the skewed nature of LS scale variables, the current study includes both mean levels of LS in a 0 to 10 scale, and the proportion of students reporting low LS (5 points or less). Linear regression models were used. Results demonstrate a global decline in mean levels of LS in 39 out of the 46 countries. In most countries, mean LS declined more among girls than among boys. Mean LS declined more, and the proportion of students reporting low LS increased more, among non-immigrant students and those of higher SES in the majority of countries. Findings regarding rural or urban communities were mixed. We advise that heterogeneity across all sociodemographic groups needs to be accounted for in public policy efforts to increase LS among young people.
EXCERPT: In terms of gender, overall changes in LS between 2015 and 2018 were more negative among girls than among boys. Specifically, the decline in mean levels of LS was greater among girls in approximately 67% of the countries. However, gender differences in the proportion of students with low LS was less pronounced, with approximately 41% of the countries demonstrating a greater decline among girls, and approximately 44% indicating greater changes among boys. This suggests that although girls overall seemed to fare worse in LS between 2015 and 2018, in a non-negligible portion of countries, more boys than girls experienced more negative changes in their satisfaction with life.
1.6.14 Blanchflower, Bryson, & Xu (2024). The Declining Mental Health Of The Young And The Global Disappearance Of The Hump Shape In Age In Unhappiness (Working Paper 32337). National Bureau of Economic Research.
ABSTRACT: Across many studies subjective well-being follows a U-shape in age, declining until people reach middle-age, only to rebound subsequently. Ill-being follows a mirror-imaged hump-shape. But this empirical regularity has been replaced by a monotonic decrease in illbeing by age. The reason for the change is the deterioration in young people’s mental health both absolutely and relative to older people. We reconsider evidence for this fundamental change in the link between illbeing and age with micro data for the United States and the United Kingdom. Beginning around 2011 there is a monotonic and declining cross-sectional association between well-being and age. In the UK the recent COVID pandemic exacerbated the trends by impacting most heavily on the wellbeing of the young, but this was not the case in the United States. We replicate the decrease in illbeing by age across 34 countries, including the United States and the United Kingdom, using five ill-being metrics for the period 2020-2024 and confirm the findings.
1.6.15 Blanchflower & Bryson (2024). The Global Loss of the U-Shaped Curve of Happiness. After Babel.
EXCERPT: There is a literature of at least 600 published papers suggesting that happiness is U-shaped in age and, conversely, that unhappiness is hump-shaped in age. In other words, when a person is young, they are as happy as they are going to be until old age (on average; individual lives vary). For many, their student days were the happiest days. The evidence of a midlife low period centered around age 50 had been found across 146 countries over the period 1973-2017. Across a variety of datasets and measures, the finding of a midlife low has been consistently replicated. The U-shape has been apparent across a whole range of well-being metrics, including life satisfaction, financial satisfaction, worthwhileness, and happiness. Every U.S. state had a U-shape.
But not anymore. It seems that the well-being of young adults (ages 18-25), especially young women, went into precipitous decline beginning around 2017 (with some evidence showing around 2014, see Figure 1).
FIGURE:
[What are we missing? ]
* * * * * * * * * * * * * * * * * * * * * * * * *
SECTION 2: SELF HARM AND SUICIDE ATTEMPTS
2.1 USA: STUDIES AND DATASETS SHOWING A RISE IN RECENT YEARS IN TEEN SELF HARM AND SUICIDE ATTEMPTS
For help in understanding the phenomenon of self harm, see this article in the New York Times, Getting a Handle on Self-Harm. It is not usually a suicide attempt; it is more often a way to blunt anxiety.
2.1.1 Mercado, Holland, Leemis, Stone, & Wang (2017). Trends in Emergency department visits for nonfatal self-inflicted injuries among youth aged 10 to 24 years in the United States, 2001-2015. Journal of the American Medical Association.
MAIN FINDING: “Youth self-inflicted injury ED visit rates were relatively stable before 2008. However, rates among females significantly increased thereafter—particularly among females aged 10 to 14 years, who experienced an 18.8% annual increase from 2009 to 2015.”
FIGURES (copied from the article, with labels added by Haidt):
Updated figures with data up through 2021 can be seen below. These data come from the most recently updated NEISS-AIP reports and were collected from the CDC’s interactive publicly accessible database, “” The figures were made by Zach Rausch.
Mercado, Holland, Leemis, Stone, & Wang (2017). Trends in Emergency Department Visits For Nonfatal Self-Inflicted Injuries. |
2.1.2 Plemmons, Hall, Doupnick, et al. (2018). Hospitalization for suicide ideation or attempt: 2008–2015.
ABSTRACT: OBJECTIVES: Suicide ideation (SI) and suicide attempts (SAs) have been reported as increasing among US children over the last decade. We examined trends in emergency and inpatient encounters for SI and SA at US children’s hospitals from 2008 to 2015.
METHODS: We used retrospective analysis of administrative billing data from the Pediatric Health Information System database.
RESULTS: There were 115,856 SI and SA encounters during the study period. Annual percentage of all visits for SI and SA almost doubled, increasing from 0.66% in 2008 to 1.82% in 2015 (average annual increase 0.16 percentage points [95% confidence intervals (CIs) 0.15 to 0.17]). Significant increases were noted in all age groups but were higher in adolescents 15 to 17 years old (average annual increase 0.27 percentage points [95% CI 0.23 to 0.30]) and adolescents 12 to 14 years old (average annual increase 0.25 percentage points [95% CI 0.21 to 0.27]). Increases were noted in girls (average annual increase 0.14 percentage points [95% CI 0.13 to 0.15]) and boys (average annual increase 0.10 percentage points [95% CI 0.09 to 0.11]), but were higher for girls. Seasonal variation was also observed, with the lowest percentage of cases occurring during the summer and the highest during spring and fall.
CONCLUSIONS: Encounters for SI and SA at US children’s hospitals increased steadily from 2008 to 2015 and accounted for an increasing percentage of all hospital encounters. Increases were noted across all age groups, with consistent seasonal patterns that persisted over the study period. The growing impact of pediatric mental health disorders has important implications for children’s hospitals and health care delivery systems.
FIGURE: Shows girls’ rates rising more than boys, and shows seasonal variation (low when school is not in session over summer, and at Christmas break). Counterpoint; this graph shows steady increases since 2008, rather than showing an elbow around 2011 or 2012.
2.1.3 Kalb, Stapp, Ballard, Holingue, Keefer, & Riley (2019). Trends in Psychiatric Emergency Department Visits Among Youth and Young Adults in the US. Pediatrics.
ABSTRACT: BACKGROUND: Visits to the emergency department (ED) for psychiatric purposes are an indicator of chronic and acute unmet mental health needs. In the current study, we examined if psychiatric ED visits among individuals 6 to 24 years of age are increasing nationwide.
METHODS: ED data came from the 2011–2015 National Hospital Ambulatory Medical Care Survey, a national survey of ED visits across the United States. Psychiatric ED visits were identified by using the International Classification of Diseases, Ninth Revision and reason-for-visit codes. Survey-weighted logistic regression analyses were employed to examine trends in as well as correlates of psychiatric ED visits. Data from the US Census Bureau were used to examine population rates.
RESULTS: Between 2011 and 2015, there was a 28% overall increase (from 31.3 to 40.2) in psychiatric ED visits per 1000 youth in the United States. The largest increases in psychiatric ED visits per 1000 US youth were observed among adolescents (54%) and African American (53%) and Hispanic patients (91%). A large increase in suicide-related visits (by 2.5-fold) was observed among adolescents (4.6–11.7 visits per 1000 US youth). Although psychiatric ED visits were long (51% were ≥3 hours in length), few (16%) patients were seen by a mental health professional during their visit.
CONCLUSIONS: Visits to the ED for psychiatric purposes among youth are rising across the United States. Psychiatric expertise and effective mental health treatment options, particular those used to address the rising suicide epidemic among adolescents, are needed in the ED.
Figures taken from the “video abstract”:
For all age groups in the study, by sex:
--By age: older group begins rising in 2014, 12-17 year olds begin rising in 2015 (but this is just one year of rise, so it not yet compelling. We’ll need to see data from 2016 and 2017)
--Focusing just on visits related to suicide attempts or ideation: rate increasing only for 12-17 year olds; it more-than-doubles between 2012 and 2015. No increase for the older group (who are millennials, not Gen Z). Also no increase for the youngest (aged 6-11), which suggests that the spike in self-harm for the youngest group (aged 10-14) shown in 2.1.1 is probably driven entirely by its oldest girls (aged 13 and 14), not by its youngest kids (the 10 and 11 year olds who show up within the purple line of the graph immediately below).
2.1.4 Burstein, Agostino, & Greenfield (2019). Suicidal Attempts and Ideation Among Children and Adolescents in US Emergency Departments, 2007-2015. JAMA Pediatrics.
ABSTRACT: [No abstract; text selected from results section:] We performed a repeated cross-sectional analysis of the National Hospital Ambulatory Medical Care Survey (NHAMCS) ED database from 2007 to 2015. [THIS IS THE SAME DATASET USED IN 2.1.2]... Over the 9-year study period, there were 59 921 unweighted ED visits for children younger than 18 years in the NHAMCS, among which 1613 (2.8%; 95% CI, 2.5%-3.0%; range, 161-198 observations annually) met the inclusion criteria for SA/SI visits. The median age was 13 years (interquartile range, 8-15 years). Most were evaluated in nonteaching and nonpediatric hospitals (Table). Notably, 43.1% of SA/SI visits were for children aged 5 to younger than 11 years and only 2.1% were hospitalized. The estimated annual visits for SA/SI between 2007 and 2015 (Figure) increased from 580 000 to 1.12 million (92.1%; 95% CI, 68.9%-130.3%; P for trend = .004). Conversely, there was no statistically significant change in total ED visits during this time (26.9 million to 31.8 million; 18.2%; 95% CI, −5.4% to 42.2%; P for trend = .67).
Figure: [Note that it’s only in 2011 that emergency room visits for suicide attempts and suicide ideation begin to rise, relative to ER visits for all causes]
2.1.5 Spiller, Ackerman, Spiller, & Casavant (2019). Sex- and Age-specific Increases in Suicide Attempts by Self-Poisoning in the United States among Youth and Young Adults from 2000 to 2018. The Journal of Pediatrics.
ABSTRACT: Study design: Retrospective review of intentional suspected-suicide self-poisoning cases reported to the National Poison Data System from US poison centers from 2000 to 2018 for patients 10-24 years old. For comparison of annual rates, population data by year of age were obtained from the US Census Bureau. We evaluated changes in the annual incidence, the annual rate per 100 000 population, and the medical outcome by patient age and sex.
RESULTS: There were 1 627 825 intentional suspected-suicide self-poisoning cases, of which 1 162 147 (71%) were female. In children 10-15 years old from 2000 to 2010, there was a decrease in number and rate per 100 000 population followed by a significant increase (from 125% to 299%) from 2011 to 2018. In children 10-18 years old, the increase from 2011 to 2018 was driven predominantly by females. In 19-24 years old age groups, there was a temporal delay and reduced increase in slope compared with the younger groups. There were 340 563 moderate outcomes, 45 857 major outcomes, and 1404 deaths. The percentage of cases with a serious outcome, major effect, or death increased over time and with age.
CONCLUSIONS: The incidence and rate of suicide attempts using self-poisoning in children less than 19 years old increased significantly after 2011, occurring predominantly in young girls. There has been an increase in the severity of outcomes independent of age or sex.
NOTES: Similar to other studies, increases begin after 2011 and are largest among girls, especially the younger girls. Self-poisonings quadrupled for girls ages 10 to 12 between 2011 and 2017, and doubled for girls ages 13 to 15. Increases are present but progressively smaller with age for girls ages 16 to 24. Boys have a lower base rate, but also show increases beginning after 2012.
Sample figures:
Comment: In the discussion, the authors mention the rise of smartphones and social media, citing previous research in the area. They also note: “Access to knowledge about specific methods to engage in lethal self-harm and exposure to the suicidal behavior of peers has undoubtedly increased with the ability to search online for readily available information, publicly posted information, and the increase of what is shared on social media networks. Youth are now increasingly exposed online to others struggling with suicidal ideation and engaging in self-injurious behaviors. Research on suicide contagion suggests that such exposure can increase suicidal behavior within peer networks and that youth are particularly vulnerable to suicide contagion. There is also an assortative component at play where youth most at risk for engaging in suicidal behavior are linking up with other youth who share information about their own suicidal thoughts or behaviors with minimal supervision or professional support to help navigate a crisis. This factor may contribute to the continued increase in the percentage of serious medical outcomes seen in our study.”
Previous theorizing has focused on cyberbullying, upward social comparison, and digital media taking time away from non-screen activities for why the rise of smartphones and social media might be linked to the rise in mental health issues among youth. This paper suggests another possible mechanism for the rise of smartphones and social media impacting mental health: These technologies have made information about self-harm techniques more accessible, as well as increased the potential for suicide contagion. Perhaps youth, especially the youngest groups, began to access that information only after they had their own smartphones or tablets (and thus could search for it without interference from parents), and/or only after more were on social media frequently.
That fits the time sequence (the iPad was introduced in 2010, the smartphone was used by the majority of Americans starting in 2012, and daily social media use exceeded 66% among teens sometime between 2011 and 2014, depending on the age group). The question is why the trend is stronger for girls. Girls use social media much more than boys, for one thing. Perhaps suicide contagion online is a stronger mechanism for girls?
2.1.6 Lo, Bridge, Shi, Ludwig, & Stanley, (2020). Children’s mental health emergency department visits: 2007–2016. Pediatrics.
ABSTRACT: BACKGROUND AND OBJECTIVES: Emergency department (ED) visits for children seeking mental health care have increased. Few studies have examined national patterns and characteristics of EDs that these children present to. In data from the National Pediatric Readiness Project, it is reported that less than half of EDs are prepared to treat children. Our objective is to describe the trends in pediatric mental health visits to US EDs, with a focus on low-volume, nonmetropolitan EDs, which have been shown to be less prepared to provide pediatric emergency care.
METHODS: Using 2007 to 2016 Nationwide Emergency Department Sample databases, we assessed the number of ED visits made by children (5–17 years) with a mental health disorder using descriptive statistics. ED characteristics included pediatric volume, children’s ED classification, and location.
RESULTS: Pediatric ED visits have been stable; however, visits for deliberate self-harm increased 329%, and visits for all mental health disorders rose 60%. Visits for children with a substance use disorder rose 159%, whereas alcohol-related disorders fell 39%. These increased visits occurred among EDs of all pediatric volumes, regardless of children’s ED classification. Visits to low-pediatric-volume and nonmetropolitan areas rose 53% and 41%, respectively.
CONCLUSIONS: Although the total number of pediatric ED visits has remained stable, visits among children with mental health disorders have risen, particularly among youth presenting for deliberate self-harm and substance abuse. The majority of these visits occur at nonchildren’s EDs in both metropolitan and nonurban settings, which have been shown to be less prepared to provide higher-level pediatric emergency care.
FIGURE: Percent change of pediatric ED visits among mental health subgroups. Line graph depicting yearly percentage change from 2007 of ED visits made by US children and those with at least one of the following mental health disorders: any mental health disorder, substance use disorder, and deliberate self-harm diagnosis.
2.1.7 Goodwin, Weinberger,, Kim, Wu, & Galea (2020). Trends in anxiety among adults in the United States, 2008-2018: Rapid increases among young adults. Journal of Psychiatric Research.
ABSTRACT: INTRODUCTION: In a time of global uncertainty, understanding the psychological health of the American public is imperative. There are no current data on anxiety trends among adults in the United States (US) over time. This study aimed to investigate prevalence of anxiety among US adults from 2008-2018.
METHODS: Data from the National Survey on Drug Use and Health (NSDUH), which is an annual, cross-sectional survey on substance use and mental health in the US, were analyzed in 2020. Prevalence of past-month anxiety was estimated among those ages ≥18, by survey year from 2008 to 2018. Time trends were tested using logistic regression.
RESULTS: Anxiety increased from 5.12% in 2008 to 6.68% in 2018 (p<.0001) among adult Americans. Stratification by age revealed the most notable increase from 7.97% to 14.66% among respondents 18-25 years old (p<.001), which was a more rapid increase than among 26-34 and 35-49 year olds (differential time trend p<.001). Anxiety did not significantly increase among those ages 50 and older. Anxiety increased more rapidly among those never married and with some college education, relative to their respective counterparts. Apart from age, marital status and education, anxiety increased consistently among sociodemographic groups.
CONCLUSIONS: Anxiety is increasing among adults under age 50 in the US, with more rapid increase among young adults. To prepare for a healthier adulthood and given direct and indirect (via 24/7 media) exposure to anxiety-provoking world events, prophylactic measures that can bolster healthy coping responses and/or treatment seeking seem warranted on a broad scale.
FIGURE 1: Percentage with self-reported anxiety, 2008-2018
FIGURE 2: Percentage with self-reported anxiety split by age group, 2008-2018
2.1.8 Sheridan, Grusing, Marshall, Lin, Hughes, Hendrickson, & Horowitz (2022). Changes in Suicidal Ingestion Among Preadolescent Children From 2000 to 2020. JAMA Pediatrics.
ABSTRACT: Pediatric emergency department visits for suicidality account for more than 1.1 million encounters per year, and this number has doubled in the past decade. Although attention has focused on suicidality among adolescents, another important group to consider is preadolescent children. A recent study of suicide risk screening in the emergency department for children aged 10 to 12 years found that more than half of these children with psychiatric complaints and almost 1 in 10 children with medical complaints received a positive screen. Approximately 1 in 5 preadolescent children in the study reported previous suicidal behavior. The objective of the present study was to examine age-specific patterns in suicidal ingestions using a nationally representative database. We hypothesized that calls for suicidal ingestions have increased.
[NOTE from Haidt: As in other studies, there is no sign of any trend up through 2012. And then, all of a sudden, the % of hospital admissions for poisoning that is coded as an attempted suicide shoots up, with the largest increase among pre-adolescents, ages, 10-12. The study notes that 68% of all admissions for poisoning are girls, although it does not allow us to know whether the increase in self-poisoning is exclusively among the girls.]
2.1.9 Youth Risk Behavior Survey Data Summary & Trends Report: 2013-2023. CDC.
NOTE: See Jon’s commentary on the 2021 YRBS survey. He argues that the new CDC report shows that Covid added little to teen mental health trends.
2.1.10 Pontes, Ayres, & Pontes (2020). Trends in Depressive Symptoms and Suicidality: Youth Risk Behavior Survey 2009–2017. Nursing Research.
ABSTRACT: BACKGROUND: Death by suicide is the second leading cause of death among adolescents globally. Healthy People 2020 set a goal to reduce by 10% the rate of suicide attempts that required treatment and designated the Youth Risk Behavior Survey (YRBS) to measure this objective.
OBJECTIVE: This study used cross-sectional YRBS data (2009–2017) to (a) investigate whether gender moderates the linear time trend for the rate of depressive symptoms, suicidal ideation, and suicide attempts among high school students within the United States; (b) estimate these same measures stratified by gender; and (c) investigate whether the Healthy People 2020 goal for reduction in suicide attempts was met in 2017 for males and females.
METHODS: Secondary analysis of nationally representative high school students using YRBS data 2009–2017 were analyzed using R and the R survey package. Based on Strengthening the Reporting of Observational Studies in Epidemiology guidelines (STROBE), additive interactions using rate difference were compared to multiplicative interactions using odds ratios.
RESULTS: Additive interactions were identified between female gender and linear trends for depressive symptoms, suicidal ideation, and suicide attempts requiring treatment. Females, but not males, had positive linear trends for depressive symptoms, suicidal ideation, and suicide attempts requiring treatment. Suicide attempts requiring treatment among females was 36% higher in 2017 than in 2009 but decreased 13% in males during the same period.
DISCUSSION: The Healthy People 2020 goal to reduce suicide attempts requiring treatment by 10% has not been met among females. The divergent trends by gender highlight the importance of surveillance measures by gender. Future research is needed to identify better suicide prevention strategies that address underlying factors and are gender specific.
2.1.11 Spiller… & Casavant (2020). Suicide attempts by self-poisoning in the United States among 10–25 year olds from 2000 to 2018: Substances used, temporal changes and demographics. Clinical Toxicology.
ABSTRACT: OBJECTIVE: To evaluate the substances used, outcomes, temporal and demographics associated with suicide attempts by self-poisoning in children and young adults aged 10–25 years old from 2000 to 2018.
METHODS: This is a retrospective review of suspected-suicide self-poisoning cases reported to the National Poison Data System (NPDS) from US Poison Centers from 2000 to 2018 for patients 10–25 years old. For comparison of annual rates, we obtained population data by year of age from the US Census Bureau. We evaluated changes in: monthly and annual incidence/rate per 100,000 population, substances used and outcome by patient age and demographics.
RESULTS: There were 1,677,435 cases of suicide attempt by self-poisoning among individuals 10–25 years old reported to US PCCs from 2000 to 2018. There were 410,940 self-poisoning cases (24.5%) with a serious medical outcome, and the proportion of exposures that resulted in a serious medical outcome increased with increasing age group. For the age groups of 10–12, 13–15 and 16–18 years of age, there was a significant increase after 2011, which was influenced primarily by females. The substance groups with the greatest number of serious medical outcomes were OTC analgesics, antidepressants, antihistamines and antipsychotics. ADHD medications were common in the younger age groups of 10–15 years, while the sedative/hypnotics occurred more commonly in the older age groups. The groups with the greatest increase in serious medical outcomes after 2011 were antidepressants, OTC analgesics, antihistamines and ADHD medications. Opiates were less commonly involved (7.4%) in cases with serious medical outcomes and decreased significantly in the 19–25 year-old age groups after 2012. States with a lower population per square mile had a greater number of reported cases with serious medical outcomes. There was a significant decrease in the number of cases in the age groups of 10–18 years during the traditional non-school months of June–August compared with September–May. This seasonal trend occurred among cases with all outcomes and among cases with serious medical outcomes. This decrease did not occur in the age group of 19–21 years, and there was an increase during summer months in the age group 22–25 years.
CONCLUSIONS: The substances used during self-poisoning varies by age group but appears to include substances available to that age group, with a significant increase after 2011, increased rates in more rural states, and a seasonal variation of increased rates during school months among adolescents but not among young adults. Two of the top substances, OTC analgesics and antihistamines, in all age groups, comprising more than a third of all substances used, are widely available over-the-counter with no restrictions regarding access. Of additional concern, ADHD medications had the highest risk of a serious medical outcome.
FIGURES:
2.1.12 Arakelyan, Freyleue, Avula, McLaren, O’Malley, & Leyenaar (2023). Pediatric Mental Health Hospitalizations at Acute Care Hospitals in the US, 2009-2019. JAMA.
ABSTRACT: IMPORTANCE: Approximately 1 in 6 youth in the US have a mental health condition, and suicide is a leading cause of death among this population. Recent national statistics describing acute care hospitalizations for mental health conditions are lacking.
OBJECTIVES: To describe national trends in pediatric mental health hospitalizations between 2009 and 2019, to compare utilization among mental health and non–mental health hospitalizations, and to characterize variation in utilization across hospitals.
DESIGN: Retrospective analysis of the 2009, 2012, 2016, and 2019 Kids’ Inpatient Database, a nationally representative database of US acute care hospital discharges. Analysis included 4 767 840 weighted hospitalizations among children 3 to 17 years of age.
EXPOSURES: Hospitalizations with primary mental health diagnoses were identified using the Child and Adolescent Mental Health Disorders Classification System, which classified mental health diagnoses into 30 mutually exclusive disorder types.
MAIN OUTCOMES: Measures included number and proportion of hospitalizations with a primary mental health diagnosis and with attempted suicide, suicidal ideation, or self-injury; number and proportion of hospital days and interfacility transfers attributable to mental health hospitalizations; mean lengths of stay (days) and transfer rates among mental health and non–mental health hospitalizations; and variation in these measures across hospitals.
RESULTS: Of 201 932 pediatric mental health hospitalizations in 2019, 123 342 (61.1% [95% CI, 60.3%-61.9%]) were in females, 100 038 (49.5% [95% CI, 48.3%-50.7%]) were in adolescents aged 15 to 17 years, and 103 456 (51.3% [95% CI, 48.6%-53.9%]) were covered by Medicaid. Between 2009 and 2019, the number of pediatric mental health hospitalizations increased by 25.8%, and these hospitalizations accounted for a significantly higher proportion of pediatric hospitalizations (11.5% [95% CI, 10.2%-12.8%] vs 19.8% [95% CI, 17.7%-21.9%]), hospital days (22.2% [95% CI, 19.1%-25.3%] vs 28.7% [95% CI, 24.4%-33.0%]), and interfacility transfers (36.9% [95% CI, 33.2%-40.5%] vs 49.3% [95% CI, 45.9%-52.7%]). The percentage of mental health hospitalizations with attempted suicide, suicidal ideation, or self-injury diagnoses increased significantly from 30.7% (95% CI, 28.6%-32.8%) in 2009 to 64.2% (95% CI, 62.3%-66.2%) in 2019. Length of stay and interfacility transfer rates varied significantly across hospitals. Across all years, mental health hospitalizations had significantly longer mean lengths of stay and higher transfer rates compared with non–mental health hospitalizations.
CONCLUSIONS: Between 2009 and 2019, the number and proportion of pediatric acute care hospitalizations due to mental health diagnoses increased significantly. The majority of mental health hospitalizations in 2019 included a diagnosis of attempted suicide, suicidal ideation, or self-injury, underscoring the increasing importance of this concern.
2.1.13 Spiller, Ackerman, Smith, Kistamgari, Funk, McDermott and Casavant (2020) Suicide Attempts By Self-Poisoning in the United States Among 10–25 Year Olds From 2000 to 2018: Substances Used, Temporal Changes and Demographics, Clinical Toxicology.
ABSTRACT: OBJECTIVE: To evaluate the substances used, outcomes, temporal and demographics associated with suicide attempts by self-poisoning in children and young adults aged 10–25 years old from 2000 to 2018.
METHODS: This is a retrospective review of suspected-suicide self-poisoning cases reported to the National Poison Data System (NPDS) from US Poison Centers from 2000 to 2018 for patients 10–25 years old. For comparison of annual rates, we obtained population data by year of age from the US Census Bureau. We evaluated changes in: monthly and annual incidence/rate per 100,000 population, substances used and outcome by patient age and demographics.
RESULTS: There were 1,677,435 cases of suicide attempt by self-poisoning among individuals 10–25 years old reported to US PCCs from 2000 to 2018. There were 410,940 self-poisoning cases (24.5%) with a serious medical outcome, and the proportion of exposures that resulted in a serious medical outcome increased with increasing age group. For the age groups of 10–12, 13–15 and 16–18 years of age, there was a significant increase after 2011, which was influenced primarily by females. The substance groups with the greatest number of serious medical outcomes were OTC analgesics, antidepressants, antihistamines and antipsychotics. (...). There was a significant decrease in the number of cases in the age groups of 10–18 years during the traditional non-school months of June–August compared with September–May. This seasonal trend occurred among cases with all outcomes and among cases with serious medical outcomes. This decrease did not occur in the age group of 19–21 years, and there was an increase during summer months in the age group 22–25 years.
2.1.14 Arakelyan M, Freyleue S, Avula D, McLaren JL, O'Malley AJ, Leyenaar JK (2023). Pediatric Mental Health Hospitalizations at Acute Care Hospitals in the US, 2009-2019. Journal of American Medical Association.
ABSTRACT: OBJECTIVE: To describe national trends in pediatric mental health hospitalizations between 2009 and 2019, to compare utilization among mental health and non–mental health hospitalizations, and to characterize variation in utilization across hospitals.
RESULTS: Adolescents aged 11 to 14 years accounted for an increasing proportion of mental health hospitalizations over time, increasing significantly from 33.5% (95% CI, 32.7%-34.2%) in 2009 to 40.4% (95% CI, 39.8%-41.0%) in 2019 (P value <.001). Mental health hospitalizations in children aged 10 years and younger and in children aged 15 years and older had commensurate decreases over this period (Table 1). The proportion of hospitalizations in females also increased significantly over time, from 51.8% (95% CI, 51.1%-52.6%) in 2009 to 61.1% (95% CI, 60.3%-61.9%) in 2019 (P value <.001). Differences across years in race and ethnicity, primary payer, community median income, co-occurring complex chronic conditions, and co-occurring disabilities were relatively small (Table 1). From 2009 to 2019, a significantly higher proportion of hospitalizations occurred at urban teaching and freestanding children’s hospitals with proportional declines in hospitalizations at rural and urban non–teaching centers. Other hospital characteristics were not notably different across years. […] From 2009 to 2019, both the absolute number and the proportion of mental health hospitalizations with an attempted suicide or self-injury diagnosis increased significantly, rising from 49 285 in 2009 (30.7% [95% CI, 28.6%-32.8%] of mental health hospitalizations) to 129 699 in 2019 (64.2% [95% CI, 62.3%-66.2%] of mental health hospitalizations; Figure 1). This represents a relative increase of 163.2% (95% CI, 115.5%-210.8%) in the number of mental health hospitalizations with an attempted suicide or self-injury diagnosis. Mental health hospitalizations with an attempted suicide or self-injury comprised a 4-fold greater proportion of all hospitalizations among children aged 3 to 17 years in 2019 vs 2009 (3.5% [95% CI, 3.2%-3.9%] to 12.7% [95% CI, 11.1%-13.9%])—an additional 9.2% percentage points (95% CI, 7.7%-10.6%; P < .001). The proportion of hospitalizations with an attempted suicide or self-injury diagnosis by primary mental health disorder group is shown in Table 2. In 2019, 69.9% (95% CI, 67.5%-72.3%) of hospitalizations with a primary diagnosis of depressive disorders had a secondary diagnosis for an attempted suicide or self-injury. The highest rate of co-occurrence was observed among the small group of hospitalizations for personality disorders in 2019 with 72.8% (95% CI, 66.3%-79.1%) having a secondary diagnosis of an attempted suicide or self-injury. Among those with diagnoses in 4 or more mental health disorder groups, 76.4% (95% CI, 75.1%-77.8%) had an attempted suicide or self-harm diagnosis in 2019.
CONCLUSION: Between 2009 and 2019, the number and proportion of pediatric acute care hospitalizations due to mental health diagnoses increased significantly. The majority of mental health hospitalizations in 2019 included a diagnosis of attempted suicide, suicidal ideation, or self-injury, underscoring the increasing importance of this concern.
2.1.15 Woolf, Wolf and Rivara (2023). The New Crisis of Increasing All-Cause Mortality in US Children and Adolescents. Journal of the American Medical Association.
EXCERPT: Suicides among individuals aged 10 to 19 years began to increase in 2007, and homicide rates in this age group began increasing in 2013. Between these nadirs and 2019, the eve of the COVID-19 pandemic, mortality rates for suicide increased by 69.5% and homicide rates increased by 32.7%.2 Likely contributors to both trends include increased access to firearms and a deepening mental health crisis among children and adolescents.4 Access to opioids (eg, fentanyl) also increased, and overdose death rates for individuals aged 10 to 19 years began increasing shortly before the COVID-19 pandemic.5
All youths did not face an equal risk of injury deaths. The increase in injury deaths that occurred in 2020 primarily involved males (Figure, A). Risk also varied by race and ethnicity. For example, non-Hispanic Black youths accounted for two-thirds (62.9%) of homicide victims aged 10 to 19 years; in 2021, the homicide rate among non-Hispanic Black youths aged 10 to 19 years was 6 times that of Hispanic youths and more than 20 times that of Asian/Pacific Islander non-Hispanic youths and White youths. Even larger racial and ethnic disparities existed across sexes: the homicide rate for non-Hispanic Black males aged 10 to 19 years was 61 times that of non-Hispanic White females
2.1.16 Slomski (2023). Slomski, A. (2023). Teen Girls Are Faring Worse Than Boys on Nearly All Mental Health Measures-Here's Why. Journal of the American Medical Association.
EXCERPT: According to the survey findings, which included data from more than 17 000 students in 152 public and private schools, 57% of teen girls reported feeling persistently sad or hopeless in the past year—a nearly 60% increase from 36% in 2011. In contrast, 29% of teen boys reported feeling this way in 2021, an increase from 21% in 2011. During this 10-year period, a nearly 60% increase also occurred in the percentage of teen girls who seriously considered suicide. In 2021, 30% of girls had these thoughts compared with 14% of boys, an increase from 19% and 13%, respectively, in 2011. And 13% of girls reported having had a suicide attempt in 2021 vs 7% of boys. “The disparity in mental health between boys and girls is not new, but the large and rapid rise in poor mental health among girls over the last decade compared with boys is particularly alarming,” said Kathleen Ethier, PhD, who directs the CDC Division of Adolescent and School Health.
2.1.17 Cushing, Liberman, Pham, Michelson, Festekjian, Chang, & Chaudhari (2023). Mental Health Revisits at US Pediatric Emergency Departments. JAMA Pediatrics.
ABSTRACT: OBJECTIVE: To describe trends in pediatric mental health ED visits and revisits and to determine factors associated with revisits.
METHODS: We conducted a multicenter, retrospective cohort study of mental health ED visits to US children’s hospitals between October 1, 2015, and February 29, 2020. Data were obtained from the Pediatric Health Information System (PHIS), an administrative database that contains billing, demographic, and resource utilization data from 49 tertiary care US children’s hospitals. We included data from 38 hospitals that had complete discharge and billing data and all variables of interest available.
RESULTS: There were 308 264 mental health ED visits from 217 865 unique patients, and 13.2% of patients had a mental health revisit within 6 months. Mental health visits increased by 8.0% annually (95% CI, 4.5%-11.4%), whereas all other ED visits increased by 1.5% annually (95% CI, 0.1%-2.9%). Factors associated with mental health ED revisits included psychiatric comorbidities, chemical restraint use, public insurance, higher area measures of child opportunity, and presence of an inpatient psychiatric unit at the presenting hospital. Patients with psychotic disorders (hazard ratio [HR], 1.42; 95% CI, 1.29-1.57), disruptive or impulse control disorders (HR, 1.36; 95% CI, 1.30-1.42), and neurodevelopmental disorders (HR, 1.22; 95% CI, 1.14-1.30) were more likely to revisit. Patients with substance use disorders (HR, 0.60; 95% CI, 0.55-0.66) were less likely to revisit.
CONCLUSION: Pediatric mental health ED visits and revisits are increasing rapidly. Factors associated with mental health revisits in this cohort study include presenting mental health diagnosis and markers of severe disease and health care access. Improved intervention services for patients with a behavioral health crisis are needed on a hospital and systems level to reduce pediatric mental health ED use and ensure access to appropriate follow-up care, with a particular focus on those patients most likely to revisit.
2.1.18 Bommersbach, McKean, Olfson and Rhee (2023). National Trends in Mental Health-Related Emergency Department Visits Among Youth, 2011-2020. Journal of the American Medical Association.
ABSTRACT: OBJECTIVE: To estimate annual trends in mental health–related ED visits among US children, adolescents, and young adults between 2011 and 2020.
METHODS: This study is based on data from the 2011-2020 NHAMCS, a cross-sectional survey of ED visits across the US conducted annually by the National Center for Health Statistics. The NHAMCS utilizes a 3-stage probability sampling design and sampling weights that allow for generalization of estimates to all nonfederal, short-stay, and general (medical, surgical, and pediatric) hospital EDs across the US. Visit information is abstracted from patient medical records by trained staff during a randomly assigned 4-week period to account for seasonal flux. For this study, we identified visits for children (6-11 years old), adolescents (12-17 years old), and young adults (18-24 years old) (unweighted = 49 515).
RESULTS: From 2011 to 2020, the weighted number of pediatric mental health–related visits increased from 4.8 million (7.7% of all pediatric ED visits) to 7.5 million (13.1% of all ED visits) with an average annual percent change of 8.0% (95% CI, 6.1%-10.1%; P < .001). Significant linearly increasing trends were seen among children, adolescents, and young adults, with the greatest increase among adolescents and across sex and race and ethnicity. While all types of mental health–related visits significantly increased, suicide-related visits demonstrated the greatest increase from 0.9% to 4.2% of all pediatric ED visits (average annual percent change, 23.1% [95% CI, 19.0%-27.5%]; P < .001). [...] As shown in Table 2, mental health–related ED visits per 1000 youth increased significantly across all age groups, but increased the most among 10- to 14-year olds (P value for interaction = .04). Aside from increases in suicide-related visits in all age groups, there were few significant age group trends by psychiatric diagnoses.
CONCLUSIONS: Our nationally representative analysis demonstrates a precipitous increase in mental health–related ED visits, especially suicide-related visits among children, adolescents, and young adults over the last 10 years. These results underscore a critical need to expand nonhospital alternatives to mental health care for young people.
2.1.19 Ormiston, Lawrence and Sulley (2024). Trends in Adolescent Suicide by Method in the US, 1999-2020. JAMA.
ABSTRACT: OBJECTIVE: To examine national trends in suicide mortality by method (firearm, poisoning, hanging and asphyxiation, and all other means) from 1999 to 2020 by demographic characteristics.
DESIGN: This serial cross-sectional study used national death certificate data of adolescent (aged 10-19 years) suicide decedents compiled by the National Center for Health Statistics from January 1, 1999, to December 31, 2020. Data analysis was performed from April 1, 2023, to July 9, 2023.
RESULTS: This study assessed data from 47 217 adolescent suicide decedents. From 1999 to 2020, suicide by firearm (AAPC, 1.0; 95% CI, 0.1-1.9), poisoning (AAPC, 2.7; 95% CI, 1.0-4.4), hanging and asphyxiation (AAPC, 2.4; 95% CI, 0.2-4.6), and other means (AAPC, 2.9; 95% CI, 1.2-4.6) increased. Rapidly increasing rates were observed among female adolescents for poisoning (AAPC, 4.5; 95% CI, 2.3-6.7) and hanging and asphyxiation (AAPC, 5.9; 95% CI, 5.0-6.8) suicides. From 2007 to 2020, firearm suicides sharply increased among female (annual percent change [APC], 7.8; 95% CI, 6.0-9.5) and male (APC, 5.3; 95% CI, 4.3-6.3) adolescents. Firearm suicide rates increased among Black adolescents from 2012 to 2020 (APC, 14.5; 95% CI, 9.7-19.5), Asian and Pacific Islander adolescents from 2008 to 2020 (APC, 12.0; 95% CI, 9.7-14.5), American Indian and Alaska Native adolescents from 2014 to 2020 (APC, 10.6; 95% CI, 2.6-19.3), and Hispanic or Latino adolescents from 2011 to 2020 (APC, 10.2; 95% CI, 6.3-13.8). During the study period, Black adolescents had the highest average increase in hanging and asphyxiation suicides (AAPC, 4.2; 95% CI, 3.2-5.2). From 2011 to 2020, poisoning suicide deaths increased (APC, 12.6; 95% CI, 8.5-16.7) among female adolescents.
CONCLUSION: Suicide rates increased across all methods from 1999 to 2020. Differences were noted by sex, age, and race and ethnicity. Increasing suicide rates among racial and ethnic minoritized youth are especially concerning, and effective prevention strategies are urgently needed.
2.1.20 Wang, Tang … & Wang (2024). Insufficient Sleep is Associated With Increasing Trends in Adolescent Suicidal Behaviors. Journal of Adolescent Health.
ABSTRACT: PURPOSE: Youth suicide has been increasing and became a public health concern worldwide. Identifying insufficient sleep as the potential risk factor is critical to reducing suicide risk and increasing trends. This study aimed to determine whether insufficient sleep is associated with increasing trends in suicidal behaviors and disparities by sex, age, and race/ethnicity among school adolescents.
METHODS: The present study used biennial data from the US nationally representative Youth Risk Behavior Survey from 2007 to 2019. Joinpoint regression models were used to estimate biennial percent changes (BPCs) and average BPCs (ABPCs) of suicidal behaviors by sleep duration. Logistic regression models were used to examine the association between insufficient sleep and suicidal behaviors.
RESULTS: Of 73,356 adolescent students included (mean [standard deviation] age, 16.11 [1.23] years), 50.03% were female. Suicidal ideation and suicide plan among insufficient sleep group increased from 2007 to 2019 (BPC = 2.88% [95% confidence interval {CI}: 1.65%, 4.13%]; BPC = 3.42% [95% CI: 2.09%, 4.77%]), but were nonsignificant among sufficient sleep group. Trends in suicidal ideation (ABPC = 3.03% [95% CI: 1.35%, 4.73%]) and suicide plan (ABPC = 4.03% [95% CI: 2.47%, 5.62%]) among female adolescents with insufficient sleep increased, but nonsignificant among male adolescents with insufficient sleep. Suicidal ideation (ABPC = 1.73% [95% CI: 0.51%, 2.97%]) and suicide plan (ABPC = 2.31% [95% CI: 0.70%, 3.95%]) increased among younger adolescents only with insufficient sleep, whereas suicide trends by sleep duration were similar among older adolescents. Suicide plan among insufficient sleep group increased across the four racial groups, with BPC highest for the White (BPC = 3.48% [95% CI: 1.31%, 5.69%]), and lowest for the Hispanic/Latino (BPC = 1.18% [95% CI: 0.15%, 2.23%]), but were nonsignificant among sufficient sleep group except for the White (BPC = 2.83% [95% CI: 0.62%, 5.09%]).
DISCUSSION: Insufficient sleep was disproportionately associated with increasing trends in suicidal behaviors among female, younger, and non-White adolescent students. Ensuring sufficient sleep can potentially reduce suicide among school adolescents.
[What are we missing?]
2.2 USA: COUNTERPOINTS AND CRITICISMS
2.2.1 Corredor-Waldron & Currie (2023). To What Extent are Trends in Teen Mental Health Driven by Changes in Reporting? The Example of Suicide-Related Hospital Visits. National Bureau of Economic Research.
ABSTRACT: Rising reports of suicidal behaviors in children and adolescents have led to the recognition of a youth mental health crisis. However, reported rates can be influenced by access to screening and changes in reporting conventions, as well as by changes in social stigma. Using data on all hospital visits in New Jersey from 2008-2019, we investigate two inflection points in adolescent suicide-related visits and show that a rise in 2012 followed changes in screening recommendations, while a sharp rise in 2016-2017 followed changes in the coding of suicidal ideation. Rates of other suicidal behaviors including self-harm, attempted suicides, and completed suicides were essentially flat over this period. These results suggest that underlying suicide-related behaviors among children, while alarmingly high, may not have risen as sharply as reported rates suggest. Hence, researchers should approach reported trends cautiously.
[What are we missing?]
2.3 UK: STUDIES AND DATASETS SHOWING A RISE IN RECENT YEARS IN TEEN SELF HARM AND SUICIDE ATTEMPTS
2.3.1 Morgan, Webb, Carr, et al. (2017). Incidence, clinical management, and mortality risk following self harm among children and adolescents: cohort study in primary care. BMJ.
ABSTRACT: Objectives: To examine temporal trends in sex and age specific incidence of self harm in children and adolescents, clinical management patterns, and risk of cause specific mortality following an index self harm episode at a young age….
SETTING: UK Clinical Practice Research Datalink—electronic health records from 674 general practices, with practice level deprivation measured ecologically using the index of multiple deprivation. Patients from eligible English practices were linked to hospital episode statistics (HES) and Office for National Statistics (ONS) mortality records.
Participants: For the descriptive analytical phases we examined data pertaining to 16 912 patients aged 10-19 who harmed themselves during 2001-14. For analysis of cause specific mortality following self harm, 8638 patients eligible for HES and ONS linkage were matched by age, sex, and general practice with up to 20 unaffected children and adolescents (n=170 274).
MAIN OUTCOME MEASURES: In the first phase, temporal trends in sex and age specific annual incidence were examined. In the second phase, clinical management was assessed according to the likelihood of referral to mental health services and psychotropic drug prescribing. In the third phase, relative risks of all cause mortality, unnatural death (including suicide and accidental death), and fatal acute alcohol or drug poisoning were estimated as hazard ratios derived from stratified Cox proportional hazards models for the self harm cohort versus the matched unaffected comparison cohort.
RESULTS: The annual incidence of self harm was observed to increase in girls (37.4 per 10 000) compared with boys (12.3 per 10 000), and a sharp 68% increase occurred among girls aged 13-16, from 45.9 per 10 000 in 2011 to 77.0 per 10 000 in 2014. Referrals within 12 months of the index self harm episode were 23% less likely for young patients registered at the most socially deprived practices, even though incidences were considerably higher in these localities. Children and adolescents who harmed themselves were approximately nine times more likely to die unnaturally during follow-up, with especially noticeable increases in risks of suicide (deprivation adjusted hazard ratio 17.5, 95% confidence interval 7.6 to 40.5) and fatal acute alcohol or drug poisoning (34.3, 10.2 to 115.7).
CONCLUSIONS: Gaining a better understanding of the mechanisms responsible for the recent apparent increase in the incidence of self harm among early-mid teenage girls, and coordinated initiatives to tackle health inequalities in the provision of services to distressed children and adolescents, represent urgent priorities for multiple public agencies.
FIGURE:
[Note: As in the American data in 2.1.1 (Mercado et al. 2017), we see a very sudden and very large increase, for teen girls only. The age brackets here are different than those used by Mercado.].
2.3.2 Cybulski, Ashcroft, Carr, Garg, Chew-Graham, Kapur, & Webb (2021). Temporal trends in annual incidence rates for psychiatric disorders and self-harm among children and adolescents in the UK, 2003–2018. BMC Psychiatry.
ABSTRACT: BACKGROUND: There has been growing concern in the UK over recent years that a perceived mental health crisis is affecting children and adolescents, although published epidemiological evidence is limited.
METHODS: Two population-based UK primary care cohorts were delineated in the Aurum and GOLD datasets of the Clinical Practice Research Datalink (CPRD). We included data from 9,133,246 individuals aged 1–20 who contributed 117,682,651 person-years of observation time. Sex- and age-stratified annual incidence rates were estimated for attention-deficit/hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) (age groups: 1–5, 6–9, 10–12, 13–16, 17–19), depression, anxiety disorders (6–9, 10–12, 13–16, 17–19), eating disorders and self-harm (10–12, 13–16, 17–19) during 2003–2018. We fitted negative binomial regressions to estimate incidence rate ratios (IRRs) to examine change in incidence between the first (2003) and final year (2018) year of observation and to examine sex-specific incidence.
RESULTS: The results indicated that the overall incidence has increased substantially in both boys and girls in between 2003 and 2018 for anxiety disorders (IRR 3.51 95% CI 3.18–3.89), depression (2.37; 2.03–2.77), ASD (2.36; 1.72–3.26), ADHD (2.3; 1.73–3.25), and self-harm (2.25; 1.82–2.79). The incidence for eating disorders also increased (IRR 1.3 95% CI 1.06–1.61), but less sharply. The incidence of anxiety disorders, depression, self-harm and eating disorders was in absolute terms higher in girls, whereas the opposite was true for the incidence of ADHD and ASD, which were higher among boys. The largest relative increases in incidence were observed for neurodevelopmental disorders, particularly among girls diagnosed with ADHD or ASD. However, in absolute terms, the incidence was much higher for depression and anxiety disorders.
CONCLUSION: The number of young people seeking help for psychological distress appears to have increased in recent years. Changes to diagnostic criteria, reduced stigma, and increased awareness may partly explain our results, but we cannot rule out true increases in incidence occurring in the population. Whatever the explanation, the marked rise in demand for healthcare services means that it may be more challenging for affected young people to promptly access the care and support that they need.
FIGURES:
[Note that all classifications are based on the researchers’ analysis of medical records for anonymized cases within the cohort. The records were created by doctors. None of the data in this study is based on self-reports or parent reports]
[Here is the self-harm graph, replotted by Haidt]
2.3.3 Trafford… & Mok (2023). Temporal trends in eating disorder and self-harm incidence rates among adolescents and young adults in the UK in the 2 years since onset of the COVID-19 pandemic: A population-based study. The Lancet Child & Adolescent Health.
ABSTRACT: BACKGROUND: Self-harm and eating disorders share multiple risk factors, with onset typically during adolescence or early adulthood. We aimed to examine the incidence rates of these psychopathologies among young people in the UK in the 2 years following onset of the COVID-19 pandemic.
METHODS: We conducted a population-based study using the primary care electronic health records of patients aged 10–24 years in the UK Clinical Practice Research Datalink (CPRD). The observation period was from Jan 1, 2010, to March 31, 2022. We calculated the monthly incidence rates of eating disorders and self-harm according to the first record of each outcome. On the basis of antecedent trends between January, 2010, and February, 2020, negative binomial regression models were fitted to predict monthly incidence rates after the pandemic began in March, 2020. Percentage differences between observed and expected incidence were calculated to indicate changes since the onset of the pandemic, with stratification by sex, age, and deprivation quintile.
FINDINGS: The primary care health records of 9 184 712 patients aged 10–24 years (4 836 226 [52·7%] female patients and 4 348 486 [47·3%] male patients; n=1881 general practices) were included for analysis. The incidence rates of eating disorders and self-harm among girls were higher than expected between March 1, 2020, and March 31, 2022. The observed incidence of eating disorders was 42·4% (95% CI 25·7–61·3) higher than expected for girls aged 13–16 years, and 32·0% (13·3–53·8) higher than expected for girls aged 17–19 years, whereas other age groups showed little difference between observed and expected incidence. Similarly, the increase in self-harm incidence was driven by girls aged 13–16 years, for whom the observed incidence was 38·4% (20·7–58·5) higher than expected. By contrast, among boys in all age groups, the incidence rates of eating disorders and self-harm were lower than, or close to, the expected rates. Among boys, the observed incidence of eating disorders was 22·8% (9·2–34·4) lower than expected, and the observed incidence of self-harm was 11·5% (3·6–18·7) lower than expected. The estimated increases in eating disorder and self-harm incidence among girls aged 13–16 years were largely attributable to increases within less deprived communities.
INTERPRETATION: Although causes are uncertain, increased incidence of eating disorder diagnoses and self-harm among teenage girls in the UK during the first 2 years of the COVID-19 pandemic highlight an urgent need for intervention. Early identification of mental health difficulties by primary care clinicians is necessary. Timely access to treatments and sufficient support from general practitioners and mental health services needs to be available to manage presenting problems and to prevent exacerbations of conditions.
FIGURE:
[What are we missing?]
2.4 UK: COUNTERPOINTS AND CRITICISMS
2.4.1 [None found so far. Are there any studies disputing the rise in self-harm?]
[What are we missing?]
2.5 WHAT IS HAPPENING IN OTHER ENGLISH-SPEAKING COUNTRIES?
2.5.1 Canada
See our separate Google doc that examines a variety of trends: The Coddling of the Canadian Mind. The short answer is that the same pattern is occurring among Canadian teens, with girls up more than boys. However, unlike the USA, it seems to be a steady rise, rather than a “hockey stick.” Here are two figures from that document:
SOURCE:Gardner… & Lima (2019). Changing rates of self-harm and mental disorders by sex in youths presenting to Ontario emergency departments: Repeated cross-sectional study. Canadian Journal of Psychiatry.
NOTES: Similar to the U.S. self-harm study (Mercado et al., 2017), increases for girls begin around 2011. For mental health visits overall, the pattern is similar. Increases for both outcomes are small for boys, and very large for girls--more than 100%.
Regraphed by Zach Rausch (see spreadsheet)
Source: Statistics Canada (2016) The health of girls and women in Canada
FIGURE: [plotted by Cameron How and Jon Haidt, from data in the report]
More recent data here: Liu, Pollock, Contreras, Xu, & Thompson (2024). Hospitalizations and emergency department visits for self-harm in Canada during the first two years of the COVID-19 pandemic: A time series analysis. Journal of Affective Disorders.
ABSTRACT: BACKGROUND: Rates of hospitalizations and emergency department (ED) visits due to self-harm are important indicators for understanding the impact of the COVID-19 pandemic on mental health. The objective of this study was to assess changes in self-harm hospitalizations and ED visits in Canada during the first two years of the pandemic.
METHODS: Rates of self-harm hospitalizations and ED visits during the pandemic were predicted based on regression analyses that modeled trends over a 5-year pre-pandemic period from fiscal year 2015 to 2019. The ratios of observed and model predicted (expected) rates in 2020 and 2021 were estimated separately to assess changes during the pandemic.
RESULTS: Overall, rates of self-harm hospitalizations and ED visits were lower than expected during the pandemic, especially in 2020. In 2021, rates for females returned to near-expected levels; but they remained lower than expected for males. Females aged 10–14 years had higher than expected rates. The rate ratio of observed rate over expected rate was 1.2 in 2020 but further increased to 1.8 in 2021 for both hospitalizations and ED visits. Higher than expected rates were also observed among females aged 15–19 years in 2021 only.
LIMITATIONS: Suicide attempts and non-suicidal self-harm cases could not be distinguished.
CONCLUSIONS: We observed lower than or close to expected rates of self-harm hospitalizations and ED visits during the pandemic for most population groups. The increased rates for young females highlights the importance of continued surveillance post-pandemic and targeted mental health services and suicide prevention programs.
2.5.2 Australia
See our separate Google doc that examines a variety of trends: The Coddling of the Australian Mind. Here is one figure from that doc
SOURCE: 2019–20 National Hospital Morbidity Database—Intentional self-harm hospitalisations. Australian Institute of Health and Welfare. Datasets can be downloaded here. Graphed by Zach Rausch — you can see his spreadsheet here.
Also see: Sara, Wu, Uesi, Jong, Perkes, Knight, O’Leary, Trudgett, & Bowden (2023). Growth in emergency department self-harm or suicidal ideation presentations in young people: Comparing trends before and since the COVID-19 first wave in New South Wales, Australia. Australian & New Zealand Journal of Psychiatry.
ABSTRACT: Self-harm presentations in children and young people have increased internationally over the last decade. The COVID-19 pandemic has the potential to worsen these trends.
OBJECTIVE: To describe trends in emergency department self-harm or suicidal ideation presentations for children and young people in New South Wales before and since the COVID-19 pandemic.
METHODS: We studied presentations for self-harm or suicidal ideation by 10- to 24-year-olds to New South Wales emergency departments, using interrupted time series analysis to compare annualised growth before COVID (2015 to February 2020) and since COVID (March 2020 to June 2021). Subgroup analyses compared age group, gender, triage category, rurality and disadvantage. Time series decomposition via generalised additive models identified long-term, seasonal and short-term trends.
RESULTS: Self-harm or suicidal ideation presentations by young people in New South Wales increased by 8.4% per annum pre-COVID. Growth accelerated since COVID, to 19.2% per annum, primarily due to increased presentations by females aged 13–17 years (47.1% per annum since COVID, from 290 per 10,000 in 2019 to 466 per 10,000 in 2021). Presentations in males aged 10–24 years did not increase since COVID (105.4 per 10,000 in 2019, 109.8 per 10,000 in 2021) despite growing 9.9% per annum before COVID. Presentation rates accelerated significantly in socio-economically advantaged areas. Presentations in children and adolescents were strongly linked to school semesters.
CONCLUSION: Emergency department self-harm or suicidal ideation presentations by New South Wales young people grew steadily before COVID. Understanding the sustained increase remains a priority. Growth has increased since COVID particularly for adolescent females, but not among adolescent males. Surprisingly, the largest post-COVID increases in annual growth occurred in socio-economically advantaged and urban regions. The COVID-19 pandemic appears to have added new challenges, particularly in females in the developmentally critical early adolescent and teenage years.
FIGURE
Figure. Growth in self-harm or suicidal ideation presentation rates before COVID and since COVID-19 first wave in NSW. Interrupted time series analysis showing weekly emergency department self-harm or suicidal ideation presentation rates (per 10,000) and linear trends before and since COVID-19 in people aged 10–24 years in NSW, January 2015 to June 2021. Spline curve included for visualisation.
2.5.3 New Zealand
See our separate Google doc that examines a variety of trends: The Coddling of the Kiwi Mind. Here is one figure from that doc
SOURCE: Hospital event data and stats, New Zealand. Public hospital discharges for intentional self-harm. Note that these are raw numbers, not rates. Graphed by Zach Rausch — you can see his spreadsheet here.
* * * * * * * * * * * * * * * * * * * * * * * * *
SECTION 3: SUICIDE
3.1 USA: STUDIES AND DATASETS SHOWING A RISE IN RECENT YEARS IN TEEN SUICIDE
3.1.1 Centers for Disease Control, Fatal Injury Data.
NOTE: There is a slight discontinuity in CDC data at 1999; slightly new methods for counting were adopted.
SOURCE: CDC Fatal Injury Reports (1981-2022)
NOTE: There is a slight discontinuity in CDC data at 1999; slightly new methods for counting were adopted.
And unlike most of the developed world, where suicide rates have been falling, rates in the USA have been rising in recent years, for nearly all age groups, and for both sexes:
Source: CDC data, grouped into 5 year age bins, originally graphed by Chris Vaccaro (updated by Zach Rausch, see his spreadsheet). Rates are per 100,000 in the population of that age/gender group.
When we look at the percentage change for each group, however, we see very large age differences and sex differences. I took as the baseline the average rate from 2000 to 2009, and then calculated how much above that baseline the rate was in 2020, for each of the age groups graphed in the charts above.
NOTE THAT THE SUICIDE RATE FOR YOUNG TEEN GIRLS HAS ALMOST TRIPLED SINCE THE 2000S, AND THE RATE FOR THE OTHER TWO GEN Z FEMALE GROUPS IS WAY UP TOO. THE RATE FOR YOUNG TEEN BOYS HAS INCREASED BY 68%, AND THE RATE FOR THE NEXT 3 AGE GROUPS IS ALSO UP A LOT. THE INCREASES FOR THOSE BORN BEFORE 1980 (GEN X AND OLDER) ARE GENERALLY MUCH SMALLER.
3.1.2 Ruch, Sheftall, Schlagbaum et al. (2019). Trends in suicide among youth aged 10 to 19 years in the United States, 1975 to 2016. JAMA.
ABSTRACT: DESIGN, SETTINGS, AND PARTICIPANTS: Cross-sectional study using period trend analysis of US suicide decedents aged 10 to 19 years from January 1, 1975, to December 31, 2016….
RESULTS: From 1975 to 2016, we identified 85 051 youth suicide deaths in the United States (68 085 male [80.1%] and 16 966 female [19.9%]) with a male to female IRR [Incidence Rate Ratios] of 3.82 (95% CI, 3.35-4.35). Following a downward trend until 2007, suicide rates for female youth showed the largest significant percentage increase compared with male youth (12.7% vs 7.1% for individuals aged 10-14 years; 7.9% vs 3.5% for individuals aged 15-19 years). The male to female IRR decreased significantly across the study period for youth aged 10 to 14 years (3.14 [95% CI, 2.74-3.61] to 1.80 [95% CI, 1.53-2.12]) and 15 to 19 years (4.15 [95% CI, 3.79-4.54] to 3.31 [95% CI, 2.96-3.69]). Significant declining trends in the male to female IRR were found in non-Hispanic white youth aged 10 to 14 years (3.27 [95% CI, 2.68-4.00] to 2.04 [95% CI, 1.45-2.89]) and non-Hispanic youth of other races aged 15 to 19 years (4.02 [95% CI, 3.29-4.92] to 2.35 [95% CI, 2.00-2.76]). The male to female IRR for firearms increased significantly for youth aged 15 to 19 years (χ2 = 7.74; P = .02 for sex × period interaction). The male to female IRR of suicide by hanging or suffocation decreased significantly for both age groups (10-14 years: χ2 = 88.83; P < .001 for sex × period interaction and 15-19 years: χ2 = 82.15; P < .001 for sex × period interaction). No significant change was found in the male to female IRR of suicide by poisoning across the study period.
3.1.3 Miron, Yu, Wilf-Miron, & Kohane (2019). Suicide rates among adolescents and young adults in the United States, 2000-2017. JAMA.
NOTE: This letter goes beyond 3.1.2 by bringing in 2017 suicide data and comparing late teens to those in early 20s, and also by trying to identify change points -- times at which trends change. The first pair of graphs shows data for 15-19 year olds. Note that Gen Z begins to enter that age range in 2011 or so, and fills it by 2016:
Note also that the authors identify 2011 as the major change point for girls. For boys, the sharp increase comes later, in 2015.
For 20-24 year olds the patterns are different. (Gen Z begins to enter the dataset in 2016 or so)
Here is an article in the Wall St. Journal showing the rise by age group, but not by gender; you can see that the 2017 data shows an acceleration, not a leveling-off:
3.1.4 Ross, Woodfin, Rege, & Holstege (2022). Pediatric suicides reported to U.S. poison centers. Clinical Toxicology.
ABSTRACT: INTRODUCTION: As the pediatric mental health crisis worsens, the rate of adolescent suicide-related cases is increasing, including adolescent cases of self-poisoning.
METHODS: Data from the National Poison Data System was analyzed for trends in rates and frequencies of all pediatric suspected suicides between 2015 and 2020.
RESULTS: There were 514,350 pediatric suspected suicides analyzed, with the largest increase in rate of suspected suicides occurring in children ages 10 to 12 years (109.3%, p = 0.002). Rates also increased significantly in children ages 13 to 15 years (30.3%, p < 0.001) and 16 to 19 years (18.1%, p < 0.05). The most commonly utilized substances were ibuprofen and acetaminophen, with the largest increase in rate of exposures seen for acetaminophen.
DISCUSSION: This data demonstrates concerning rises in cases of self-poisoning, suggesting that the pediatric mental health crisis is worsening and extending into younger populations. Pediatric populations have easier access to over-the-counter medications, potentially explaining the likelihood of utilization of these medications in pediatric suspected suicides.
CONCLUSIONS: Initiation of appropriate mental health screenings and interventions should be considered in these young age groups in order to prevent further rises in self-poisoning cases and associated morbidity and mortality.
FIGURE 1:
3.1.5 Marcotte & Hansen (2023). The Re-Emerging Suicide Crisis in the U.S.: Patterns, Causes and Solutions (Working Paper No. 31242). National Bureau of Economic Research.
ABSTRACT: The suicide rate in the United States has risen nearly 40 percent since 2000. This increase is puzzling because suicide rates had been falling for decades at the end of the 20th Century. In this paper, we review important facts about the changing rate of suicide. General trends miss the story of important differences across groups – suicide rates rose substantially among middle aged persons between 2005 and 2015 but have fallen since. Among young people, suicide rates began a rapid rise after 2010 that has not abated. We review empirical evidence to assess potential causes for recent changes in suicide rates. The economic hardship caused by the Great Recession played an important role in rising suicide among prime-aged Americans. We illustrate that the increase in the prevalence of depression among young people during the 2010s was so large it could explain nearly all the increase in suicide mortality among those under 25. Bullying victimization of LGBTQ youth could also account for part of the rise in suicide. The evidence that access to firearms or opioids are major drivers of recent suicide trends is less clear. We end by summarizing evidence on the most promising policies to reduce suicide mortality.
[What are we missing?]
3.2 USA: COUNTERPOINTS AND CRITICISMS
3.2.1: Critique from Chris Ferguson: [which began as a discussion on the Social Media literature review; Haidt pasted all text into this newer lit review, where we continued the dialogue]
FERGUSON: Just as one thing, I wonder if this is a false frame. My understanding of the federal data is that self-harm/suicide has been rising for all age groups, other than elderly adults (and tiny children of course). I believe the worst increases are actually for middle-aged adults. So it's not a "teen thing", and I wonder if that might argue for other explanations than technology if rates of increases are higher among lower tech adopters?
HAIDT: Thanks Chris, rates are rising 10-40% or so for all age groups, but rising much faster for teen girls, and extraordinarily fast for pre-teens, who have lowest total rates, but by far the largest percent increases. I'll create a google doc for this question too, later this week. [meaning: this google doc] It is a crucial empirical question that is debated.
FERGUSON: No problem. I wonder if it's one of those things that the data can be parsed different ways and there are multiple competing reports. I know I read one recently that showed both the highest increase and highest overall among middle-aged adults, but of course I'll have to dig around to find it. If you look at crude rates, highest crude rates remain highest for middle-aged adults. Thus, if suicide rates are changing for any cause it wouldn't be unexpected to see higher increases for lower frequency groups (i.e. regression to the mean) and it probably isn't possible to ascribe a clear cause due to any social cause (whether, say, technology, the 2008 economic crisis, etc.): https://webappa.cdc.gov/sasweb/ncipc/mortrate.html
HAIDT: Hi Chris, yes, it is percent increase i'm focused on; i think that's the proper stat. I'm creating another google doc for these mental health stats [i.e., this document] All groups are up, as you say, but a 151% increase for pre-teen girls is.... stunning, out of the ballpark. if you have any comments/critiques on that file please do add them.
FERGUSON: I think you're making a mistake and this isn't the proper stat. This is a long-standing problem in CJ research that "percent increase" is misleading given low base rate behaviors and regression to the mean. Take, for instance, a town that in 2018 experiences 1 murder. In 2019 they experience 2 murders and news media report "This year our town experienced a 100% increase in homicides." Technically true, but also deeply misleading, as the "increase" is not put into proper context. That's why, particularly with low base-rate behaviors, we have to be careful using percent increase stats. To make an argument for this you have to ignore a.) this didn't happen for boys to the same degree, b.) overall rates for middle-aged adults are much higher, c.) the raw increase for middle-aged adults is much higher. To be direct (and I'm just being honest) I'd fail a senior student research project for trying to make the inferential leaps I think you're trying to make with these numbers. It's not even really correlational data, very cherry-picked. Again, I'm not trying to be overly critical. I do my best to try to warn people about where they're going to get slammed with criticism (hopefully, you'll be like, literally, the first to listen, haha).
HAIDT: You are certainly right that with low baserates, big percentage increases are easier to obtain, and you are right that such increases would be misleading IF the raw numbers were low (like 4 going to 6, rather than 100 going to 150), and the increase was within the range of normal variation, so that it could reflect nothing more than regression to the mean. But in this case, neither of those conditions apply. I went to the CDC WISQARS website and requested the raw numbers and rates, just for 10-14 year old girls, for deaths deemed suicides since 1999 (when the main dataset begins). Here is the output. You can see that we are not dealing with small numbers, and the two big jumps, in 2009 and 2013 are not blips, they are not regression to the mean, they are each the start of sustained and large increases.
The rate for the last 5 years is nearly triple the rate for the first five years. I think that is an accurate way to say it. When we combine that with the many other findings in this lit review, it tells us that something is changing in the lives of pre-teen American girls, and it tells us to pay attention. I do not think it would be better to say that the rate is essentially unchanged, since it has merely increased by about one out of 100,000 girls per year (i.e., from 0.51 to 1.66), an increase that is smaller in absolute terms than is happening to older girls and women.
Here is the corresponding table for boys aged 10-14. Here the regression to the mean argument is clearly more relevant. The rise around 2012 put them up well above the mean, for a sustained time, but there was one year (2000) that had had a similar rate. The jump up in 2017, to 3.28 could be a one-year blip, if the number for 2018 comes in much lower, but not if it stays high.
FERGUSON: But here's why the proportional increase is the wrong statistic on which to base your claim. I looked at the same data for 45-49 year old males for the same time period (the age range was picked randomly to be somewhere in middle age.) You are saying that an increase in raw deaths among preteen girls of about 100 is "stunning"...yeah the raw increase in men in the age category from 1999 to 2017 is 1000. If you want to talk in terms of proportions...that's actually a 900% (!!!) increase in raw deaths over those seen among teen girls. And the population of the two groups is roughly the same (about 10 million, give or take.) THAT is why you are misleading people with your approach. You are implying that preteens and teens are more influenced by this suicide trend but that is clearly not the case and this is why you cannot use proportion increase as your main stat.
HAIDT: If our concern is to save the maximum number of lives then you are right, we should be focusing on the age ranges that have the highest suicide rates. But the point of this lit review, and the companion one on social media, is to figure out what is happening to Gen Z, and whether Gen Z is being adversely affected by social media. If the number of deaths jumps by 1000 among a slice of middle age men, and it’s a 20% increase from 10 years before, it’s alarming, but the 20% increase doesn’t tell us that something fundamental has changed for middle aged men as clearly as the finding that for pre-teen girls, the extra 100 deaths is a 150% increase. That 100 deaths may not seem like a lot, compared to the 1000 for middle aged men, but when you link it to the consistent finding throughout this lit review that girls are up a lot, and young girls are up the most, on most measures, then I think this 100-death increase, sustained for multiple years, has more informational value. I dwell on this finding because skeptics like Friedman (1.2.1) say that the rise is just found in self-report data. It is not. The rise in self-reports of distress from teen girls is mirrored in the rising number of suicides, and in the rising (and much larger) number of hospital admissions for self-harm (see study 2.1.1, Mercado). These are not based on self report; they tell us that there is a big change in the mental health of teen girls, bigger than the change in mental health for middle aged men. I think it is important to look at the percentage change in deaths, as well as at the raw numbers of extra deaths, for each age/sex category.
FERGUSON: [raising an additional point, in response to Haidt’s statement that the “main dataset” of CDC suicide data begins in 1999, because the CDC posts older datasets too]: Actually, it begins in 1981. If you start back then you'll see something of a sine wave for this age group and gender. Granted, the numbers are still higher, but you can see some of that regression to the mean effect bounce back and forth. In a population of 8-10 million people, these are still fairly small numbers. Not saying that there isn't a trend...I agree with you that there is. But when that trend takes the form of a sine wave, trying to interpret it in regards to specific historical events becomes a fraught process.
HAIDT: Yes, there is what looks like a sine wave for older teen boys. Here is the graph from the CDC website:
Suicide Rates for Teens Aged 15–19 Years, by Sex — United States, 1975–2015
Source: CDC: https://www.cdc.gov/mmwr/volumes/66/wr/mm6630a6.htm
So you might say that for boys, the recent rise could just be regression to the mean; it is certainly NOT the highest rate its ever been. That is a valid point. But I believe that the huge wave of male teen suicides in the 1980s and 1990s was part and parcel of the huge male wave of violence in those years. We don’t know what caused that wave (although as I said earlier, i think that mass lead poisoning from the 1950s through the banning of leaded gas around 1980 is a big contributor to the rise in violence of all sorts in the 1960s and 1970s, and the sudden drop in the 1990s.) Look how the violent crime and suicide rates used to move in tandem, until around 2007:
Figure 3.12, from Twenge (2017). Suicide and homicide rates for US teens (ages 15-19, both sexes combined). “In 2011, for the first time in twenty-four years, the teen suicide rate was higher than the teen homicide rate. The gap grew larger from 2011 to 2014, with the suicide rate 32% higher than the homicide rate by 2014—the largest gap since records have been kept.”
This graph shows data for both sexes combined, but both lines reflect mostly male behavior. Since around 2007 or so, and especially since 2013, the boys’ suicide rate is rising even as their violent crime rate falls, or stays flat. So I will grant your point that we should look at the suicide rates going back as far as we can, and we do indeed see an earlier period when the rates were higher than today. Today’s boys’ suicide rate is not the highest in modern times, although it is very close (16 to 18 per 100k). But the main story emerging in this document is that in both the US and UK, something has changed for girls, more sharply than for boys. For teen girls in the USA, the recent rise takes the rate to the highest it has been, for as long as CDC has been collecting data.
UPDATE TO THIS DEBATE, IN FEB. 2023:
UPDATED GRAPHS IN 2024 WITH DATA THROUGH 2022: USA Suicide Rates per 100,000 (Ages 15-19). CDC Fatal Injury Reports.
UPDATED GRAPHS IN 2023 WITH DATA THROUGH 2021: USA Suicide Rates per 100,000 by Sex (Ages 10-14). CDC Fatal Injury Reports.
3.2.2 Males (2018). The Truth About Teen Suicide: Smartphones and social media aren’t driving the increase in teen suicide. The real story is far more complicated. Washington Monthly.
ABSTRACT: This is a magazine article that argues that claims about teen suicide rising are another moral panic. Main point is that rates for all groups are rising, and the rise is higher in rural areas, where smartphone usage rates are lower:
COMMENT: [Jon or someone will need to check this out, particularly whether the rate within race/sex is rising mostly in small towns. Whites have a higher suicide rate than African Americans. The fact that the rate is rising quickly since 2013 in suburbs suggests it’s not just a rural issue. We need to get 2017 data into this graph. Also: Is it really true that smartphone use is lower in rural areas?]
UPDATE TO HOMICIDE x SUICIDE GRAPHS (May 15th, 2024):
Figure. Male homicide by suicide Rates (CDC). See spreadsheet.
[What are we missing?]
3.3 UK: STUDIES AND DATASETS SHOWING A RISE IN RECENT YEARS IN TEEN SUICIDE
3.3.1 According to the Office of National Statistics, the overall suicide rate in the UK has been falling since the 1980s:
The decline has been happening for all age groups, except for the youngest, ages 10-29. Below is the graph for females only [need to get males]
When we look more closely at this youngest age group, by sex and by age, we find that the rate for teen girls in recent years has been rising, as explained in this Guardian article from Sept. 2018: [note that all that matters is rates per 100k, given that population is growing]
There were 177 suicides among 15- to 19-year-olds in 2017, compared with 110 in 2010 and more than in every year since then except 2015, when the toll was 186, the Office of National Statistics data shows. Fifty-six girls and women in the age group killed themselves last year, the highest number since records began in 1981. The suicide rate among that group, 3.5 per 100,000 people, was also the highest on record, and well up on the rate of 2.1 per 100,000 in 2010. The suicide rate among boys and men that age climbed to 7.1 per 100,000. There were 121 young male suicides last year, compared to 74 in 2010.
Zach Rausch created this Google spreadsheet to copy over the relevant data from the ONS site: raw numbers of suicides, and rates per 100k, for the 2 youngest age groups, for England and Wales. The sheet shows that the raw numbers of suicides for 10-14 year olds are so low --single digits -- that we are in the territory that Chris Ferguson warned about: A jump from 5 to 10 should not be called a 100% increase, it should be ignored. Let us therefore focus on the next age group that ONS breaks out: teens 15-19, where raw numbers are much higher. Here is the subset of that spreadsheet with the most important information. You can see that for both sexes, the rates are higher, in the last five years, than they were during the several years before that. But you can also see that, for boys, the history is similar to that in the USA: rates have gone up a bit in the last few years, but rates were far higher in the 1980s and 1990s, and into the early 2000s. So it would not be correct to say that there is a clear rise in suicides among UK boys aged 15-19.
Year | England & Wales Boys, 15-19 Deaths | England & Wales Boys, 15-19 Rate | England & Wales Girls, 15-19 Deaths | England & Wales Girls, 15-19 Rate |
2022 | 117 | 6.9 | 51 | 3.2 |
2021 | 135 | 8.2 | 63 | 4.0 |
2020 | 115 | 6.8 | 45 | 2.8 |
2019 | 121 | 7.2 | 64 | 4 |
2018 | 134 | 8 | 59 | 3.7 |
2017 | 121 | 7.1 | 56 | 3.5 |
2016 | 116 | 6.7 | 44 | 2.7 |
2015 | 135 | 7.7 | 51 | 3.1 |
2014 | 114 | 6.5 | 42 | 2.5 |
2013 | 112 | 6.3 | 23 | 1.4 |
2012 | 99 | 5.5 | 26 | 1.5 |
2011 | 100 | 5.6 | 41 | 2.4 |
2010 | 74 | 4.1 | 36 | 2.1 |
2009 | 99 | 5.5 | 33 | 1.9 |
2008 | 97 | 5.4 | 33 | 1.9 |
2007 | 96 | 5.3 | 32 | 1.8 |
2006 | 87 | 4.9 | 34 | 2 |
2005 | 103 | 5.9 | 42 | 2.5 |
2004 | 112 | 6.4 | 48 | 2.9 |
2003 | 107 | 6.2 | 35 | 2.1 |
2002 | 120 | 7.1 | 32 | 2 |
2001 | 148 | 9 | 32 | 2 |
2000 | 122 | 7.5 | 47 | 3 |
1999 | 131 | 8.1 | 43 | 2.7 |
1998 | 154 | 9.6 | 40 | 2.6 |
1997 | 134 | 8.5 | 36 | 2.3 |
1996 | 118 | 7.6 | 51 | 3.4 |
1995 | 104 | 6.8 | 41 | 2.8 |
1994 | 114 | 7.6 | 27 | 1.9 |
1993 | 122 | 8.1 | 27 | 1.9 |
1992 | 131 | 8.3 | 33 | 2.2 |
1991 | 153 | 9.2 | 47 | 3 |
1990 | 154 | 8.8 | 39 | 2.3 |
1989 | 158 | 8.7 | 54 | 3.1 |
1988 | 183 | 9.7 | 51 | 2.8 |
1987 | 156 | 8.1 | 52 | 2.8 |
1986 | 138 | 7 | 46 | 2.4 |
1985 | 132 | 6.5 | 39 | 2 |
1984 | 129 | 6.3 | 35 | 1.8 |
1983 | 117 | 5.5 | 39 | 1.9 |
1982 | 113 | 5.3 | 38 | 1.9 |
1981 | 140 | 6.6 | 52 | 2.6 |
For girls, the story is different. As in the USA, there was no big rise in suicides back in the 1980s and 1990s. As in the USA, rates for girls are more stable than for boys over the decades. The rates for girls were about as high as they are now from roughly 1995 to 2005, then they dipped, and then they increased in the last few years. So it would not be correct to say that there is an “epidemic” or that recent suicide rates among girls are unprecedented, or that there is a “suicide generation.” But, taken in the context of rising rates of self harm, anxiety, and depression, for girls more than for boys, the recent rise in rates of suicide for girls (not for boys) may well be more than random fluctuations.
FIGURE (Using the data from the table above):
[Note how similar this graph of UK data is to the graph of USA data in section 3.2.1, the one that plots suicide rates by sex since 1975. Both show a big rise for boys in the 1980s -- the days of the crime wave--dropping down in the 1990s for the USA and in the 2000s for the UK, before rising again in the 2010s. For the girls, both graphs show no trend until an increase in the 2010s]
FIGURE 2: Here’s a look at this same dataset but broken down by different age groups (Female and Males).
3.4 UK: COUNTERPOINTS AND CRITICISMS
3.4.1 Chivers (2019). Do we really have a ‘suicidal generation’? Unherd.
EXCERPT: “First, it’s worth noting that very few teenagers kill themselves. The total number of suicide deaths among 15- to 19-year-olds in 2017 in England and Wales was 177, out of about 3.25 million. That means that small changes can look like big percentage swings. More important, though, the Sunday Times story did exactly what the climate deniers did. The year 2010 had the lowest rate of teen suicides of any year since at least 1981, when the ONS records begin. You could compare it with literally any other year and you’d see a rise.”
COMMENT: I think Chivers is correct that there is not a “suicide generation” in the UK, and that the choice of a single year--2010--is a bad choice for a baseline because it was an unusually low year for boys (though not for girls). However, as we see above (section 3.3), and in most sections of this document, when you examine girls and boys separately, things look worse for girls. If we take the average of the years 2000 to 2009 as the baseline (that is roughly the decade before social media became popular with teens), how much has the rate increased in 2019, compared to that baseline? 13.9% percent for boys, and 81% for girls:
Teens aged 15-19 | Avg of 2000-2009 | 2019 | % increase in the rate |
Male | 6.32 | 7.2 | 13.9% |
Female | 2.21 | 4 | 81% |
[What are we missing?]
3.5 WHAT IS HAPPENING IN OTHER ENGLISH-SPEAKING COUNTRIES?
3.5.1 Canada
See our separate Google doc that examines a variety of trends: The Coddling of the Canadian Mind.
Suicide Statistics for Canadian Teens
There is journalistic reporting on a rise in teen suicide, e.g., a BBC article: Teen suicide on the rise among Canadian girls.
--Here are various pages at Statistics Canada.
Deaths and age-specific mortality rates, by selected grouped causes. Statistics Canada.
[note from Haidt: the only group that shows a general increase in the 2010s are the girls.]
Here are Canadian suicide rates split by age groups and gender:
--Here is the Canada graph from 3.5.1.1, Padmanathan. Boys suicide rate is the top line; girls is the bottom; combined is the middle:
3.5.2 Australia
See our separate Google doc that examines a variety of trends: The Coddling of the Australian Mind. From that document:
Suicide Statistics for Australian Teens
From Australia Bureau of Statistics, graphed by Zach Rausch from the individual yearly reports
Note: The suicide rate in the years after teen social life moved onto social media (roughly 2009 to 2011) is generally higher than before, for both sexes, but the rise is not as clear and sharp as in the USA. Girls leveled off after 2017, while boys have generally increased.
--Here is the Australia graph from 3.5.1.1, Padmanathan. Boys suicide rate is the top line; girls is the bottom; combined is the middle:
3.5.3 New Zealand
See our separate Google doc that examines a variety of trends: The Coddling of the Kiwi Mind. From that document:
New Zealand Ministry of Health Suicide Data. Graph created by Zach Rausch.
Note: New Zealand teen suicide rates do NOT seem to be increasing in the period under study -- post 2012.
[What are we missing?]
3.6 WHAT IS HAPPENING OUTSIDE THE ANGLOSPHERE?
For country-specific data, see our additional international docs at jonathanhaidt.com/reviews. The studies included below only include cross-national studies. Also see Zach’s series of international posts that examine trends in the five Nordic countries and throughout Europe.
3.6.1 Padmanathan, Bould, Winstone et al. (2020). Social media use, economic recession and income inequality in relation to trends in youth suicide in high-income countries: a time trends analysis. Journal of Affective Disorders
ABSTRACT: We analysed trends in suicide rates in 15-24 year olds over the period 2000-2017 in high-income countries with populations >20 million using Joinpoint analysis. We investigated differences in the following population-level exposures between countries that are and are not experiencing suicide rates rises: 1) 2008 economic recession as indexed by changes in GDP; 2) Gini income inequality; 3) daily social media use.
Results: Four of the 11 countries studied are experiencing youth suicide rate rises: Australia, Canada, the UK, the USA. The year the increase began ranged from 2003 (95% confidence interval: 2002, 2007) in the UK to 2009 (95% CI: 2007, 2012) in Australia. There was little evidence of an association between social media use and youth suicide trends, and inconsistent evidence regarding the impact of the 2008 economic recession. Suicide rate rises were seen in countries with higher GDP per capita (Wilcoxon rank sum (WRS) z=-2.27; p=0.02) and income inequality (WRS z=-2.45; p=0.01) in 2008.
[Note from Haidt: The online appendices for this article are very important, since they show trends for boys and girls separately. Korea and Japan are very different from the 9 Western countries: they have low rates of social media use, and they have high suicide rates that peaked around 2009 and have been declining since then. In the 9 Western countries rates of social media use are high and rising, and rates of suicide for boys are declining in 7, rising only in the UK and USA. But rates for girls are only declining (slightly) in France and Italy. In the other seven countries they are either flat (Germany and Poland) or rising (Australia, Canada, Spain, UK, USA). So trends are more negative for girls, especially in the English speaking countries.]
FIGURES:
3.6.2 OECD Data (2017). Teenage suicides (15 - 19 years old). Directorate of Employment, Labour and Social Affairs.
ABSTRACT: Teenage suicide rates are calculated as the total number of deaths due to ‘intentional self-harm’ among the population aged 15-19 years old in a given year, divided by the total population of 15-19 years old that year, and multiplied by a factor of 100 000. The result is expressed in units of suicides per 100 000 individuals (aged 15-19) per year.
KEY FINDINGS: Teenage suicides rates have, on average, declined slightly over the past two decades or so (Chart CO4.4.A). While in 1990 there were, on average across the OECD, 8.5 suicides per 100 000 teenagers (15-19), by 2015 this rate had fallen to 7.4. Much of this decline occurred during the 2000s. Between 1990 and 1999 the OECD average teenage suicide remained fairly stable at around 8.4 suicides per 100,000, but this average fell across the 2000s before reaching a low of 6.3 per 100,000 in 2007. With the exception of 2008, the average rate remained lower than 7.0 until 2014, although it increased slightly in 2014 and 2015.
NOTE: There was a sharp increase in suicides among Iceland teens between 2013 and 2015.
3.6.3 World Health Organization (2019). Global Health Estimates 2019: Estimated deaths by age, sex, and cause. Graphs made by Zach Rausch (see his spreadsheet).
NOTE: Details to come. See global suicide trends, along with trends broken down by the Anglosphere, Northern Europe, and Southern Europe.
3.6.4 Rausch (2024). The Youth Mental Health Crisis is International, Unless You Rely on a Flawed International Dataset. After Babel.
EXCERPT: Many researchers and organizations, such as the World Health Organizations rely on the Global Burden of Disease study (entirely or partially) for their data on international suicide rates. That turns out to be a big problem. That turns out to be a big problem. I show that the GBD—in every case that I looked at—systematically and substantially underestimates official statistics, and fails to detect the large increases that happened in the 2010s. Therefore the GBD should not be used when trying to understand changes in youth mental health or the causes of those changes.
Figure. According to CDC data (left side), U.S. adolescent (ages 10-14) emergency department visits for self-harm have been rising rapidly, particularly among girls. According to Global Burden of Disease (GBD) estimates (right side), self-harm prevalence rates among U.S. young adolescents (ages 10-14) have risen slightly for girls and declined for boys. Sources: Institute for Health Metrics and Evaluation, Global Burden of Disease. Centers for Disease Control and Prevention, National Center for Injury Prevention and Control WISQARS database. See Zach’s CDC spreadsheet for data points.
3.6.5 Rausch & Haidt (2023). Suicide Rates Are up for Gen Z Across the Anglosphere, Especially for Girls. After Babel.
EXCERPT: It turns out that in each of the five Anglosphere nations, suicide rates among adolescent girls (ages 10-19) are at all-time highs (since modern records have been kept). Suicide rates for Gen Z girls now surpass those seen during the adolescent and young adult years of Millennials (born 1981-1995), Gen X (born 1965-1980), and Boomers (born 1945-1964).
For adolescent boys, I’ll show that the changes are more varied across nations and that only some nations are reaching historical peaks. I was surprised to discover that it was the Gen X boys who showed the most consistent peaks in suicide rates (back in the 1980s and early 1990s) across the five Anglosphere nations.
In short, the suicide data among Gen Z girls is consistent with the story we have told on this Substack about rising anxiety, depression, and self-harm. The adolescent mental health crisis is real, international, and gendered. And if that is true, then the many theories put forth to explain the mental health crisis in the U.S. may have only limited explanatory power. We must also look for something that changed in the lives of adolescents, and especially adolescent girls, in many countries simultaneously, in the years around 2010.
Figure. U.S. suicide rates across the lifespan by generation. Gen Z females have higher rates than all other generations at the same age. Gen Z males have higher rates than Boomers and Millennials but look similar to Gen X until their early 20s. Note that Gen Z data for those ages 20-24 only include 2020 and 2021, the COVID years (this is true for all nations) (Source: CDC WISQARS Fatal Injury Report, and CDC WONDER). (spreadsheet with graphs and data points).
[What are we missing?]
* * * * * * * * * * * * * * *
SECTION 4: SUMMARY OF FINDINGS
The studies we listed above, and the graphs we excerpted or plotted ourselves, lead to the following 6 tentative findings:
1) Depression and anxiety: The rise seems to be real; it is found using a variety of questions and measures, in the USA and the UK, as well as in Canada, Australia, and New Zealand. The increases for girls are found consistently, and are often large. The story for boys is less clear: Most studies find increases, but some do not.
2) Self harm and suicide attempts: The rise seems to be real, in the USA, UK, Australia, New Zealannd, and (with much less data) Canada, for girls only. There seems to be no increase at all for teen boys in any of those three countries for self-harm, and smaller increases for boys in suicide attempts.
3) Suicide: The rise seems to be real in the USA. The rise for teen boys is not much larger than the rise for older male age groups, so this may not reflect a big change for boys. But the rise for teen girls is much larger, as a percentage change, than for older groups of women, and it is now the highest it has been since the CDC began collecting data in 1980. (But see the debate between Haidt and Ferguson about whether percent change is the proper statistic to examine). Trends in the UK are weaker, with no clear sign of an increase for teen boys. The rate for teen girls in 2017 is up 30% from the average rate between 2000 and 2009, but this does not bring teen girls up to their highest level ever. Yet in the context of the UK, where suicide rates are falling overall, it is noteworthy that the only group for which they are rising is teen girls. We need to find more recent data (up through 2017) to draw conclusions about Canada, Australia, and New Zealand.
4) Gender differences: These are large and consistent across studies and nations: In most studies that allow comparisons, girls are doing worse than boys in absolute terms (except for completed suicides, where boys’ rates are always higher). In nearly all studies that examine changes since the years before 2009, the increases in poor mental health are larger for girls.
5) Age differences (younger teens vs. older teens): Some studies allow us to compare younger teens (ages 10-13) and older teens (14 or 15 to 19 or 20). The few studies that allow us to do so indicate that the percentage increase in mental health issues, self-harm, and suicide attempts for younger girls (ages 10 to 15) in recent years is generally larger than for older teen girls (16+). More work is needed here.
6) USA vs. UK: Trends are generally similar for anxiety, depression, and self-harm. The trends for suicide are larger in the USA, where it is more clear that there is a serious problem for teenagers, with higher rates for boys (as as long been the case), but larger increases (percentagewise) for girls.
5. CONCLUSION
[written in 2020; need update for 2022]
Many studies, using different methods and asking different questions, indicate that rates of adolescent anxiety, depression, and self harm are rising for girls in the years after (roughly) 2012, compared to the years before (roughly) 2009, in the US and in the UK, as well as in other English speaking countries. The increases for boys are sometimes smaller and are found less consistently. But even though boys generally have lower rates of internalizing disorders, the increases after 2012 are often just as large as those for girls in percentage terms, even though a 30% increase does not look dramatic on a graph when the initial rate is lower.
The increases in teen suicide rates are more clear in the USA than in the UK, but in both countries, the percent increase is larger for girls. Because the changes are seen in behavioral acts (such as self harm, suicide attempts, and suicide) which increase at roughly the same time as the self-reports of anxiety, depression, and suicidal ideation, these increases seem to reflect more than just changing diagnostic criteria or changes in the willingness of Gen Z to report distress. The increase appears to be real, and we must look for its causes. The timing of the rise is consistent with the hypothesis that the widespread entrance of teens to social media platforms between 2009 and 2012, and the smartphone gaining market saturation (50%+ ownership) in 2012 is a contributor to the rise in mental health problems after 2012. Of course, correlation does not imply causation, so we have created a separate literature review to examine whether social media has played a causal role in this increase.
[more to come. What is the main point that skeptics would like to note in the conclusion?]
ADDITIONAL NOTES AND STUDIES:
APPENDICES
APPENDIX A: VARIATION BY SEX
[Temporary placeholder; this will be filled out in summer 2023] This document has focused on internalizing disorders: depression and anxiety, plus their most-studied behavioral consequences: self harm and suicide. Internalizing disorders are more common in girls and women than in boys and men; the gap has long been known to widen at puberty. Boys and men have higher rates of externalizing disorders (e.g., crime, violence, drug use….). I (Haidt) have been more focused on internalizing disorders since that is the specific area that is at the center of The Coddling of the American Mind; the fearfulness and fragility that seem to be produced by a childhood surrounded by overprotection and social media. But all along on this project, I have heard from boys, and parents of boys, about just how badly American boys are doing, and their problems are somewhat different from those of girls. They seem to be related to getting sucked into their screens: video games, porn, online radicalization, with terrible outcomes that seem more related to failure to grow or launch than to internalizing disorders.
We are working on a second google doc for boys mental health issues.
APPENDIX B: VARIATION BY AGE (INCLUDING ADULTS)
In order to determine whether trends are unique to Gen Z, we looked at adulthood trends of depression, anxiety, self-harm, and suicide. Below is a brief look at the literature. Basically, levels are flat or declining after 2010, which is the period in which we see sharp and steady rises for teens.
B.1 Yu, Zhang, Wang, Sun, Jin, & Liu (2020). Trends in depression among Adults in the United States, NHANES 2005–2016. Journal of Affective Disorders, 263, 609–620.
ABSTRACT: OBJECTIVE: To describe the prevalence and trends of mild, moderate and severe depression among adults and all age groups in the US from 2005 to 2016, and analysis the risk factors for depression.
METHODS: This study analyzed the prevalence and association with risk factors of depression using weighted univariate logistic regression model. Data of NHANES 2005-2016 were used.
RESULTS: This study analyzed 29,303 participants. From 2005 through 2016, the trends of severe depression increased among ≥ 20 years (p for trend = 0.026, difference, 0.109[0.012,0.207]); severe depression increased among ≥ 65 years (p for trend <0.001, difference, 0.302[0.170,0.435]); and moderate depression increased among 20–39 years (p for trend = 0.028, difference, 0.137[0.045,0.229]). In adults, the odds ratios (OR) (95% confidence intervals (CI)) of mild depression for < 25000$ was 2.24 (1.96, 2.55), moderate depression was 4.94(3.91,6.24), and severe depression was 6.45 (4.78,8.71); the OR (95%CI) of mild depression for smoking was 1.69 (1.55,1.84), moderate depression was 2.94 (2.57,3.35), and severe depression was 3.36 (2.87,3.93); the OR(95%CI) of mild depression for hypertension was1.27 (1.18,1.38), moderate depression was 1.50 (1.31,1.73), and severe depression was 1.94 (1.61,2.34); the OR(95%CI) of mild depression for diabetes mellitus was1.45 (1.30,1.61), moderate depression was 1.83 (1.51,2.22), and severe depression was 2.05 (1.70,2.48).
CONCLUSIONS: There was an increasing trend of severe depression in American adults, which was mainly manifested in the increasing trend of severe depression in the population ≥65 years. And the trend of moderate depression increased in 20-39 years. In addition, lower income, smoking, hypertension and diabetes mellitus increased the risk of depression, and the risk increased with the degree of depression.
FIGURE:
B.2 Goodwin, Weinberger, Kim, Wu, & Galea (2020). Trends in anxiety among adults in the United States, 2008–2018: Rapid increases among young adults. Journal of Psychiatric Research. (This also includes young adults (see 2.1.7 for the stats on young adults)
ABSTRACT: INTRODUCTION: In a time of global uncertainty, understanding the psychological health of the American public is imperative. There are no current data on anxiety trends among adults in the United States (US) over time. This study aimed to investigate prevalence of anxiety among US adults from 2008 to 2018.
METHODS: Data from the National Survey on Drug Use and Health (NSDUH), which is an annual, cross-sectional survey on substance use and mental health in the US, were analyzed in 2020. Prevalence of past-month anxiety was estimated among those ages ≥18, by survey year from 2008 to 2018. Time trends were tested using logistic regression.
RESULTS: Anxiety increased from 5.12% in 2008 to 6.68% in 2018 (p < 0.0001) among adult Americans. Stratification by age revealed the most notable increase from 7.97% to 14.66% among respondents 18–25 years old (p < 0.001), which was a more rapid increase than among 26–34 and 35–49 year olds (differential time trend p < 0.001). Anxiety did not significantly increase among those ages 50 and older. Anxiety increased more rapidly among those never married and with some college education, relative to their respective counterparts. Apart from age, marital status and education, anxiety increased consistently among sociodemographic groups.
CONCLUSIONS: Anxiety is increasing among adults under age 50 in the US, with more rapid increase among young adults. To prepare for a healthier adulthood and given direct and indirect (via 24/7 media) exposure to anxiety-provoking world events, prophylactic measures that can bolster healthy coping responses and/or treatment seeking seem warranted on a broad scale.
FIGURE:
B.3 Daly (2022). Prevalence of psychological distress among working-age adults in the United States, 1999–2018. American Journal of Public Health.
ABSTRACT: OBJECTIVES: To test whether the prevalence of reported psychological distress increased among working-age adults in the United States between 1999 and 2018.
METHODS: I examined psychological distress in the past 30 days using the Kessler-6 Distress Scale, completed by 403 223 participants aged 25 to 64 years across 20 annual waves of the National Health Interview Study conducted from 1999 to 2018. I examined overall and demographic-specific trends.
RESULTS: The prevalence of psychological distress in the past 30 days increased from 16.1% in 1999–2000 to 22.6% in 2017–2018, an increase of 6.5 percentage points (95% confidence interval [CI] = 5.6, 7.3) or 40% from 1999–2000 levels. Statistically significant increases in the prevalence of distress were observed across all age, gender, race/ethnicity, and educational attainment subgroups examined. Rates of serious psychological distress increased from 2.7% in 1999–2000 to 4% in 2017–2018, an increase of 1.3 percentage points (95% CI = 0.9, 1.6).
[note from Haidt: the youngest people in this study are millennials; there are no Gen Z]
FIGURE 1:
FIGURE 2:
B.4 Giuntella et al. (2022). The Midlife Crisis. Working paper, NBER.
ABSTRACT: This paper documents a longitudinal crisis of midlife among the inhabitants of rich nations. Yet middle-aged citizens in our data sets are close to their peak earnings, have typically experienced little or no illness, reside in some of the safest countries in the world, and live in the most prosperous era in human history. This is paradoxical and troubling. The finding is consistent, however, with the prediction – one little-known to economists – of Elliott Jaques (1965). Our analysis does not rest on elementary cross-sectional analysis. Instead the paper uses panel and through-time data on, in total, approximately 500,000 individuals. It checks that the key results are not due to cohort effects. Nor do we rely on simple life-satisfaction measures. The paper shows that there are approximately quadratic hill-shaped patterns in data on midlife suicide, sleeping problems, alcohol dependence, concentration difficulties, memory problems, intense job strain, disabling headaches, suicidal feelings, and extreme depression. We believe the seriousness of this societal problem has not been grasped by the affluent world’s policy-makers.
[Note from Haidt: until the 2010s, young adults had low rates of mental distress, compared to those in their 40s and 50s. But as you can see in some of the studies in this Collaborative Review document, it is primarily teens and young adults (gen z) whose mental distress has increased rapidly, closing the gap on some measures with middle aged adults. The graphs below are from p. 28 of the report]
B.5 Richesson & Hoenig (2020) Key Substance Use and Mental Health Indicators in the United States: Results from the 2020 National Survey on Drug Use and Health.
EXCEPRT: Among adults aged 18 or older in 2020, 8.4 percent (or 21.0 million people) had a past year MDE, and 6.0 percent (or 14.8 million people) had a past year MDE with severe impairment. Percentages for past year MDE and MDE with severe impairment were highest among young adults aged 18 to 25, followed by adults aged 26 to 49, then by adults aged 50 or older. Among young adults, 17.0 percent (or 5.6 million people) had a past year MDE, and 12.1 percent (or 4.0 million people) had a past year MDE with severe impairment… Among adolescents aged 12 to 17 in 2020, 17.0 percent (or 4.1 million people) had a past year major depressive episode (MDE), and 12.0 percent (or 2.9 million people) had a past year MDE with severe impairment.
FIGURES:
B.6 Centers for Disease Control, Fatal Injury Data.
Unlike most of the developed world, where suicide rates have been falling, rates in the USA have been rising in recent years, for nearly all age groups, and for both sexes:
Source: CDC data, grouped into 5 year age bins, originally graphed by Chris Vaccaro (updated by Zach Rausch, see his spreadsheet). Rates are per 100,000 in the population of that age/gender group.
When we look at the percentage change for each group, however, we see very large age differences and sex differences. I took as the baseline the average rate from 2000 to 2009, and then calculated how much above that baseline the rate was in 2020, for each of the age groups graphed in the charts above.
APPENDIX C: VARIATION BY RACE/ETHNICITY
C.1 Lipson… & Eisenberg (2022). Trends in college student mental health and help-seeking by race/ethnicity: Findings from the national healthy minds study, 2013–2021. Journal of Affective Disorders.
BACKGROUND: A considerable gap in knowledge exists around mental health trends in diverse racial and ethnic adolescent and young adult populations. The purpose of this study is to examine annual trends for mental health and help-seeking by race/ethnicity in a national sample of college students.
METHODS: Survey data come from >350,000 students at 373 campuses that participated in the Healthy Minds Study between 2013 and 2021. Analyses are descriptive in nature focusing on year-by-year prevalence and help-seeking rates for each racial/ethnic group.
RESULTS: In 2020–2021, >60% of students met criteria for one or more mental health problems, an early 50% increase from 2013. Mental health worsened among all groups over the study period. American Indian/Alaskan Native students experienced the largest increases in depression, anxiety, suicidal ideation, and meeting criteria for one or more mental health problem. Students of color had the lowest rates of mental health service utilization. The highest annual rate of past-year treatment for Asian, Black, and Latinx students was at or below the lowest rate for White students. Although Arab American students experienced a 22% increase in prevalence, there was an 18% decrease in treatment.
LIMITATIONS: Response rates raise the potential of nonresponse bias. Sample weights adjust along known characteristics, but there may be differences on unobserved characteristics.
CONCLUSIONS: Findings have important implications for campus mental health programming and underscore the urgency of reducing mental health inequalities in college student populations through the identification and implementation of best practices both in clinical settings and through system-level change.
FIGURE: Created by Zach Rausch, see spreadsheet.
C.2 National Survey on Drug Use and Health: Cases of Major Depressive Disorders by Race/Ethnicity (2004-2022). Substance Abuse and Mental Health Services Administration Center for Behavioral Health Statistics and Quality.
Figure created by Zach Rausch. See spreadsheet.
C.3 Youth Risk Behavior Survey Data Summary & Trends Report: 2013-2023. CDC.
NOTE: See Jon’s commentary on the 2021 YRBS survey. He argues that the new CDC report shows that Covid added little to teen mental health trends.
C.4 Self-Harm Rates by Race and Gender. Center’s for Disease Control (2001-2020). Graphs created by Zach Rausch. See his spreadsheet.
C.5 Suicide Rates by Race and Gender. Center for Disease Control (2000-2020). Graphs created by Zach Rausch. See his spreadsheet.
APPENDIX D: VARIATION BY GENDER / SEXUALITY
D.1 Bettis & Liu (2019). Population-Based Analysis of Temporal Trends in the Prevalence of Depressed Mood Among Sexual Minority and Heterosexual Youths From 1999 Through 2017. JAMA Pediatrics.
ABSTRACT: Depression in adolescence is highly prevalent and associated with negative long-term outcomes. Despite decades of research on treatment for adolescent depression, sexual minority youths remain a particularly at-risk group.2 Temporal trends inform progress in addressing the need to eliminate health disparities among sexual minority populations.3 To our knowledge, this study presents the first population-representative analysis of temporal trends in depressed mood among sexual minority and heterosexual youths. An 18-year period is examined.
METHODS: The Youth Risk Behavior Surveillance System (YRBSS) obtains biannual data representative of students in grades 9 through 12 using a multistage cluster-sample design. Data were drawn from the Massachusetts YRBSS for calendar years 1999 through 2017. This study did not undergo institutional review board review at any institution but was believed by the authors to be exempt because the dataset used is publicly available for use (https://www.cdc.gov/healthyyouth/data/yrbs/data.htm). These analyses used previously collected data through the YRBSS, a national survey conducted by the US Centers for Disease Control and Prevention, who works with local schools to collect data and obtain parental permission. The unweighted total study population was 33 456 individuals. The percentage of non-Hispanic white youths in the sample ranged from 42.5% among sexual minority youths reporting sexual identity in 2017 to 73.3% among heterosexual youths reporting sexual behavior in 2003 (Table 1).
RESULTS: Table 1 presents depressed mood prevalence rates stratified by sexual identity and behavior. Analyses based on sexual identity (Table 2) revealed heterosexual youths demonstrated a significant decrease in depressed mood from 1999 to 2013 (annual percentage change, −2.31% [95% CI, −3.67% to −0.92%]; P = .01), with no significant change from 2013 to 2017. A significant decline was not observed for sexual minorities from 1999 to 2017. In a sensitivity analysis, the trend for sexual minorities remained nonsignificant. When sexual orientation was based on sexual behavior, a significant decrease in depressed mood was observed for heterosexual youths (annual percentage change, −3.33% [95% CI, −4.93% to −1.70%]; P = .003), but not sexual minority youths, between 1999 and 2009.
DISCUSSION: Prevalence of depressed mood across all years was high, with especially concerning rates reported in sexual minority youths across the study period. The current study found evidence of a decline in depressed mood among heterosexual youths over time, while depressed mood rates among sexual minority peers have remained largely unchanged in nearly 2 decades. These findings collectively suggest that disparities in rates of feeling depressed in sexual minority youths populations have not improved over the last 18 years.
D.2 Ross, Salway, Tarasoff, MacKay, Hawkins, & Fehr (2018). Prevalence of Depression and Anxiety Among Bisexual People Compared to Gay, Lesbian, and Heterosexual Individuals:A Systematic Review and Meta-Analysis. Journal of Sex Research.
ABSTRACT: Over the past decade, evidence has accumulated to suggest that bisexual people experience higher rates of poor mental health outcomes compared to both heterosexual and gay/lesbian individuals. However, no previous meta-analyses have been conducted to establish the magnitude of these disparities. To address this research gap, we conducted a systematic review and meta-analysis of studies that reported bisexual-specific data on standardized measures of depression or anxiety. Of the 1,074 full-text articles reviewed, 1,023 were ineligible, predominantly because they did not report separate data for bisexual people (n = 562 studies). Ultimately, 52 eligible studies could be pooled in the analysis. Results indicate that across both outcomes, there is a consistent pattern of lowest rates of depression and anxiety among heterosexual people, while bisexual people exhibit higher or equivalent rates in comparison to lesbian/gay people. On the basis of empirical and theoretical literature, we propose three interrelated contributors to these disparities: experiences of sexual orientation-based discrimination, bisexual invisibility/erasure, and lack of bisexual-affirmative support. Implications for interventions to improve the health and well-being of bisexual people are proposed.
D.3 Liu (2019). Temporal Trends in the Prevalence of Nonsuicidal Self-injury Among Sexual Minority and Heterosexual Youth From 2005 Through 2017. JAMA Pediatrics.
ABSTRACT: Nonsuicidal self-injury (NSSI) has received increasing attention in recent years as a clinically important phenomenon; it has been included in the DSM-5 as a syndrome warranting further investigation. Sexual minority youth are particularly at risk for NSSI. An important step toward addressing this issue is to characterize the prevalence of NSSI among these individuals. Additionally, data on temporal trends are needed to inform progress in addressing the stated need to eliminate health disparities among sexual minorities. The current study presents the first (to my knowledge) population-representative analysis of temporal trends in NSSI among sexual minority and heterosexual youth over a 13-year period.
METHODS: The Youth Risk Behavior Surveillance System obtains biannual data representative of students in grades 9 through 12 (with age group ranging from 12 years and younger to 18 years and older).3 Data were drawn from the Massachusetts Youth Risk Behavior Surveillance System for 2005 to 2017. Massachusetts was the first state to assess sexual orientation and started assessing NSSI in 2005. This study used publicly available secondary data and was exempt from institutional review board review. The deidentified nature of the data rendered informed consent for the current study unnecessary. Sexual orientation was assessed with an item of self-reported sexual identity and another of same-sex behavior. For sexual identity, respondents self-identifying as gay, lesbian, bisexual, and not sure were classified as sexual minorities based on identity. For same-sex behavior, respondents who had had same-sex partners in their lifetime were classified as sexual minorities based on behavior. Those who had had no sexual partners were excluded from analyses associated with sexual behavior. Respondents were asked a single item of past-12-month NSSI (ie, intentional self-harm without wanting to die).
RESULTS: Table 1 presents NSSI prevalence rates from 2005 to 2017, stratified by sexual identity and sexual behavior. The unweighted number of participants was 21, 213. The NSSI prevalence rates ranged from 10.79% (SE, 0.63%) to 20.41% (SE, 1.58%) among heterosexual youth and from 38.04% (SE, 2.89) to 52.97% (SE, 4.32%) among sexual minority youth across the study period. When sexual orientation was based on sexual identity, a significant decrease in NSSI was observed across this period for heterosexual youth (annual percentage change, −2.51 [95% CI, −4.75 to −0.21]; P = .04) but not for sexual minority youth (Table 2). In a sensitivity analysis excluding respondents unsure of their sexual identity, the trend for sexual minorities remained nonsignificant. When sexual orientation was based on sexual behavior, a similar pattern of results was obtained, but no significant changes over time in any group were noted.
DISCUSSION: The NSSI prevalence rates ranged from 11% to 20% among heterosexual youth and 38% to 53% among sexual minority youth across the same period. Given that single-item measures of NSSI have been found to yield lower prevalence estimates,4 it is possible that the actual rates may be even higher. Although the current study found a decline in NSSI among heterosexual youth since 2005, prevalence rates were nonetheless generally high across all years. Furthermore, among sexual minority peers, the rates have remained largely unchanged, suggesting that disparities in NSSI rates in sexual minority youth populations have not improved over the last 13 years. The absence of a decline in NSSI prevalence among sexual minority youth across this time period is all the more striking given that the very high rates of NSSI allow for more potential room for improvement. Collectively, these findings indicate that there remains much opportunity for progress in addressing this public health concern. The need for progress in this area is all the more pressing when the potential long-term mental health outcomes of adolescent NSSI are considered.
TABLE:
D.4 Liu, Sheehan, Walsh, Sanzari, Cheek, & Hernandez (2019). Prevalence and correlates of non-suicidal self-injury among lesbian, gay, bisexual, and transgender individuals: A systematic review and meta-analysis. Clinical Psychology Review.
ABSTRACT: The current review presents a meta-analysis of the existing empirical literature on the prevalence of non-suicidal self-injury (NSSI) among lesbian, gay, bisexual, and transgender (LGBT) individuals, as well as on correlates of NSSI within sexual and gender minority populations. Eligible publications (n = 51) were identified through a systematic search of PsycINFO, MEDLINE, and Embase, supplemented by a search of references of prior reviews on this topic. NSSI prevalence rates were quite elevated among sexual (29.68% lifetime) and gender (46.65% lifetime) minority individuals compared to heterosexual and/or cisgender peers (14.57% lifetime), with transgender (46.65% lifetime) and bisexual (41.47% lifetime) individuals being at greatest risk. Even among these group findings, sexual minority youth emerged as an especially vulnerable population. Moreover, current evidence suggests these rates and differences between LGBT and heterosexual and/or cisgender peers have not declined over time. These findings may in some measure be due to the existence of LGBT-specific risk correlates combined with general risk correlates being more severe among sexual and gender minority populations. Additional research, particularly employing a longitudinal design, is needed in this area to advance efforts to reduce risk for NSSI among sexual and gender minority individuals.
D.5 Liu, Walsh, Sheehan, Cheek, & Carter (2020). Suicidal Ideation and Behavior Among Sexual Minority and Heterosexual Youth: 1995–2017. Pediatrics.
NOTE: Supplemental material
ABSTRACT: OBJECTIVES: In this study, we determined trends in prevalence of suicidal thoughts and behaviors among lesbian, gay, bisexual, and heterosexual youth from 1995 to 2017 using population-based surveillance data.
METHODS: Data were drawn from the Massachusetts Youth Risk Behavior Survey from 1995 to 2017 (unweighted N = 41 636). The annual percent change (APC) in prevalence of suicidal ideation, plans, and attempts was stratified by sexual orientation as indexed by sexual identity and sexual behavior.
RESULTS: Among sexual minority youth, prevalence rates declined over the entire study period for suicidal ideation (APCsexual identity = −1.25; APCsexual behavior = −1.83), plans (APCsexual identity = −1.88; APCsexual behavior = −1.95), and attempts (APCsexual identity = −2.64; APCsexual behavior = −2.47). Among heterosexual youth, prevalence rates declined from 1995 to 2007 for suicidal ideation (APCsexual identity = −6.67; APCsexual behavior = −6.77) and plans (APCsexual identity = −5.73; APCsexual behavior = −6.25). These declines in ideation and plans were steeper than those for sexual minority youth. Prevalence of suicide attempts declined across the entire study period among heterosexual youth (APCsexual identity = −3.66; APCsexual behavior = −4.01). Prevalence of all 3 outcomes remained markedly high among sexual minority youth across the 23-year study period.
CONCLUSIONS: Although suicidal thoughts and behavior have generally declined among sexual minority and heterosexual youth, disparities in these outcomes persist, and their prevalence among sexual minority youth has remained consistently elevated. Prioritized screening for risk for suicidal thoughts and behaviors in this vulnerable population is imperative to reduce disparities and prevalence of these outcomes.
ADDITIONAL EXCERPT: Across this 23-year period, the prevalence rates of all suicide-related outcomes declined for both sexual minority and heterosexual youth. These results held true regardless of whether sexual orientation was based on self-identification or sexual behavior. In the case of suicidal ideation, a significant decline in prevalence was observed across the entire study period for sexual minority youth. Among heterosexual youth, a significant decrease was found from 1995 to 2007, after which a plateau appeared to emerge from 2007 to 2017. Additionally, the decline from 1995 to 2007 was significantly sharper for heterosexual youth than for sexual minority peers. For suicide plans, a similar set of findings was revealed. Specifically, sexual minority youth experienced a decrease in prevalence of suicide plans across the 23-year period, whereas heterosexual counterparts saw a significantly steeper decrease from 1995 to 2007 before reaching a plateau from 2007 to 2017. In the case of suicide attempts, however, both sexual minority and heterosexual youth saw a significant decrease in prevalence from 1995 to 2017.
D.6 Raifman… & McConnell (2020). Sexual Orientation and Suicide Attempt Disparities Among US Adolescents: 2009–2017. Pediatrics.
ABSTRACT: BACKGROUND: Sexual minority adolescents face mental health disparities relative to heterosexual adolescents. We evaluated temporal changes in US adolescent reported sexual orientation and suicide attempts by sexual orientation.
METHODS: We used Youth Risk Behavioral Surveillance data from 6 states that collected data on sexual orientation identity and 4 states that collected data on sex of sexual contacts continuously between 2009 and 2017. We estimated odds ratios using logistic regression models to evaluate changes in reported sexual orientation identity, sex of consensual sexual contacts, and suicide attempts over time and calculated marginal effects (MEs).
RESULTS: The proportion of adolescents reporting minority sexual orientation identity nearly doubled, from 7.3% in 2009 to 14.3% in 2017 (ME: 0.8 percentage points [pp] per year; 95% confidence interval [CI]: 0.6 to 0.9 pp). The proportion of adolescents reporting any same-sex sexual contact increased by 70%, from 7.7% in 2009 to 13.1% in 2017 (ME: 0.6 pp per year; 95% CI: 0.4 to 0.8 pp). Although suicide attempts declined among students identifying as sexual minorities (ME: −0.8 pp per year; 95% CI: −1.4 to −0.2 pp), these students remained >3 times more likely to attempt suicide relative to heterosexual students in 2017. Sexual minority adolescents accounted for an increasing proportion of all adolescent suicide attempts.
CONCLUSIONS: The proportion of adolescents reporting sexual minority identity and same-sex sexual contacts increased between 2009 and 2017. Disparities in suicide attempts persist. Developing and implementing approaches to reducing sexual minority youth suicide is critically important.
NOTE: Changes in reported sexual minority orientation identity and same-sex sexual contacts over time. A, Sexual minority orientation identity over time. B, Sex of sexual contacts over time. C, Sexual orientation among students with same-sex sexual contacts. D, Sexual orientation among students with opposite-sex sexual contacts. Note that y-axis scales range from 0% to 10% for all graphs except for the bottom left (C), where the scale is from 0% to 60%. Heterosexual sexual orientation identity and opposite-sex sexual contacts are excluded from the graphs, so the graphs can be presented at a scale that improves visualization of temporal changes in sexual minority identity and in same-sex sexual contacts.
NOTE: Changes in the proportion of adolescents reporting ≥1 suicide attempt over time, by sexual orientation identity and sex of sexual contacts. A, Suicide attempts by sexual orientation identity. B, Suicide attempts by sex of sexual contacts.
D.7 The Trevor Project National Survey (2021).
SUMMARY: The Trevor Project’s 2021 National Survey on LGBTQ Youth Mental Health sheds light on many of these challenges by capturing the experiences of nearly 35,000 LGBTQ youth ages 13–24 across the United States.
FIGURES 2021:
FIGURES 2020:
FIGURES 2019:
D.8 Marshal, Dietz, Friedman, Stall, Smith, McGinley, Thoma, Murray, D’Augelli, & Brent (2011). Suicidality and Depression Disparities Between Sexual Minority and Heterosexual Youth: A Meta-Analytic Review. Journal of Adolescent Health.
ABSTRACT: PURPOSE: To examine disparities between sexual minority youth (SMY) and heterosexual youth in rates of suicidality and depression symptoms.
METHODS: Separate meta-analyses were conducted to examine suicidality and depression disparities. Studies were included if the average age of the participants was <18 years, and if suicidality or depression symptoms were compared across SMY and heterosexual youth.
RESULTS: SMY reported significantly higher rates of suicidality (odds ratio [OR] = 2.92) and depression symptoms (standardized mean difference, d = .33) as compared with the heterosexual youth. Disparities increased with the increase in the severity of suicidality (ideation [OR = 1.96], intent/plans [OR = 2.20], suicide attempts [OR = 3.18], suicide attempts requiring medical attention [OR = 4.17]). Effects did not vary across gender, recruitment source, and sexual orientation definition.
CONCLUSIONS: Disparities in suicidality and depression may be influenced by negative experiences including discrimination and victimization. Clinicians should assess sexual orientation, analyze psychosocial histories to identify associated risk factors, and promote prevention and intervention opportunities for SMY and their families.
D.9 Kerr, Bae, & Dermody (2022). Trends and disparities in suicidal thoughts and behaviors and mental health symptoms among sexual and gender minority college students in the U.S., 2008–2018. Psychology of Sexual Orientation and Gender Diversity.
ABSTRACT: Prevalence estimates of mental health problems including suicidal thoughts and behaviors among college students have increased over the last decade, but time trends for sexual and gender minority (SGM) students remain undocumented. We compared trends in the adjusted prevalence of these outcomes by gender and sexual orientation in college students (n = 816,461) ages 18–24 years sampled from 4-year colleges that self-selected to participate in the cross-sectional National College Health Assessment in 2008–2018. Prevalence of 30-day depressive and anxiety symptoms increased significantly for every group examined—cisgender women, cisgender women, transgender students, and heterosexual, gay/lesbian, bisexual/pansexual, and unsure/questioning students—and 30-day suicidal ideation increased significantly for all but transgender and unsure/questioning students. Disparities that sexual minority and transgender students showed relative to heterosexual and cisgender peers widened for depressive and anxiety symptoms, and the narrowing of disparities on suicidal ideation and past-year suicide attempt were explained by increases in prevalence among cisgender women and heterosexual students. A redoubling of prevention efforts, particularly for SGM young people, is needed to reverse recent trends in psychological distress and suicide risk.
APPENDIX E: VARIATION BY IDEOLOGY
E.1 Gimbrone, Bates, Prins, & Keyes (2022). The politics of depression: Diverging trends in internalizing symptoms among US adolescents by political beliefs. SSM - Mental Health.
ABSTRACT: Adolescent internalizing symptoms (e.g. depressive affect) have increased over the past decade in the US, particularly among girls. The reasons for these increases are unclear. We hypothesize that increasing exposure to politicized events has contributed to these trends in adolescent internalizing symptoms, and that effects may be differential by political beliefs and sociodemographic characteristics. We analyzed nationally-representative data from 2005 to 2018 Monitoring the Future annual cross-sectional samples of 12th-grade students (N = 86,138). We examined self-reported political beliefs, sex, and parental education as predictors of four internalizing symptom scales over time, including depressive affect. From 2005 to 2018, 19.8% of students identified as liberal and 18.1% identified as conservative, with little change over time. Depressive affect (DA) scores increased for all adolescents after 2010, but increases were most pronounced for female liberal adolescents (b for interaction = 0.17, 95% CI: 0.01, 0.32), and scores were highest overall for female liberal adolescents with low parental education (Mean DA 2010: 2.02, SD 0.81/2018: 2.75, SD 0.92). Findings were consistent across multiple internalizing symptoms outcomes. Trends in adolescent internalizing symptoms diverged by political beliefs, sex, and parental education over time, with female liberal adolescents experiencing the largest increases in depressive symptoms, especially in the context of demographic risk factors including parental education. These findings indicate a growing mental health disparity between adolescents who identify with certain political beliefs. It is therefore possible that the ideological lenses through which adolescents view the political climate differentially affect their mental wellbeing.
E.2 Pew Research Center for the People & the Press. Pew Research Center: American Trends Panel Wave 64 (fielded March 2020).
The interaction of politics and mental health was first noticed by Zach Goldberg, in this Twitter thread. Haidt then downloaded the data and plotted it in a way that makes it easier to see the 3 way interaction of politics, gender, and age:
For more details see this chapter of the Afterword for The Coddling of the American Mind.
E.3 Gallup (via the AEI Survey Center on American Life) finds that there has been a sudden and large shift to the left in the politics of young women, beginning around 2015. Gen Z women begin to turn 18 in 2013 and enter this dataset (although we cannot tell if the shift that begins in 2015 is due to the entry of Gen Z women, or to a change in the politics of late millennial women). In any case, something has been changing in the politics of young women that does not seem to be happening to young men:
E.4 YouGov (2022) American Family Survey
Reported in this UnHerd essay;
Here is the survey writeup from 2022, but i don’t see the data for that question [ZR]
E.5 Anderson, Vogels, Perrin, & Rainie (2022). Connection, Creativity and Drama: Teen Life on Social Media in 2022. Pew Research Center.
Finds that Democrat-leaning teens are much more likely to use social media for political activity. It would be good to find cross-tabs of sex and ideology.
APPENDIX F: VARIATION URBAN VS. RURAL AND U.S. REGION
F.1 National Survey on Drug Use and Health: Cases of Major Depressive Disorders by USA Region (2004-2022). Substance Abuse and Mental Health Services Administration Center for Behavioral Health Statistics and Quality.
Figure created by Zach Rausch. See spreadsheet.
NOTE: Item details: “Had at Least One Major Depressive Episode (MDE) in Past Year among Persons Aged 12 to 17, by Demographic Characteristics: Percentages”
F.2 Evans, Huizink, Greaves-Lord, Tulen, Roelofs, & Ende (2020). Urbanicity, biological stress system functioning and mental health in adolescents. PLOS ONE.
ABSTRACT: Growing up in an urban area has been associated with an increased chance of mental health problems in adults, but less is known about this association in adolescents. We examined whether current urbanicity was associated with mental health problems directly and indirectly via biological stress system functioning. Participants (n = 323) were adolescents from the Dutch general population. Measures included home and laboratory assessments of autonomic nervous system and hypothalamic-pituitary-adrenal axis functioning, neighborhood-level urbanicity and socioeconomic status, and mother- and adolescent self-reported mental health problems. Structural equation models showed that urbanicity was not associated with mental health problems directly. Urbanicity was associated with acute autonomic nervous system and hypothalamic-pituitary-adrenal axis reactivity such that adolescents who lived in more urban areas showed blunted biological stress reactivity. Furthermore, there was some evidence for an indirect effect of urbanicity on mother-reported behavioral problems via acute autonomic nervous system reactivity. Urbanicity was not associated with overall autonomic nervous system and hypothalamic-pituitary-adrenal axis reactivity or basal hypothalamic-pituitary-adrenal axis functioning. Although we observed some evidence for associations between urbanicity, biological stress reactivity and mental health problems, most of the tested associations were not statistically significant. Measures of long-term biological stress system functioning may be more relevant to the study of broader environmental factors such as urbanicity.
F.3 Fontanella, Hiance-Steelesmith, Phillips, Bridge, Lester, Sweeney, & Campo, (2015). Widening Rural-Urban Disparities in Youth Suicides, United States, 1996-2010. JAMA Pediatrics.
ABSTRACT: IMPORTANCE: Little is known about recent trends in rural-urban disparities in youth suicide, particularly sex- and method-specific changes. Documenting the extent of these disparities is critical for the development of policies and programs aimed at eliminating geographic disparities.
OBJECTIVE: To examine trends in US suicide mortality for adolescents and young adults across the rural-urban continuum.
DESIGN: Longitudinal trends in suicide rates by rural and urban areas between January 1, 1996, and December 31, 2010, were analyzed using county-level national mortality data linked to a rural-urban continuum measure that classified all 3141 counties in the United States into distinct groups based on population size and adjacency to metropolitan areas. The population included all suicide decedents aged 10 to 24 years.
MEASURES: Rates of suicide per 100 000 persons.
RESULTS: Across the study period, 66 595 youths died by suicide, and rural suicide rates were nearly double those of urban areas for both males (19.93 and 10.31 per 100 000, respectively) and females (4.40 and 2.39 per 100 000, respectively). Even after controlling for a wide array of county-level variables, rural-urban suicide differentials increased over time for males, suggesting widening rural-urban disparities (1996-1998: adjusted incidence rate ratio [IRR], 0.98; 2008-2010: adjusted IRR, 1.19; difference in IRR, P = .02). Firearm suicide rates declined, and the rates of hanging/suffocation for both males and females increased. However, the rates of suicide by firearm (males: 1996-1998, 2.05; and 2008-2010: 2.69 times higher) and hanging/suffocation (males: 1996-1998, 1.24; and 2008-2010: 1.63 times higher) were disproportionately higher in rural areas, and rural-urban differences increased over time (P = .002 for males; P = .06 for females).
CONCLUSIONS: Suicide rates for adolescents and young adults are higher in rural than in urban communities regardless of the method used, and rural-urban disparities appear to be increasing over time. Further research should carefully explore the mechanisms whereby rural residence might increase suicide risk in youth and consider suicide-prevention efforts specific to rural settings.
F.4 Chinni (2021). Unpacking the Geography of America’s Youth Suicide Epidemic. American Communities Project.
FIGURES:
APPENDIX G: VARIATION BY SOCIOECONOMIC STATUS
G.1 National Survey on Drug Use and Health: Cases of Major Depressive Disorders by Social Class (2004-2022). Substance Abuse and Mental Health Services Administration Center for Behavioral Health Statistics and Quality.
Figure created by Zach Rausch. See spreadsheet.
NOTE: Poverty Level: Three Categories for Poverty Level Defined Relative to the Poverty Threshold: Less Than 100 Percent, 100 to 199 Percent, 200 Percent or More. Values: “Had at Least One Major Depressive Episode (MDE) in Past Year among Persons Aged 12 to 17, by Demographic Characteristics: Percentages.”
G.2 Rivenbark… & Odgers (2020). Adolescents’ perceptions of family social status correlate with health and life chances: A twin difference longitudinal cohort study. Proceedings of the National Academy of Sciences.
ABSTRACT: Children from lower-income households are at increased risk for poor health, educational failure, and behavioral problems. This social gradient is one of the most reproduced findings in health and social science. How people view their position in social hierarchies also signals poor health. However, when adolescents’ views of their social position begin to independently relate to well-being is currently unknown. A cotwin design was leveraged to test whether adolescents with identical family backgrounds, but who viewed their family’s social status as higher than their same-aged and sex sibling, experienced better well-being in early and late adolescence. Participants were members of the Environmental Risk Longitudinal Twin Study, a representative cohort of British twins (n = 2,232) followed across the first 2 decades of life. By late adolescence, perceptions of subjective family social status (SFSS) robustly correlated with multiple indicators of health and well-being, including depression; anxiety; conduct problems; marijuana use; optimism; not in education, employment, or training (NEET) status; and crime. Findings held controlling for objective socioeconomic status both statistically and by cotwin design after accounting for measures of childhood intelligence (IQ), negative affect, and prior mental health risk and when self-report, informant report, and administrative data were used. Little support was found for the biological embedding of adolescents’ perceptions of familial social status as indexed by inflammatory biomarkers or cognitive tests in late adolescence or for SFSS in early adolescence as a robust correlate of well-being or predictor of future problems. Future experimental studies are required to test whether altering adolescents’ subjective social status will lead to improved well-being and social mobility.
G.3 Rivenbark… & Odgers (2019). Perceived social status and mental health among young adolescents: Evidence from census data to cellphones. Developmental Psychology.
ABSTRACT: Adolescents in the United States live amidst high levels of concentrated poverty and increasing income inequality. Poverty is robustly linked to adolescents’ mental health problems; however, less is known about how perceptions of their social status and exposure to local area income inequality relate to mental health.
Participants comprised a population representative sample of over 2,100 young adolescents (aged 10–16), 395 of whom completed a 14-day ecological momentary assessment (EMA) study. Participants’ subjective social status (SSS) was assessed using a ladder measure at the start of the EMA, and mental health symptoms were measured both at baseline for the entire sample and daily in the EMA sample.
Adolescents’ SSS tracked family, school and neighborhood economic indicators (|r| ranging from 0.12 to 0.30), and associations did not differ by age, race, or gender. SSS was independently associated with mental health, with stronger associations among older (aged 14 to 16) versus younger (aged 10 to 13) adolescents. Adolescents with lower SSS reported higher psychological distress and inattention problems, as well as more conduct problems in daily life. Those living in areas with higher income inequality reported significantly lower subjective social status, but this association was explained by family and neighborhood income. Findings illustrate that adolescents’ SSS is correlated with both internalizing and internalizing mental health problems, and by age 14 becomes a unique predictor of mental health problems.
G.6 Zou, Xu, Hong, & Yuan (2020). Higher Socioeconomic Status Predicts Less Risk of Depression in Adolescence: Serial Mediating Roles of Social Support and Optimism. Frontiers in Psychology.
ABSTRACT: Family socioeconomic status (SES) is known to have a powerful influence on adolescent depression. However, the mechanisms underlying this association are unclear. Here, we explore this issue by testing the potential mediating roles of social support (interpersonal resource) and optimism (intrapersonal resource), based on the predictions of the reserve capacity model (RCM). Participants were 652 adolescents [age range: 11–20 years old, Mage = 14.55 years, SD = 1.82; 338 boys (51.80%)] from two junior and two senior high schools in Wuhan, China. They completed questionnaires measuring family SES, perceived social support, optimism, and depression. Results showed, as predicted, (1) SES negatively predicted adolescent depression; (2) social support and optimism serially mediated the relations between SES and depression, consistent with the predictions by the RCM. Specifically, higher SES predicted greater social support and increased optimism, which in turn contributed to reduced depression. The implications of these data to the prevention and interventions of adolescent depression were discussed.
G.4 Reiss… & Ravens-Sieberer (2019). Socioeconomic status, stressful life situations and mental health problems in children and adolescents: Results of the German BELLA cohort-study. PLoS ONE.
ABSTRACT: AIM: Children and adolescents with low socioeconomic status (SES) suffer from mental health problems more often than their peers with high SES. The aim of the current study was to investigate the direct and interactive association between commonly used indicators of SES and the exposure to stressful life situations in relation to children’s mental health problems.
METHODS: The prospective BELLA cohort study is the mental health module of the representative, population-based German National Health Interview and Examination Survey for children and adolescents (KiGGS). Sample data include 2,111 participants (aged 7–17 years at baseline) from the first three measurement points (2003–2006, 2004–2007 and 2005–2008). Hierarchical multiple linear regression models were conducted to analyze associations among the SES indicators household income, parental education and parental unemployment (assessed at baseline), number of stressful life situations (e.g., parental accident, mental illness or severe financial crises; 1- and 2-year follow-ups) and parent-reported mental health problems (Strength and Difficulties Questionnaire; 2-year follow-up).
RESULTS: All indicators of SES separately predicted mental health problems in children and adolescents at the 2-year follow-up. Stressful life situations (between baseline and 2-year follow-up) and the interaction of parental education and the number of stressful life situations remained significant in predicting children’s mental health problems after adjustment for control variables. Thereby, children with higher educated parents showed fewer mental health problems in a stressful life situation. No moderating effect was found for household income and parental employment. Overall, the detected effect sizes were small. Mental health problems at baseline were the best predictor for mental health problems two years later.
CONCLUSIONS: Children and adolescents with a low SES suffer from multiple stressful life situations and are exposed to a higher risk of developing mental health problems. The findings suggest that the reduction of socioeconomic inequalities and interventions for families with low parental education might help to reduce children’s mental health problems.
APPENDIX H: VARIATION BY RELIGIOSITY
H.1 Monitoring the Future study. The degree to which U.S. 12th graders—split by gender and religiosity—agree with the following statements: I feel I do not have much to be proud of, Sometimes I think I am no good at all, I feel that I can't do anything right, and I feel that my life is not very useful.
Figure. Self-derogation trends by sex and religiosity, averaging four items from the Monitoring the Future Study. The scale runs from 1 (strongly disagree with each statement) to 5 (strongly agree). Source: Monitoring the Future, 2-year buckets. (see Zach’s spreadsheet) Note that the sample sizes were smaller than usual in 2020 and 2021 (and were during COVID).
NOTE: We operationalized religiosity by splitting respondents on the basis of how they answer the following questions: “How important is religion in your life?” and “How often do you attend religious serivces” Those who responded with “not important” or “a little important”, and responded saying they attend services “rarely” or “never” were considered not religious. Those who responded “pretty important” or “very important” and attended services at least once a month were considered religious.
H. 2 Lawrence, Oquendo, Stanley (2020). Religion and Suicide Risk: A Systematic Review. Archive of Suicide Research.
ABSTRACT: Although religion is reported to be protective against suicide, the empirical evidence is inconsistent. Research is complicated by the fact that there are many dimensions to religion (affiliation, participation, doctrine) and suicide (ideation, attempt, completion). We systematically reviewed the literature on religion and suicide over the last ten years (89 articles) with a goal of identifying what specific dimensions of religion are associated with specific aspects of suicide. We found that religious affiliation does not necessarily protect against suicidal ideation, but does protect against suicide attempts. Whether religious affiliation protects against suicide attempts may depend on the culture-specific implications of affiliating with a particular religion, since minority religious groups can feel socially isolated. After adjusting for social support measures, religious service attendance is not especially protective against suicidal ideation, but does protect against suicide attempts, and possibly protects against suicide. Future qualitative studies might further clarify these associations.
H. 3 Dervic, Oquendo, Grunebaum, Ellis, Burke and Mann (2004). Religious Affiliation and Suicide Attempt. American Journal of Psychiatry.
ABSTRACT: OBJECTIVE: Few studies have investigated the association between religion and suicide either in terms of Durkheim’s social integration hypothesis or the hypothesis of the regulative benefits of religion. The relationship between religion and suicide attempts has received even less attention.
METHOD: Depressed inpatients (N=371) who reported belonging to one specific religion or described themselves as having no religious affiliation were compared in terms of their demographic and clinical characteristics.
RESULTS: Religiously unaffiliated subjects had significantly more lifetime suicide attempts and more first-degree relatives who committed suicide than subjects who endorsed a religious affiliation. Unaffiliated subjects were younger, less often married, less often had children, and had less contact with family members. Furthermore, subjects with no religious affiliation perceived fewer reasons for living, particularly fewer moral objections to suicide. In terms of clinical characteristics, religiously unaffiliated subjects had more lifetime impulsivity, aggression, and past substance use disorder. No differences in the level of subjective and objective depression, hopelessness, or stressful life events were found.
CONCLUSIONS: Religious affiliation is associated with less suicidal behavior in depressed inpatients. After other factors were controlled, it was found that greater moral objections to suicide and lower aggression level in religiously affiliated subjects may function as protective factors against suicide attempts. Further study about the influence of religious affiliation on aggressive behavior and how moral objections can reduce the probability of acting on suicidal thoughts may offer new therapeutic strategies in suicide prevention.
H 4. O’Reilly and Rosato (2018). Religion and the risk of suicide: longitudinal study of over 1 million people. British Journal of Psychiatry.
ABSTRACT: BACKGROUND: Durkheim's seminal historical study demonstrated that religious affiliation reduces suicide risk, but it is unclear whether this protective effect persists in modern, more secular societies.
AIMS: To examine suicide risk according to Christian religious affiliation and by inference to examine underlying mechanisms for suicide risk. If church attendance is important, risk should be lowest for Roman Catholics and highest for those with no religion; if religiosity is important, then ‘conservative’ Christians should fare best.
METHODS: A 9-year study followed 1 106 104 people aged 16–74 years at the 2001 UK census, using Cox proportional hazards models adjusted for census-based cohort attributes.
RESULTS: In fully adjusted models analysing 1119 cases of suicide, Roman Catholics, Protestants and those professing no religion recorded similar risks. The risk associated with conservative Christians was lower than that for Catholics (HR = 0.71, 95% CI 0.52–0.97).
CONCLUSIONS: The relationship between religious affiliation and suicide established by Durkheim may not pertain in societies where suicide rates are highest at younger ages. Risks are similar for those with and without a religious affiliation, and Catholics (who traditionally are characterised by higher levels of church attendance) do not demonstrate lower risk of suicide. However, religious affiliation is a poor measure of religiosity, except for a small group of conservative Christians, although their lower risk of suicide may be attributable to factors such as lower risk behaviour and alcohol consumption.
H 5. Lawrence, Brent, Mann, Burke, Grunebaum, Galfalvy and Oquendo (2016). Religion as a Risk Factor for Suicide Attempt and Suicide Ideation Among Depressed Patients. Journal of Nervous and Mental Disorders.
ABSTRACT: We aimed to examine the relationship between religion and suicide attempt and ideation. 321 depressed patients were recruited from mood-disorder research studies at the New York State Psychiatric Institute. Participants were interviewed using the SCID, Columbia University Suicide History form, Scale for Suicide Ideation, and Reasons for Living Inventory. Participants were asked about their religious affiliation, importance of religion, and religious service attendance. We found that past suicide attempts were more common among depressed patients with a religious affiliation (OR 2.25, p=.007). Suicide ideation was greater among depressed patients who considered religion more important (Coeff. 1.18, p=.026), and those who attended services more frequently (Coeff. 1.99, p=.001). We conclude that the relationship between religion and suicide risk factors is complex, and can vary among different patient populations. Physicians should seek deeper understanding of the role of religion in an individual patient’s life in order to understand the person’s suicide risk factors more fully.
H 6. Stacks and Laubepin (2019). Religiousness as a Predictor of Suicide: An Analysis of 162 European Regions. The American Association of Suicidology.
ABSTRACT: Research on religion as a protective factor has been marked by four recurrent limitations: (1) an overemphasis on the United States, a nation where religiosity is relatively high; (2) a neglect of highly secularized zones of the world, where religiousness may be too weak to affect suicide; (3) restriction of religiousness to religious affiliation, a construct which may miss capturing other dimensions of religiousness such as the importance of religion in one’s life; and (4) an overwhelming use of the nation as a unit of analysis, which masks variation in religiousness within nations. The present article addresses these limitations by performing a cross-national test of the following hypothesis: The greater the strength of subjective religiousness, the lower the suicide rate, using small units of analysis for a secularized area of the world. All data refer to 162 regions within 22 European nations. Data were extracted from two large databases, EUROSTAT and the European Social Surveys (ESS Round 4), and merged using NUTS-2 (Nomenclature of Statistical Territorial Units) regions as the unit of analysis. Controls are incorporated for level of economic development, education, and measures of economic strain. The results of a multiple regression analysis demonstrated that controlling for the other constructs in the model, religiousness is associated with lower suicide rates, confirming the hypothesis. Even in secularized European nations, where there is a relatively weak moral community to reinforce religion, religiousness acts as a protective factor against suicide. Future work is needed to explore the relationship in other culture zones of the world.
H 7. Pelham and Nyri (2008). More Religious Countries, Lower Suicide Rates: Lower suicide rates not a matter of national income. Gallup Polls.
EXCERPT: Comparing the Religiosity Index scores of different countries with suicide statistics published in 2007 by the World Health Organization reveals a clear pattern: Countries that are more religious tend to have lower suicide rates. For example, whereas the Philippines has one of the world's highest religiosity scores (79), Japan has one of the world's lowest scores (29). Suicide rates in the Philippines are almost 12 times lower than rates in Japan. Paraguayans, who are much more religious than Uruguayans are, also have suicide rates about five times lower than in Uruguay. The United States falls near the middle of the international pack in religiosity, at 61. The United States also falls near the middle of the international pack in suicide rates, having the 32nd highest suicide rate out of 67 countries.
Does religiosity per se truly affect suicide rates? Countries that differ in religiosity often differ in other ways. For example, these same data showed that less religious countries are usually wealthier (e.g., have a higher GDP per capita) than more religious countries. Further, these data showed that suicide rates are also slightly higher in wealthier countries. However, the relation between GDP and suicide is not nearly as strong as the relation between religiosity and suicide. Thus, national wealth cannot explain the connection between religiosity and suicide. Another concern is that countries that are more religious might tend to underreport suicides -- because of subpar medical documentation, or the added social stigma suicide carries in countries that are more religious. However, an analysis focusing only on wealthy countries, where documentation of suicide is likely to be excellent, still reveals a robust association between religiosity and national suicide rates.
Do these country-level findings translate into the behavior of individual people? Large-scale studies of individual respondents suggest so. In 2002, statistician Sterling Hilton and colleagues showed that among young men who were actively involved in the Mormon church, suicide rates were three to five times lower than those of either non-members or less active church members. Moreover, in a review of 42 studies of religiosity and mortality, Michael McCullough and associates showed that, compared with less religious people, highly religious people are slightly less prone to mortality from several specific causes, including suicide. Finally, recent respondent level data from other Gallup Polls show that religious people are much less likely than the general public to believe that suicide is "morally acceptable." Perhaps the most extreme example of this comes from France, where 40% of the general population but only 4% of Muslims living in Paris consider suicide morally acceptable.
H8. Tsomokos and Dunbar (2023). The role of religion in adolescent mental health: faith as a
moderator of the relationship between distrust and depression. Religion, Brain & Behavior
It has recently been shown that interpersonal distrust predicts depressive symptoms in middle adolescence, and this finding has been interpreted in light of Social Safety Theory, which views distrust as an index of social threat. Here we hypothesize that religiousness provides social safety and may counteract the sense of social threat indexed by distrust. Religiousness should therefore act as a moderator between interpersonal distrust and depression. Using a nationally representative birth cohort from the UK, we provide evidence in favor of this hypothesis, even after controlling for stratum disadvantage and socioeconomic characteristics, sex, ethnicity, and multiple confounders on the level of the individual (BMI, chronic illness, cognitive ability, risk-taking, experiencing bullying, dietary habits, chronotype, physical activity and screen time), family context (frequency of eating meals together, maternal mental health), and neighborhood ecology (NO2 levels of air pollution).
APPENDIX I: WHAT HAPPENED DURING COVID?
I.1 CDC (2022, March 31). Adolescent Behaviors and Experiences Survey. Centers for Disease Control and Prevention.
KEY FINDINGS:
- More than 1 in 3 high school students experienced poor mental health during the pandemic and nearly half of students felt persistently sad or hopeless.
- Female students and those who identify as lesbian, gay, bisexual, other or questioning (LGBQ) are experiencing disproportionate levels of poor mental health and suicide-related behaviors. For example, in 2021, 12% of female students, more than 25% of LGB students, and 17% of other or questioning students attempted suicide during the past year compared to 5% of their male peers and 5% of their heterosexual peers, respectively.
Daily Life Was Disrupted
- The range of impacts on youth’s daily lives was broad – including difficulties, family economic impacts, hunger, and abuse in the home.
- More than half of students experienced emotional abuse in the home and more than 10% reported physical abuse in the home.
- Lesbian, gay, and bisexual students were far more likely to report physical abuse, with 20% reporting that they had been physically abused by a parent or other adult in their home, compared to 10% of heterosexual students.
- Black students were most likely to report hunger, with nearly a third reporting that there was not enough food in their home during the pandemic.
Racism Is a Public Health Problem
- More than one third of all U.S. high school students felt they had been treated badly or unfairly at school because of their race or ethnicity.
- Asian, Black, and Multiracial students reported the highest levels of experiencing racism.
- Students who reported racism were also more likely to experience poor mental health and less likely to feel connected to people at school.
I.2 The U.S. Surgeon General’s Advisory (2021). Protecting youth mental health.
See excerpts from the advisory in section 1.1.10.
I.3 Lebrun-Harris, Ghandour, Kogan, & Warren (2022). Five-Year Trends in US Children’s Health and Well-being, 2016-2020. JAMA Pediatrics.
ABSTRACT: IMPORTANCE: Ensuring the well-being of the 73 million children in the United States is critical for improving the nation’s health and influencing children’s long-term outcomes as they grow into adults.
OBJECTIVE: To examine recent trends in children’s health-related measures, including significant changes between 2019 and 2020 that might be attributed to the COVID-19 pandemic.
DESIGN: Annual data were examined from the National Survey of Children’s Health (2016-2020), a population-based, nationally representative survey of randomly selected children. Participants were children from birth to age 17 years living in noninstitution settings in all 50 states and the District of Columbia whose parent or caregiver responded to an address-based survey by mail or web. Weighted prevalence estimates account for probability of selection and nonresponse. Adjusted logistic regression models tested for significant trends over time.
OUTCOMES AND MEASURES: Diverse measures pertaining to children’s current health conditions, positive health behaviors, health care access and utilization, and family well-being and stressors.
RESULTS: A total of 174 551 children were included (annual range = 21 599 to 50 212). Between 2016 and 2020, there were increases in anxiety (7.1% [95% CI, 6.6-7.6] to 9.2% [95% CI, 8.6-9.8]; +29%; trend P < .001) and depression (3.1% [95% CI, 2.9-3.5] to 4.0% [95% CI, 3.6-4.5]; +27%; trend P < .001). There were also decreases in daily physical activity (24.2% [95% CI, 23.1-25.3] to 19.8% [95% CI, 18.9-20.8]; −18%; trend P < .001), parent or caregiver mental health (69.8% [95% CI, 68.9-70.8] to 66.3% [95% CI, 65.3-67.3]; −5%; trend P < .001), and coping with parenting demands (67.2% [95% CI, 66.3-68.1] to 59.9% [95% CI, 58.8-60.9]; −11%; trend P < .001). In addition, from 2019 to 2020, there were increases in behavior or conduct problems (6.7% [95% CI, 6.1-7.4] to 8.1% [95% CI, 7.5-8.8]; +21%; P = .001) and child care disruptions affecting parental employment (9.4% [95% CI, 8.0-10.9] to 12.6% [95% CI, 11.2-14.1]; +34%; trend P = .001) as well as decreases in preventive medical visits (81.0% [95% CI, 79.7-82.3] to 74.1% [95% CI, 72.9-75.3]; −9%; trend P < .001).
CONCLUSIONS: Recent trends point to several areas of concern that can inform future research, clinical care, policy decision making, and programmatic investments to improve the health and well-being of children and their families. More analyses are needed to elucidate varying patterns within subpopulations of interest.
I.4 Jones (2022). Mental health, suicidality, and connectedness among high school students during the COVID-19 pandemic—Adolescent Behaviors and Experiences Survey, United States, January–June 2021. MMWR Supplements.
ABSTRACT: Disruptions and consequences related to the COVID-19 pandemic, including school closures, social isolation, family economic hardship, family loss or illness, and reduced access to health care, raise concerns about their effects on the mental health and well-being of youths. This report uses data from the 2021 Adolescent Behaviors and Experiences Survey, an online survey of a probability-based, nationally representative sample of U.S. public- and private-school students in grades 9–12 (N = 7,705), to assess U.S. high school students’ mental health and suicidality during the COVID-19 pandemic. The study also examines whether mental health and suicidality are associated with feeling close to persons at school and being virtually connected to others during the pandemic. Overall, 37.1% of students experienced poor mental health during the pandemic, and 31.1% experienced poor mental health during the preceding 30 days. In addition, during the 12 months before the survey, 44.2% experienced persistent feelings of sadness or hopelessness, 19.9% had seriously considered attempting suicide, and 9.0% had attempted suicide. Compared with those who did not feel close to persons at school, students who felt close to persons at school had a significantly lower prevalence of poor mental health during the pandemic (28.4% versus 45.2%) and during the past 30 days (23.5% versus 37.8%), persistent feelings of sadness or hopelessness (35.4% versus 52.9%), having seriously considered attempting suicide (14.0% versus 25.6%), and having attempted suicide (5.8% versus 11.9%). The same pattern was observed among students who were virtually connected to others during the pandemic (i.e., with family, friends, or other groups by using a computer, telephone, or other device) versus those who were not. Comprehensive strategies that improve feelings of connectedness with others in the family, in the community, and at school might foster improved mental health among youths during and after the COVID-19 pandemic.
I.5 Ravens-Sieberer… & Hölling (2021). Quality of life and mental health in children and adolescents during the first year of the COVID-19 pandemic: Results of a two-wave nationwide population-based study. European Child & Adolescent Psychiatry.
ABSTRACT: BACKGROUND: The COVID-19 pandemic has disrupted the lives of children and adolescents worldwide. The German COPSY study is among the first population-based longitudinal studies to examine the mental health impact of the pandemic. The objective of the study was to assess changes in health-related quality of life (HRQoL) and mental health in children and adolescents and to identify the associated risk and resource factors during the pandemic.
METHODS: A nationwide longitudinal survey was conducted with two waves during the pandemic (May/June 2020 and December 2020/January 2021). In total, n = 1923 children and adolescents aged 7 to 17 years and their parents participated (retention rate from wave 1 to wave 2: 85%). The self-report and parent-proxy surveys assessed HRQoL (KIDSCREEN-10), mental health problems (SDQ with the subscales emotional problems, conduct problems, hyperactivity, and peer problems), anxiety (SCARED), depressive symptoms (CES-DC, PHQ-2) and psychosomatic complaints (HBSC-SCL). Mixed model panel regression analyses were conducted to examine longitudinal changes in mental health and to identify risk and resource factors.
RESULTS: The HRQoL of children and adolescents decreased during the pandemic, and emotional problems, peer-related mental health problems, anxiety, depressive and psychosomatic symptoms increased over time, however the change in global mental health problems from wave 1 to wave 2 was not significant, and some changes were negligible. Socially disadvantaged children and children of mentally burdened parents were at particular risk of impaired mental health, while female gender and older age were associated with fewer mental health problems. A positive family climate and social support supported the mental health of children and adolescents during the pandemic.
DISCUSSION: Health promotion, prevention and intervention strategies could support children and adolescents in coping with the pandemic and protect and maintain their mental health.
I.6 Racine, McArthur, Cooke, Eirich, Zhu, & Madigan (2021). Global Prevalence of Depressive and Anxiety Symptoms in Children and Adolescents During COVID-19: A Meta-analysis. JAMA Pediatrics.
ABSTRACT: IMPORTANCE: Emerging research suggests that the global prevalence of child and adolescent mental illness has increased considerably during COVID-19. However, substantial variability in prevalence rates have been reported across the literature.
OBJECTIVE: To ascertain more precise estimates of the global prevalence of child and adolescent clinically elevated depression and anxiety symptoms during COVID-19; to compare these rates with prepandemic estimates; and to examine whether demographic (eg, age, sex), geographical (ie, global region), or methodological (eg, pandemic data collection time point, informant of mental illness, study quality) factors explained variation in prevalence rates across studies.
DATA SOURCES: Four databases were searched (PsycInfo, Embase, MEDLINE, and Cochrane Central Register of Controlled Trials) from January 1, 2020, to February 16, 2021, and unpublished studies were searched in PsycArXiv on March 8, 2021, for studies reporting on child/adolescent depression and anxiety symptoms. The search strategy combined search terms from 3 themes: (1) mental illness (including depression and anxiety), (2) COVID-19, and (3) children and adolescents (age ≤18 years). For PsycArXiv, the key terms COVID-19, mental health, and child/adolescent were used.
STUDY SELECTION: Studies were included if they were published in English, had quantitative data, and reported prevalence of clinically elevated depression or anxiety in youth (age ≤18 years).
DATA EXTRACTION AND SYNTHESIS: A total of 3094 nonduplicate titles/abstracts were retrieved, and 136 full-text articles were reviewed. Data were analyzed from March 8 to 22, 2021.
MAIN OUTCOMES AND MEASURES: Prevalence rates of clinically elevated depression and anxiety symptoms in youth.
RESULTS: Random-effect meta-analyses were conducted. Twenty-nine studies including 80 879 participants met full inclusion criteria. Pooled prevalence estimates of clinically elevated depression and anxiety symptoms were 25.2% (95% CI, 21.2%-29.7%) and 20.5% (95% CI, 17.2%-24.4%), respectively. Moderator analyses revealed that the prevalence of clinically elevated depression and anxiety symptoms were higher in studies collected later in the pandemic and in girls. Depression symptoms were higher in older children.
CONCLUSIONS: Pooled estimates obtained in the first year of the COVID-19 pandemic suggest that 1 in 4 youth globally are experiencing clinically elevated depression symptoms, while 1 in 5 youth are experiencing clinically elevated anxiety symptoms. These pooled estimates, which increased over time, are double of prepandemic estimates. An influx of mental health care utilization is expected, and allocation of resources to address child and adolescent mental health concerns are essential.
I.7 Bouter… & Grootendorst-van Mil (2022). A longitudinal study of mental health in at-risk adolescents before and during the COVID-19 pandemic. European Child & Adolescent Psychiatry.
ABSTRACT: Although cross-sectional studies have shown that the COVID-19 pandemic has negatively affected the mental health of adolescents, the effect of the pandemic on adolescents with pre-pandemic symptoms is unclear. We, therefore, tested the hypothesis that adolescents had increased emotional and behavioral problems during the lockdowns imposed during the pandemic.This study included three measurements in a prospective cohort of 1022 adolescents who were oversampled based on their high risk of developing psychopathology. Before the pandemic, we assessed depressive, anxiety, stress, oppositional defiant problems, psychotic experiences and suicidality, using the Youth Self-Report; 445 and 333 of these 1,022 adolescents subsequently completed the online questionnaire in the first lockdown (in April 2020) and in the second lockdown (in January 2021), respectively. Multilevel random intercept regression models were used to determine the change in psychiatric symptoms, including an interaction term to assess whether these changes differed based on the severity of symptoms prior to the pandemic. Throughout the pandemic, the majority of the participating adolescents reported having emotional and behavioral symptoms that were within the normal range. Moreover, the mean symptom scores for all six outcomes decreased significantly among adolescents with high clinical severity prior to the pandemic. In contrast to our original hypothesis, the effects of the COVID-19 pandemic may not necessarily be detrimental, at least among a specific subgroup of adolescents with pre-existing mental health problems. Moreover, our finding that most adolescents in this at-risk sample did not report experiencing clinically relevant symptoms during the pandemic reflects their resilience during the pandemic.
FIGURE:
Fig. Summary of the estimated scores for anxiety problems, depressive problems, oppositional defiant problems, stress problems, psychotic experiences, and suicidality reported at baseline, C1, and C2, stratified by emotional and behavioral problem scores measured before the COVID-19 pandemic. *p < 0.05 and **p < 0.001 (post hoc univariate test)
I.8 Ludwig-Walz, Dannheim, Pfadenhauer, Fegert, & Bujard (2022). Increase of depression among children and adolescents after the onset of the COVID-19 pandemic in Europe: A systematic review and meta-analysis. Child and Adolescent Psychiatry and Mental Health.
ABSTRACT: BACKGROUND: Research points to a high depression burden among youth during the COVID-19 pandemic; however, a lack of systematic evidence exists. We determine the change in depression symptoms among children and adolescents during COVID-19 compared to pre-pandemic baselines. By using country differences in pandemic-related restrictions and school closures in Europe as quasi-experimental design, we evaluate policy impacts on depression.
METHODS: In this systematic review and meta-analysis, following the PRISMA statement, we searched six databases (MEDLINE, EMBASE, PsycINFO, Cochrane Central, Web of Science, WHO COVID-19) using a peer-reviewed search string up until March 18, 2022 with citation tracking and grey literature searches. No limitations regarding language and effect measures existed. We included studies that compared (1) general depression symptoms or (2) clinically relevant depression rates in children and adolescents (≤ 19 years) before and during the COVID-19 pandemic in Europe. The validated Oxford Stringency Index was used as indicator for pandemic-related restrictions. Screening for eligibility, extracting data from published reports and from unpublished data requested directly from study authors, assessing the study risk of bias and grading certainty of evidence using the GRADE approach, were all done in duplicate. Data were pooled in a random-effects model.
RESULTS: Of 7,422 nonduplicate records, 22 studies with data from 868,634 participants pre-pandemic and 807,480 during pandemic, met full inclusion criteria. For the comparison of depression symptoms before and during the COVID-19 pandemic, moderate certainty of evidence was observed for general depression symptoms (standardized mean difference, 0.21 [95%CI, 0.12-0.30]; I2 = 94%) and low certainty of evidence for clinically relevant depression rates (odds ratio, 1.36 [95%CI, 1.05-1.76]; I2 = 95%) for total population. Increase in general depression symptoms was higher for male adolescents, whereas increase in clinically relevant depression rates was higher for females. Effect estimates were significantly higher when pandemic-related restrictions were more stringent or school closure occurred.
CONCLUSION: An increase in depression symptoms occurred in a pre-pandemic vs. during-pandemic comparison within the COVID-19 pandemic, whereby pandemic-related restrictions (such as school closures) resulted in a considerable effect increase. Ensuring adequate supply of mental health recovery services and long-term monitoring is of high public health relevance.
I.9 Madigan, Korczak, Vaillancourt, Racine, Hopkins, Pador, Hewitt, AlMousawi, McDonald, & Neville (2023). Comparison of paediatric emergency department visits for attempted suicide, self-harm, and suicidal ideation before and during the COVID-19 pandemic: A systematic review and meta-analysis. The Lancet Psychiatry.
ABSTRACT: BACKGROUND: There is a lack of consensus about the effect of the COVID-19 pandemic on the mental health of children and adolescents. We aimed to compare rates of paediatric emergency department visits for attempted suicide, self-harm, and suicidal ideation during the pandemic with those before the pandemic.
METHODS: For this systematic review and meta-analysis, we searched MEDLINE, Embase, and PsycINFO for studies published between Jan 1, 2020, and Dec 19, 2022. Studies published in English with data on paediatric (ie, those aged <19 years) emergency department visits before and during the COVID-19 pandemic were included. Case studies and qualitative analyses were excluded. Changes in attempted suicide, self-harm, suicidal ideation, and other mental-illness indicators (eg, anxiety, depression, and psychosis) were expressed as ratios of the rates of emergency department visits during the pandemic compared with those before the pandemic, and we analysed these with a random-effects meta-analysis. This study was registered with PROSPERO, CRD42022341897.
FINDINGS: 10,360 non-duplicate records were retrieved, which yielded 42 relevant studies (with 130 sample-estimates) representing 11·1 million emergency department visits for all indications of children and adolescents across 18 countries. The mean age of the samples of children and adolescents across studies was 11·7 years (SD 3·1, range 5·5–16·3), and there were on average 57·6% girls and 43·4% boys as a proportion of emergency department visits for any health reasons (ie, physical and mental). Only one study had data related to race or ethnicity. There was good evidence of an increase in emergency department visits for attempted suicide during the pandemic (rate ratio 1·22, 90% CI 1·08–1·37), modest evidence of an increase in emergency department visits for suicidal ideation (1·08, 0·93–1·25), and good evidence for only a slight change in self-harm (0·96, 0·89–1·04). Rates of emergency department visits for other mental-illness indications showed very good evidence of a decline (0·81, 0·74–0·89), and paediatric visits for all health indications showed strong evidence of a reduction (0·68, 0·62–0·75). When rates for attempted suicide and suicidal ideation were combined as a single measure, there was good evidence of an increase in emergency department visits among girls (1·39, 1·04–1·88) and only modest evidence of an increase among boys (1·06, 0·92–1·24). Self-harm among older children (mean age 16·3 years, range 13·0–16·3) showed good evidence of an increase (1·18, 1·00–1·39), but among younger children (mean age 9·0 years, range 5·5–12·0) there was modest evidence of a decrease (0·85, 0·70–1·05).
INTERPRETATION: The integration of mental health support within community health and the education system—including promotion, prevention, early intervention, and treatment—is urgently needed to increase the reach of mental health support that can mitigate child and adolescent mental distress. In future pandemics, increased resourcing in some emergency department settings would help to address their expected increase in visits for acute mental distress among children and adolescents.
* * * * * * * * * * * * * * *
APPENDIX J: STUDIES ON PERFECTIONISM
J.1 Curran & Hill (2019). Perfectionism is increasing over time: A meta-analysis of birth cohort differences from 1989 to 2016. Psychological Bulletin.
ABSTRACT: From the 1980s onward, neoliberal governance in the United States, Canada, and the United Kingdom has emphasized competitive individualism and people have seemingly responded, in kind, by agitating to perfect themselves and their lifestyles. In this study, the authors examine whether cultural changes have coincided with an increase in multidimensional perfectionism in college students over the last 27 years. Their analyses are based on 164 samples and 41,641 American, Canadian, and British college students, who completed the Multidimensional Perfectionism Scale (Hewitt & Flett, 1991) between 1989 and 2016 (70.92% female, Mage = 20.66). Cross-temporal meta-analysis revealed that levels of self-oriented perfectionism, socially prescribed perfectionism, and other-oriented perfectionism have linearly increased. These trends remained when controlling for gender and between-country differences in perfectionism scores. Overall, in order of magnitude of the observed increase, the findings indicate that recent generations of young people perceive that others are more demanding of them, are more demanding of others, and are more demanding of themselves.
[Note that this article does not directly implicate social media. It is included here because perfectionism is often said to be one of the effects that Instagram has on girls, and it is interesting that it is the “socially prescribed perfectionism” that has increased most rapidly in recent years. We should try to graph trends for boys and girls separately]
ADDITIONAL EXCERPTS: “Perhaps the most important finding from this research is that more recent generations of college students are reporting higher levels of socially prescribed perfectionism than previous generations. This finding suggests that young people are perceiving that their social context is increasingly demanding, that others judge them more harshly, and that they are increasingly inclined to display perfection as a means of securing approval. We highlight the salience of this finding because of the size of the comparative increase, twice that of the other two dimensions, and the larger association between socially prescribed perfectionism and psychopathology (Limburg, Watson, Hagger, & Egan, 2017). Rising socially prescribed perfectionism dovetails with observations of rising externality of control, anxiety, and neurosis among young people, in addition to a rising sense of social disconnection (e.g., Paik & Sanchagrin, 2013; Twenge, 2000; Twenge, Zhang, & Im, 2004). These are worrying trends and suggest that young people may be increasingly more sensitive to perceived external pressures and are finding it more difficult than previous generations to cope with them.
…Socially prescribed perfectionism is the most debilitating of the three dimensions of perfectionism. This is because the perceived expectations of others are experienced as excessive, uncontrollable, and unfair, making failure experiences and negative emotional states common (Hewitt & Flett, 1991). The debilitating nature of socially prescribed perfectionism is evident in research on college students, which has found this dimension of perfectionism to be positively associated with major psychopathology (e.g., anxiety, depressive symptoms, and suicide ideation; Martin, Flett, Hewitt, Krames, & Szanto, 1996; Hewitt, Flett, & Weber, 1994; Sherry, Hewitt, Flett, & Harvey, 2003). These relationships have been replicated in longitudinal and experimental studies (e.g., Flett, Endler, Tassone, & Hewitt, 1995; Hewitt, Flett, & Ediger, 1995; O’Connor, O’Connor, O’Connor, Smallwood, & Miles, 2004). Like self-oriented perfectionism, the reviews of Smith et al. (2016, 2017) showed that socially prescribed perfectionism predicted increases in depressive symptoms and suicide ideation over time, but to a much greater degree.”
FIGURE:
J.2 Curran & Hill (2022). Young people’s perceptions of their parents’ expectations and criticism are increasing over time: Implications for perfectionism. Psychological Bulletin.
ABSTRACT: Recent evidence demonstrates rising self-oriented, other-oriented, and socially prescribed perfectionism among young people from the United States, United Kingdom, and Canada (Curran & Hill, 2019). One reason why perfectionism is increasing may be that rising competitiveness and individualism are requiring parents to engage in anxious, overly involved, and/or overly controlling forms of parenting. Yet, data to support this claim are limited and contested. In two meta-analyses, we expanded upon and tested this claim by examining whether excessive parental expectations and harsh parental criticism are correlated with perfectionism (Study 1) and whether these perceived practices are changing over time among American, Canadian, and British college students (Study 2). In Study 1, meta-analyses found small-to-moderate positive mean weighted effects of parental expectations and parental criticism on self-oriented and other-oriented perfectionism, and large positive mean weighted effects of parental expectations and parental criticism on socially prescribed perfectionism. In Study 2, using cross-temporal meta-analysis, we found that mean levels of parental expectations and parental criticism had linearly increased between 1989 and 2019 among college students. With rising competitiveness, individualism, economic inequality, and pressure to excel at school and college as the societal background, increases in parental expectations and parental criticism offer the most plausible explanation for rising perfectionism to date.
IMPACT STATEMENT: Public Significance Statement Two meta-analyses support a possible explanation for rising perfectionism among young people: changing parenting practices. Parental expectations and criticism were positively correlated with perfectionism in a first meta-analysis (Study 1), and these practices were found to be increasing over time in a second meta-analysis (Study 2). The latter finding has special public significance. Rising expectations and criticism likely reflect one response parents are making to escalating societal competitiveness, individualism, inequality, and pressures to excel at school and college.
J.3 Sand, Bøe, Shafran, Stormark, & Hysing (2021). Perfectionism in Adolescence: Associations With Gender, Age, and Socioeconomic Status in a Norwegian Sample. Frontiers in Public Health.
ABSTRACT: BACKGROUND: Perfectionism in adolescence has received increased attention, but few studies have examined this in non-clinical samples. This study investigated perfectionism among adolescents from the general population in relation to demographic factors.
METHODS: The present study is cross-sectional and draws on the epidemiological study. The sample consisted of 10.217 adolescents aged 16–19 years (52.9% girls). Self-reported perfectionism was assessed by the EDI-P scale from the Eating Disorder Inventory with two dimensions of perfectionism, namely self-oriented (SOP) and socially prescribed (SPP) perfectionism, and a total score. Perfectionism was analyzed in relation to age, gender, and socioeconomic status (SES) by perceived economic well-being and parental education level. Chi-squared tests, t-tests, and regression analyses were performed.
RESULTS: There were few gender differences on the mean scores on perfectionism, with similar levels on the total score of EDI-P and SOP, while girls scored slightly higher on SPP (p < 0.001). The latter gender difference represented a small effect size (Cohen's d = 0.053). Chi-square analyses with perfectionism split at the 90th percentile across gender showed that there were significantly more girls than boys among the high scorers both for EDI-P, EDI-SOP, and EDI-SPP. There were no significant differences between levels of perfectionism between the three age groups. The logistic regression analyses adjusted by age and gender showed that adolescents with a better perceived economic well-being had increased odds of high perfectionism. This was evident for overall EDI-P (OR = 1.760, 95% CI = 1.493–2.076), SOP (OR = 1.543, 95% CI = 1.292–1.843), and SPP (OR = 1.836, 95% CI = 1.559–2.163). Parental education was not significantly associated with perfectionism scores among the adolescents.
CONCLUSIONS: The levels of perfectionism were relatively similar between the genders in the present study, besides slightly higher SPP among girls than boys. There were also significantly more girls than boys among the high scorers on overall perfectionism, SOP, and SPP, respectively. High perfectionism was related to SES for perceived economic well-being, but not for parental education level. Implications for further research and clinical interventions were suggested.
J.4 Smith, Sherry, Vidovic, Saklofske, Stoeber, & Benoit (2019). Perfectionism and the Five-Factor Model of Personality: A Meta-Analytic Review. Personality and Social Psychology Review.
EXCERPT: The moderating effect of gender on perfectionistic strivings and neuroticism, other-oriented perfectionism and neuroticism, discrepancy and conscientiousness, socially prescribed perfectionism and agreeableness, and self-oriented perfectionism and agreeableness remained significant (p < .05) after controlling for age, year of data collection, perfectionism subscale, and FFM versus non-FFM measure. However, inspection of funnel plots suggested that the moderating effect of gender on the other-oriented perfectionism–neuroticism link and the discrepancy–conscientiousness link was driven by outliers and therefore should be interpreted with caution.
…Gender moderated the perfectionistic strivings–neuroticism link (β = .12, p = .018, R2 = .07), the other-oriented perfectionism–neuroticism link (β = .44, p = .001, R2 = .39), the discrepancy–conscientiousness link (β = −.30, p = .005, R2 =.68), the socially prescribed perfectionism–agreeableness link (β = .85, p = .033, R2 =.27), and the self-oriented perfectionism–agreeableness link (β = .68, p = .002, R2 =.76).
Notably, perfectionistic strivings’ positive relationship with neuroticism increased as the percentage of females increased. The implied disattenuated correlations between perfectionistic strivings and neuroticism for an all-male, a 50% female, and an all-female sample were rc + = .04, rc + = .10, and rc + = .16. Likewise, other-oriented perfectionism’s positive relationship with neuroticism increased as the percentage of females increased. The implied disattenuated correlations for other-oriented perfectionism and neuroticism were rc + = –.20, rc + = .02, and rc + = .24.
Similarly, the negative relationship between discrepancy and conscientiousness increased as the percentage of females increased. The implied disattenuated correlations between discrepancy and conscientiousness for an all-male, a 50% female, and an all-female sample were rc + = –.04, rc + = –.20, and rc + = –.34. Also, socially prescribed perfectionism’s and self-oriented perfectionism’s negative relationships with agreeableness decreased as the percentage of females increased.
J.5 Uz Baş (2011). Dimensions of Perfectionism in Elementary School-Aged Children: Associations with Anxiety, Life Satisfaction, and Academic Achievement. Eğitim ve Bilim.
ABSTRACT:The main purpose of the study was to examine the relationship of the dimensions of perfectionism with anxiety, life satisfaction, and academic achievement, and to determine whether dimensions of perfectionism significantly predict anxiety, overall life satisfaction, and academic achievement in elementary school-aged children. Differences in the level of the dimensions of perfectionism according to gender and grade level were also investigated. Four hundred and eighteen children (198 girls and 220 boys) completed the Adaptive-Maladaptive Perfectionism Scale, Trait Anxiety Inventory for Children, and The Brief Multidimensional Students' Life Satisfaction Scale. Results revealed that sensitivity to mistakes and need for admiration were significantly and positively correlated with anxiety, while contingent self-esteem and compulsiveness were significantly and positively correlated with all life satisfaction domains, overall life satisfaction, and academic achievement. Sensitivity to mistakes was found as a significant predictor of anxiety, while contingent self-esteem and compulsiveness were found as significant predictors of both life satisfaction and academic achievement. Findings also indicated that there were significant effects of gender on sensitivity to mistakes, contingent self-esteem, and compulsiveness, and that there were significant grade level effects on compulsiveness and need for admiration.
J.6 Stornelli, Flett, & Hewitt (2009). Perfectionism, Achievement, and Affect in Children: A Comparison of Students From Gifted, Arts, and Regular Programs. Canadian Journal of School Psychology.
EXCERPT: Analyses of variance were conducted on levels of self-oriented and socially prescribed perfectionism. The between-subjects variables were program type and gender. There were no significant differences for levels of socially prescribed perfectionism. Analyses of the self-oriented perfectionism data yielded no significant main effects, but there was an interaction effect, F(1, 281) = 5.36, p < .01. Post hoc analyses found that female students in the arts program had substantially elevated levels of self-oriented perfectionism relative to the male arts students and students in the other programs. There was no evidence indicating that levels of perfectionism were elevated among students in the gifted program.
J.7 Hewitt, Caelian, Flett, Sherry, Collins, & Flynn (2002). Perfectionism in children: Associations with depression, anxiety, and anger. Personality and Individual Differences.
EXCERPT: In terms of gender differences, there were no differences in mean levels of the variables for boys versus girls. There were also no differences between boys and girls in the magnitude of correlations, although there was a trend for the relationship between socially prescribed perfectionism and anger suppression to be greater for boys (r=0.40) than girls (r=0.05, Z=1.89).
APPENDIX K: OTHER RELEVANT STUDIES
K.1 Hansen, Sabia, & Schaller (2022). In-Person Schooling and Youth Suicide: Evidence from School Calendars and Pandemic School Closures (Working Paper No. 30795). National Bureau of Economic Research.
ABSTRACT: This study explores the effect of in-person schooling on youth suicide. We document three key findings. First, using data from the National Vital Statistics System from 1990-2019, we document the historical association between teen suicides and the school calendar. We show that suicides among 12-to-18-year-olds are highest during months of the school year and lowest during summer months (June through August) and also establish that areas with schools starting in early August experience increases in teen suicides in August, while areas with schools starting in September don’t see youth suicides rise until September. Second, we show that this seasonal pattern dramatically changed in 2020. Teen suicides plummeted in March 2020, when the COVID-19 pandemic began in the U.S. and remained low throughout the summer before rising in Fall 2020 when many K-12 schools returned to in-person instruction. Third, using county-level variation in school reopenings in Fall 2020 and Spring 2021—proxied by anonymized SafeGraph smartphone data on elementary and secondary school foot traffic—we find that returning from online to in-person schooling was associated with a 12-to-18 percent increase teen suicides. This result is robust to controls for seasonal effects and general lockdown effects (proxied by restaurant and bar foot traffic), and survives falsification tests using suicides among young adults ages 19-to-25. Auxiliary analyses using Google Trends queries and the Youth Risk Behavior Survey suggests that bullying victimization may be an important mechanism.
FIGURES:
APPENDIX L: COLLABORATIVE REVIEW DOCS
Here are all of the open-source collaborative reviews that I have created, with various co-curators, to study what is happening to the mental health and life outcomes of Gen Z, and also a few related to what is happening to liberal democracies
- Social media and mental health: A collaborative review
- Social media and political dysfunction: A collaborative review
- What is happening to boys? A collaborative review
- Adolescent porn use and mental health: A collaborative review
- Dating app use and mental health: A collaborative review
- All international coddling collaborative docs
- Evidence for and against the coddling: A collaborative review
- Free Play and Mental Health: A Collaborative Review
- Digital Gambling and Adolescent Health: A Collaborative Review.
- Online Communities and Adolescent Health: A Collaborative Review.
- Changes in Cognitive Abilities Among Children and Adults. A Review.
APPENDIX M: IS THE RISE JUST BECAUSE OF CHANGES IN REPORTING?
This inclues both the changes in willingness to self-report as well as changes in actual reporting of psychiatric illnesses and suicide.
[COMING SOON]
In the meantime, see our document on alternative hypotheses to the mental health crisis
* * * * * * * * * * * * * * *