The pediatric mental health crisis

The October 2021 joint declaration by the AAP, AACAP, and CHA was an unusual document. Professional medical societies rarely use the word "emergency" outside of acute public-health threats like pandemics or natural disasters. By applying it to a slow-moving population-level shift, they signaled both the magnitude of the problem and a frustration with the policy response to date. The U.S. Surgeon General's advisory in December 2021 followed the same logic and used similar language. These were not internal documents — they were public alarms designed to mobilize attention and resources. Whether they succeeded is the question the next decade will answer. So far, the resources have not followed the alarm.

CDC's Youth Risk Behavior Survey shows the share of high school students reporting persistent feelings of sadness or hopelessness rising from 26% in 2009 to 42% in 2021. For girls, the figure reached 57%. Reports of seriously considering suicide rose from 14% to 22% overall and to 30% among girls. Emergency department data show suspected suicide attempts among adolescent girls rising 51% in early 2021 compared with the same period in 2019. Inpatient psychiatric admissions for self-harm and suicidal ideation roughly doubled in many systems. The pattern is consistent across self-report, clinician-reported, and hospital-administrative data sources, which makes pure detection bias unlikely as a full explanation.

Jonathan Haidt and Jean Twenge have argued that the rapid rewiring of adolescent social life around smartphones and social media, completed between roughly 2010 and 2015, is the most plausible primary driver of the trend. Their evidence: the timing fits, the gender pattern fits, the dose-response pattern fits, and similar trends appear across Anglophone countries with similar adoption curves. The mechanism is plausible — algorithmically curated comparison, sleep displacement, reduced face-to-face contact, exposure to harmful content, and a loss of unstructured time. Critics note that effect sizes in well-controlled studies are modest, that experimental evidence for causation remains limited, and that other plausible mechanisms operate during the same window. The honest read is that the case is strong enough to act on, but not closed.

A wider lens captures changes that predate smartphones and continued in parallel. Free play declined steadily from the 1970s onward. Academic pressure intensified, with homework loads, standardized testing, and college-admissions anxiety reshaping adolescence. Time outdoors collapsed. Sleep duration fell. Community institutions — religious congregations, youth groups, extended families nearby — thinned. The Great Recession introduced sustained economic anxiety in households raising the current adolescent generation. The political environment after 2016 generated chronic stress signals to children paying any attention at all. Smartphones plausibly accelerated and concentrated distress, but the substrate was changing already.

COVID-19 did not start the trend, but it bent the curve sharply upward. School closures, social isolation, family stress, grief, and the disruption of routines hit adolescents hard, and especially hit kids who were already vulnerable. Emergency department visits for mental health crises spiked. Eating disorder admissions surged. The reopening did not reverse the spike fully. Some of the pandemic effect is plausibly receding; some appears to have produced lasting cohort effects in kids whose key developmental years occurred during the disruption. Treating the crisis as a pre-pandemic trend amplified by an acute shock is the most defensible synthesis.

A counter-narrative argues that distress hasn't risen — only reporting has. Reduced stigma, expanded screening, and clinician willingness to apply mental health labels have increased measured prevalence without changing underlying conditions. There is something to this: language shifts, awareness campaigns, and TikTok-mediated self-diagnosis have all increased the salience of mental health categories for adolescents. But the hospitalization and mortality data are harder to inflate this way. Suicide deaths, ICU admissions for overdose, emergency department visits for self-harm — these are events, not labels, and they have risen. Detection accounts for some of the survey trend, not all of it, and not the most consequential parts.

The U.S. has roughly one child and adolescent psychiatrist per 9,000 children. The estimated need is one per 2,000 or fewer. Roughly half of U.S. counties have zero. The workforce did not collapse — it failed to expand while demand exploded. Training pipelines are slow (it takes a decade to produce a board-certified child psychiatrist from medical-school entry), reimbursement for child psychiatry is lower than for many specialties, and burnout in the field is high. The result is that even families with means, insurance, and information cannot find a clinician for their child. Workforce expansion would take years to bend the curve even if it started today, which makes the lack of urgency in current policy especially costly.

Schools have become the largest de facto provider of pediatric mental health services in the United States, mostly by accident. School counselors, social workers, and psychologists screen, refer, and increasingly provide direct services. The ratio of school counselors to students recommended by the American School Counselor Association is 1:250. The actual U.S. average is closer to 1:400, with many states above 1:600. School-based health centers, where they exist, provide some of the highest-yield interventions because they are accessible to kids who would otherwise never reach care. Federal funding for school mental health expanded after 2021 but remains a fraction of what the scale requires.

The shortage of pediatric psychiatric beds is acute and worsening. Children in crisis routinely wait days or weeks in emergency rooms while placement is sought, a phenomenon called "boarding." Many hospitals have closed pediatric psychiatric units due to staffing and reimbursement pressures even as demand has risen. The 988 Suicide and Crisis Lifeline, launched in 2022, provides a national entry point but routes to local services whose capacity varies enormously by geography. Mobile crisis teams that can respond to a home or school are available in some jurisdictions and absent in most. The absence of a functional crisis system means families default to police and emergency rooms, neither of which is designed for or effective at pediatric mental health crises.

Mental health parity laws nominally require insurers to cover mental health services at the same level as physical health services. Enforcement is weak, and pediatric mental health is particularly underpaid. Many child psychiatrists do not accept insurance at all because reimbursement does not cover the cost of complex visits, parent contact, school coordination, and care management. Medicaid pays better in some states than others. The result is a two-tier system in which families with resources pay out of pocket and families without resources go without. Tightening parity enforcement, raising Medicaid rates for pediatric behavioral health, and reimbursing collaborative-care and integrated-behavioral-health models would shift practice patterns within a few years. None of this is technically hard. The obstacle is political.

If social-media exposure is a meaningful driver of adolescent distress — which the evidence increasingly suggests — then platform design choices matter at a population scale. Age verification, restrictions on algorithmic feeds for minors, defaults toward private accounts, limits on overnight use, and stronger protections against harmful content are all technically feasible and politically contested. Several states have passed legislation in this direction; courts have blocked some of it on First Amendment grounds. The federal Kids Online Safety Act has moved through Congress in various forms. Whatever one thinks of specific bills, the basic point is that the current platform-design choices were made for adult users and applied to children by default. That default is the policy question.

The system as it stands requires families to be their child's case manager. That means: insist on screening at every well-child visit, push for warm handoffs and not just referrals, use every door (pediatrician, school, telehealth, crisis line, primary care) because no single door reliably opens, document everything, escalate when responses are vague, learn to recognize warning signs early, and protect sleep, time outdoors, and face-to-face connection as if they were medications. None of this should be the family's job. All of it is, until the structural changes catch up. Naming the system failure does not relieve the family of operating inside it.

A coherent response would include: expanding the clinical workforce through training subsidies, loan forgiveness, and faster pipelines; integrating behavioral health into pediatric primary care as the default; funding school-based services to recommended ratios; building functional crisis systems including mobile teams and stabilization units; expanding inpatient capacity; enforcing parity and raising reimbursement; regulating platform design choices that affect minors; and protecting time for play, sleep, and unstructured childhood. None of this is novel. All of it is contained in the AAP, AACAP, and Surgeon General documents. The question is whether the country will fund what it has declared an emergency, or whether the declaration will become its own conclusion. The next cohort of teenagers is being shaped by the answer.

American Academy of Pediatrics, American Academy of Child and Adolescent Psychiatry, Children's Hospital Association. "Declaration of a National Emergency in Child and Adolescent Mental Health." October 19, 2021.

Office of the U.S. Surgeon General. Protecting Youth Mental Health: The U.S. Surgeon General's Advisory. Washington, DC: U.S. Department of Health and Human Services, 2021.

Centers for Disease Control and Prevention. Youth Risk Behavior Survey Data Summary & Trends Report: 2011–2021. Atlanta: CDC, 2023.

Haidt, Jonathan. The Anxious Generation: How the Great Rewiring of Childhood Is Causing an Epidemic of Mental Illness. New York: Penguin Press, 2024.

Twenge, Jean M. Generations: The Real Differences Between Gen Z, Millennials, Gen X, Boomers, and Silents — and What They Mean for America's Future. New York: Atria Books, 2023.

Benton, Tami D., Rhonda Boyd, and Wanjikũ F. M. Njoroge. "Addressing the Global Crisis of Child and Adolescent Mental Health." JAMA Pediatrics 175, no. 11 (2021): 1108–1110.

Koplewicz, Harold S. The Scaffold Effect: Raising Resilient, Self-Reliant, and Secure Kids in an Age of Anxiety. New York: Harmony Books, 2021.

Shatkin, Jess P. Born to Be Wild: Why Teens Take Risks, and How We Can Help Keep Them Safe. New York: TarcherPerigee, 2017.

Yard, Ellen, Lakshmi Radhakrishnan, Michael F. Ballesteros, et al. "Emergency Department Visits for Suspected Suicide Attempts Among Persons Aged 12–25 Years Before and During the COVID-19 Pandemic." MMWR Morbidity and Mortality Weekly Report 70, no. 24 (2021): 888–894.

Orben, Amy, and Andrew K. Przybylski. "The Association Between Adolescent Well-Being and Digital Technology Use." Nature Human Behaviour 3, no. 2 (2019): 173–182.

Hoffmann, Jennifer A., Bonnie T. Zima, Layla Soares, et al. "Pediatric Mental Health Boarding in U.S. Emergency Departments." Pediatrics 152, no. 4 (2023): e2023062612.

Bitsko, Rebecca H., Angelika H. Claussen, Jesse Lichstein, et al. "Mental Health Surveillance Among Children — United States, 2013–2019." MMWR Supplements 71, no. 2 (2022): 1–42.