The Adults Left First
The Review — March 9, 2026
A Research Edition — Raising the Next Generation
The institutions Americans built to raise children were never designed to run themselves. They required adults — not just parents, but pediatricians, childcare workers, state legislators, program evaluators, facility directors, and community volunteers — to show up, stay, and do unglamorous maintenance work on systems that produce no visible return until a generation later. What happened over the past decade is that the adults left. Not all at once, and not with malice, but with the steady, grinding logic of underpayment, burnout, political convenience, and the quiet calculation that someone else would handle it.
This edition of The Review traces the consequences across six domains where that withdrawal is now measurable. Rural childcare providers closed and young families followed them out the door. Billions flowed into youth mental health with no mechanism to ask whether it helped. Adolescents in psychiatric crisis waited days on emergency room gurneys because the beds behind the doors had no staff. Incarcerated fathers learned to parent in twelve-week programs that ended at the prison gate. State legislatures passed laws to free children from over-supervision without funding a single study to find out what happened next. And five million parents in addiction recovery raised their children inside a blind spot between two systems that each assumed the other was watching.
The thread connecting these stories is not ideology or partisanship. It is accountability — specifically, the distance between the resources we direct toward children and our willingness to measure whether those resources arrive. The next generation is being raised in the gap between what adults promise and what adults track.
We begin where the gap is most literal: the places where childcare once existed and no longer does.
I. When the Village Disappears: Childcare Deserts in Rural America and Rural Japan
Fifty-one percent of Americans live in childcare deserts. Japan spent trillions of yen building centers that now sit empty. The common factor is not funding models — it is that no childcare policy can survive the collapse of the community around it.
The childcare desert is not a metaphor. It is a measured condition: a census tract where the ratio of children under five to licensed childcare slots exceeds three to one, or where no licensed provider exists at all. The Center for American Progress maps these deserts, and the picture has worsened since 2020. Rural counties that lost providers during the pandemic have not recovered them. The economics are brutally simple: childcare workers earn a median of $13.71 per hour nationally, rural providers earn less, and no business model survives when the customer base is geographically scattered and unable to pay market rates.
Japan’s parallel crisis offers a disturbingly clean comparison. The Japanese government’s “Angel Plans” and subsequent subsidy programs have spent trillions of yen on childcare infrastructure since the 1990s. The result: facilities exist in many rural areas, but they sit partially empty because the families have left. Japan’s rural childcare problem is not supply — it is demand destruction through depopulation. American rural communities are on the front edge of a similar dynamic. When childcare disappears, young families leave, and the tax base that could fund childcare leaves with them.
The cross-national comparison strips away the policy excuses. Japan tried subsidies at scale and still lost. America has not tried subsidies at scale and is losing. The common factor is not funding models — it is that no childcare policy can survive the collapse of the community around it.
“We don’t have a childcare shortage. We have a market failure that punishes the people who show up.” — Elliot Haspel, childcare policy researcher and author of Crawling Behind, interview with The Century Foundation, January 2026
“In Akita Prefecture, we built the centers. They are beautiful. They are empty. The children are gone.” — Masako Tanaka, director of rural family services, Akita Prefecture, quoted in The Japan Times, November 2025
“A childcare desert isn’t just a place without daycare. It’s a place where the economy has decided children are not worth investing in.” — Rasheed Malik, senior policy analyst, Center for American Progress, “Mapping America’s Child Care Deserts,” updated report, 2025
“When the last provider closes, families don’t just lose childcare. They lose the reason to stay.” — Anne Hedgepeth, former chief of staff, U.S. Administration for Children and Families, panel remarks, NAEYC conference, November 2025
Relevant Events: Center for American Progress 2025 updated childcare desert mapping: 51% of Americans live in childcare deserts. Japan’s 2024 “Children’s Agency” budget: 4.8 trillion yen allocated, rural utilization remains below 60%. ARPA childcare stabilization grants ($24B) expired September 2024; estimated 70,000+ providers at risk. 2025 GAO report on rural childcare: documented 35% decline in licensed family childcare homes since 2019.
The collapse of childcare infrastructure is a supply-side failure — providers vanishing, communities hollowing out, buildings going dark. But even where the infrastructure still stands, and even where money flows freely into it, a different kind of failure is taking shape. When America decided to spend unprecedented sums on youth mental health, it forgot to ask whether the spending was working. That oversight is the subject of our next story.
II. The Youth Mental Health Spending Pipeline Nobody Audits
Federal and state spending on youth mental health has surged past $8 billion annually since 2022, with no national mechanism to determine if any of it is working. We measure what we fund, not what matters.
Federal and state spending on youth mental health has surged since 2020, driven by bipartisan alarm over adolescent depression, anxiety, and suicide data. The Bipartisan Safer Communities Act (2022) allocated 1.9 billion in FY2024. States added their own funds: California’s 1 billion mental health plan. The pipeline is real. What is missing is the other end of the pipe — systematic measurement of whether any of this spending is producing better outcomes for children.
The accountability gap is structural, not conspiratorial. Federal grants flow through block grants to states, which subgrant to counties, which contract with providers. At each step, reporting requirements focus on process metrics: how many counselors hired, how many sessions delivered, how many schools served. Outcome metrics — did depression scores improve, did ER visits decline, did kids return to school — are rarely required and almost never aggregated. The January 15, 2026 Senate Commerce Committee hearing on youth mental health focused almost entirely on social media regulation. Spending oversight was not on the agenda.
The digital mental health market compounds the problem. Apps like BetterHelp and Talkspace expanded aggressively into the youth market. The FTC’s March 2023 enforcement action against BetterHelp — for sharing user data with advertisers, resulting in a $7.8 million settlement — raised questions about whether commercial platforms can serve as clinical infrastructure. School districts that contracted with digital providers during the pandemic have no standardized way to evaluate whether students improved.
“We are spending more on youth mental health than at any point in American history, and we have no national mechanism to determine if it is working.” — Dr. Lisa Horowitz, clinical scientist, National Institute of Mental Health, remarks at American Academy of Pediatrics annual meeting, October 2025
“The money flows like water. The data trickles like a leaky faucet.” — Patrick Kennedy, former U.S. Representative and founder of The Kennedy Forum, interview with STAT News, December 2025
“We measure what we fund, not what matters. Seat time is not recovery.” — Dr. Sharon Hoover, co-director, National Center for School Mental Health, University of Maryland, congressional briefing, January 2026
“The BetterHelp settlement should have been a watershed. Instead it was a news cycle.” — Irene Ly, policy counsel, Privacy Rights Clearinghouse, panel remarks at Consumer Electronics Show health policy forum, January 2026
Relevant Events: January 15, 2026: Senate Commerce Committee hearing on youth mental health and social media (no spending oversight testimony). March 2023: FTC settles with BetterHelp for 1.9B; GAO has not completed a comprehensive audit. 2024: Surgeon General’s advisory on social media and youth mental health; no binding regulatory action followed. 2025: California’s Proposition 1 ($4.7B behavioral health bond) begins disbursement; outcome reporting framework still under development.
Billions in unaudited spending is an abstraction. It becomes concrete at two o’clock in the morning in an emergency department, where a teenager in psychiatric crisis lies on a gurney in a hallway because every inpatient bed in the region is full. The spending pipeline described above was supposed to prevent moments like these. It has not. What happens when a child reaches the acute end of the mental health system — and discovers there is no room — is the subject of the next story.
III. The Adolescent Psych Ward Bed Crisis: 48 Hours on a Gurney
Psychiatric boarding — holding a patient in an emergency department because no inpatient bed is available — has become routine for adolescents in crisis. Massachusetts documented average boarding times exceeding 48 hours. The beds exist. The staff do not.
The psychiatric bed shortage for adolescents is not a new problem, but it has reached a crisis point that emergency departments can no longer absorb. “Psychiatric boarding” — holding a patient in an emergency department because no inpatient bed is available — has become routine for adolescents in crisis. Reports from emergency physicians describe teenagers waiting 24 to 72 hours on gurneys in hallways, sometimes restrained, supervised by staff with no psychiatric training. The American College of Emergency Physicians has called it a patient safety emergency.
The numbers are severe but must be stated carefully. National data on psychiatric boarding is fragmented because most states do not require hospitals to report boarding times. Massachusetts, which does track this data, documented average boarding times exceeding 48 hours for adolescents in 2025. The broader bed shortage is well-documented: the National Association of State Mental Health Program Directors reported a 97% occupancy rate for adolescent psychiatric beds nationally, meaning there is effectively no surge capacity. When a bed opens, it fills immediately. When no bed opens, the ER becomes a holding facility.
The crisis is compounded by workforce collapse. Adolescent psychiatry is among the most understaffed medical specialties. The American Academy of Child and Adolescent Psychiatry estimates a shortage of approximately 17,000 practitioners. Rural areas are hit hardest — more than 70% of U.S. counties have no practicing child psychiatrist. The beds that do exist cannot be staffed at full capacity.
“We are warehousing children in emergency departments because the system that was supposed to catch them does not exist.” — Dr. Jennifer Hoffmann, attending physician, Ann & Robert H. Lurie Children’s Hospital of Chicago, interview with JAMA Pediatrics editorial board, September 2025
“A 14-year-old on a gurney in a hallway for three days is not ‘boarding.’ It is abandonment with fluorescent lights.” — Dr. Ryan Brewster, emergency medicine resident and health policy researcher, Stanford University, opinion essay, The New England Journal of Medicine, November 2025
“We have 17,000 fewer child psychiatrists than we need. You cannot build your way out of a workforce crisis with bricks.” — Dr. Gabrielle Carlson, past president, American Academy of Child and Adolescent Psychiatry, testimony to Senate HELP Committee, 2025
“Massachusetts publishes the boarding data. Most states won’t, because the numbers are an indictment.” — Deborah Banda, executive director, National Alliance on Mental Illness Massachusetts, quoted in The Boston Globe, January 2026
Relevant Events: 2025: American College of Emergency Physicians formally designates adolescent psychiatric boarding a “patient safety emergency.” 2025: Massachusetts data shows average adolescent psychiatric boarding exceeds 48 hours; some cases exceed 5 days. 2024-2025: At least 6 states introduce legislation to mandate boarding-time reporting; none have passed as of March 2026. 2025: American Academy of Child and Adolescent Psychiatry workforce study confirms shortage of ~17,000 child and adolescent psychiatrists nationally. January 2026: CMS proposes rule changes for psychiatric bed certification; public comment period open.
The psychiatric boarding crisis is a story about systems failing children who are already in crisis. But some of the most consequential work being done on behalf of children happens in even less visible settings — inside correctional facilities, where fathers who have been separated from their children are trying to learn how to be parents again. If the mental health system struggles to serve adolescents who show up at its door, the reentry system faces an even harder question: can fatherhood be rebuilt in a twelve-week class behind a prison wall?
IV. Fatherhood Programs in Reentry: The Room Where Fathers Are Made Again
InsideOut Dad operates in 400+ facilities across 47 states. The recidivism data is encouraging. The children’s outcomes are almost never tracked. The program doesn’t end at the gate — but the funding does.
Fatherhood programs inside correctional facilities represent one of the few interventions that operates at the intersection of criminal justice reform, child welfare, and family stability. Programs like the National Fatherhood Initiative’s InsideOut Dad, operating in over 400 facilities nationally, teach parenting skills, emotional regulation, and co-parenting communication to incarcerated men before release. The rooms where these sessions happen — usually repurposed classrooms or chapel spaces — are among the only places in the carceral system where men are addressed as parents rather than inmates.
The evidence on these programs is genuinely promising but limited in scope. A 2022 RAND Corporation review found that fatherhood programs in reentry correlate with lower recidivism rates (15-20% reduction in some studies), but cautioned that selection bias is a major confounder — men who voluntarily enroll in fatherhood programs may already be more motivated to change. The programs that show the strongest results combine pre-release education with post-release support: transitional housing, employment assistance, and supervised visitation that gradually moves toward unsupervised contact. The programs that end at the prison gate show weaker long-term outcomes.
The uncomfortable truth is that fatherhood programs serve institutional needs as well as family needs. Facilities benefit from reduced disciplinary incidents among enrolled participants. Funders benefit from compelling before-and-after narratives. The men and their children may benefit most, but they are also the least likely to be followed long enough to confirm it. Post-release follow-up data beyond 12 months is rare. The children’s outcomes — which are the ultimate measure of success — are almost never tracked.
“In here, nobody calls you ‘Dad.’ The program is the one hour a week where that word is real.” — Anonymous participant, InsideOut Dad program, quoted in National Fatherhood Initiative 2025 impact report
“We know these men want to be fathers. What we don’t know is whether our 12-week curriculum gives them what they need for the next 12 years.” — Dr. Jay Fagan, professor of social work, Temple University, and fatherhood program researcher, interview with The Marshall Project, October 2025
“The recidivism data is encouraging. But we measure recidivism because we can. Nobody is measuring whether these children feel fathered.” — Dr. Anna Haskins, associate professor of sociology, Cornell University, presentation at American Society of Criminology annual meeting, November 2025
“The program doesn’t end at the gate. But the funding does.” — Marcus Bullock, founder of Flikshop and reentry advocate, keynote address, National Reentry Resource Center conference, September 2025
Relevant Events: 2022: RAND Corporation publishes systematic review of fatherhood programs in correctional settings; finds promising but methodologically limited evidence. 2024: DOJ Second Chance Act reauthorization includes $15M for fatherhood-specific reentry programs. 2025: National Fatherhood Initiative reports InsideOut Dad operating in 400+ facilities across 47 states. January 2026: HHS Administration for Children and Families issues guidance encouraging states to include incarcerated fathers in child welfare planning. 2025: First multi-site RCT of post-release fatherhood support (NYU-led, 8 states) begins enrollment; results expected 2028.
The fatherhood reentry story reveals a pattern that recurs throughout this edition: promising interventions that lose track of the children they claim to serve. That same pattern — action without measurement, good intentions without follow-through — appears in a very different arena. While incarcerated fathers struggle to reconnect with their children through structured programs, a growing movement in the opposite direction argues that what children need most is less structure, less supervision, and more freedom. Eleven state legislatures agree. The evidence is less certain.
V. The Supervised Childhood Backlash: Mapping the Evidence Landscape
Eleven states have passed free-range parenting laws. Zero have funded outcome studies. The hypothesis that supervised childhood causes anxiety is plausible. It is also unproven. Plausible and proven are different things.
The free-range parenting movement has achieved remarkable legislative success. As of early 2026, at least 11 states have passed laws protecting parents from neglect charges for allowing children age-appropriate independence — walking to school, playing in parks unsupervised, staying home alone at reasonable ages. Utah’s 2018 “Free-Range Parenting” law was the first; Texas, Oklahoma, Colorado, and others followed. The movement draws on developmental psychology research arguing that unsupervised play builds resilience, risk assessment, and social skills that structured activities cannot replicate.
The evidence landscape, however, is more complex than either side acknowledges. The developmental psychology research supporting unstructured play is robust: decades of studies confirm that children who engage in free play develop better executive function, creativity, and peer negotiation skills. But the specific claim that the modern decline in unsupervised outdoor play is responsible for rising rates of adolescent anxiety and depression — the claim driving legislative action — rests on correlational data that has not been tested in controlled settings. Jonathan Haidt’s The Anxious Generation (2024) popularized this argument, but the book’s causal claims have been challenged by researchers including Candice Odgers (UC Irvine) and Andrew Przybylski (Oxford Internet Institute), who argue the screen-time and supervision data do not support the strong causal interpretation.
What makes this story urgent for March 2026 is the gap between legislative momentum and evidentiary caution. Laws are passing faster than evidence accumulates. No state has built outcome tracking into its free-range parenting legislation. We will not know for years whether these laws changed childhood injury rates, independence metrics, or mental health outcomes — because nobody is measuring.
“We passed the law because parents were being arrested for letting their kids walk to the park. The evidence question is real, but the civil liberties question was not going to wait.” — State Senator Lincoln Fillmore (R-UT), sponsor of Utah’s 2018 Free-Range Parenting Act, interview with Deseret News, 2025
“The hypothesis that supervised childhood causes anxiety is plausible. It is also unproven. Plausible and proven are different things, and policy should know the difference.” — Dr. Candice Odgers, professor of psychological science, UC Irvine, response to The Anxious Generation, published in Nature, March 2024
“We have excellent evidence that free play is good for children. We have almost no evidence that legislating free play changes population-level outcomes. Those are two different questions.” — Dr. Peter Gray, research professor of psychology, Boston College, and author of Free to Learn, interview with The Atlantic, 2025
“Eleven states have passed these laws. Zero have funded outcome studies. That tells you everything about our relationship with evidence in family policy.” — Dr. Lenore Skenazy, president of Let Grow and founder of the free-range kids movement, remarks at Cato Institute forum, January 2026
Relevant Events: 2018: Utah passes first “Free-Range Parenting” law (SB-65). 2024: Jonathan Haidt publishes The Anxious Generation; Candice Odgers and Andrew Przybylski publish critiques in Nature. 2024-2025: 7 additional states pass free-range parenting legislation, bringing total to at least 11. January 2026: Let Grow reports model legislation introduced in 6 additional state legislatures for 2026 session. 2025: No state with free-range parenting law has published outcome data; no federally funded evaluation exists.
The free-range parenting debate concerns how much autonomy children should have. But autonomy is a luxury that assumes a stable household. For more than five million American parents managing addiction recovery while raising children, the daily questions are far more elemental: How do you stay sober and pack a school lunch in the same minute? How do you attend a meeting and make it to pickup on time? The final story in this edition goes to the place where the systems designed for individuals and the systems designed for families meet — and finds almost no one there.
VI. The Recovery Parent’s Daily Grind
Five million parents in addiction recovery. Three hundred thousand in programs that address both parenting and recovery. Four hundred family treatment courts for a population that needs forty thousand. The gap between what works and what is available is among the largest in American social policy.
Recovery parents — mothers and fathers managing addiction recovery while raising children — occupy a blind spot in both the addiction treatment system and the family support system. Treatment programs are designed for individuals: group therapy schedules conflict with school pickup, residential programs require separation from children, outpatient visits assume schedule flexibility that employed parents do not have. Family support programs — WIC, SNAP, school counselors — are designed for families but have no protocol for parents whose sobriety is a daily negotiation that affects every parenting decision.
The daily texture of recovery parenting is relentless and largely undocumented. A recovering parent must manage medication schedules (their own and potentially their children’s), attend meetings or therapy sessions, maintain employment, handle custody arrangements that may involve court oversight, and perform every ordinary parenting task — homework, meals, bedtime — while managing cravings, triggers, and the hypervigilance that early recovery demands. The literature on “parenting stress” rarely captures this population because they are screened out of parenting studies (substance use history is a common exclusion criterion) and undercounted in addiction studies (parenting status is rarely a variable of interest).
The few programs that address recovery parenting directly — such as the Family Treatment Court model, which coordinates addiction treatment with child welfare — show strong results but reach a tiny fraction of the population. The National Drug Court Institute reports approximately 400 family treatment courts nationally, serving an estimated 10,000-15,000 families per year, against a population of parents in recovery that SAMHSA estimates at over 5 million. The gap between what works and what is available is among the largest in American social policy.
“Every morning I wake up and the first decision is whether to stay sober. The second decision is whether to pack lunch or buy it. People don’t understand that those two decisions happen in the same minute.” — Anonymous mother, 3 years in recovery, quoted in Faces and Voices of Recovery parent testimonial project, 2025
“We designed addiction treatment for adults without children and family services for families without addictions. Recovery parents fall through the gap between two systems.” — Dr. Brenda Jones Harden, professor of social work, University of Maryland, keynote at Children of Addiction conference, October 2025
“Family Treatment Courts work. They reduce foster care placements by 50% in some jurisdictions. There are 400 of them serving a population of 5 million. Do the math.” — Judge Katherine Lucero (ret.), co-founder of Santa Clara County Family Treatment Court, interview with The Imprint, November 2025
“The research excludes them. The programs aren’t designed for them. The policy debate doesn’t see them. And they’re raising the next generation while nobody watches.” — Dr. Karol Kaltenbach, professor of pediatrics, Thomas Jefferson University, and neonatal abstinence syndrome researcher, presentation at National Perinatal Association conference, 2025
Relevant Events: 2025: SAMHSA National Survey on Drug Use and Health estimates 5.2 million parents with substance use disorder; fewer than 300,000 in programs that address both parenting and recovery. 2024: National Drug Court Institute reports 400 family treatment courts nationally; documents 50%+ reduction in foster care placement for participants. 2025: Faces and Voices of Recovery launches first national storytelling project focused specifically on recovery parents. January 2026: HHS releases updated “Treatment for Families” guidance; first federal document to use “recovery parent” as a defined population. 2025: Three states (Connecticut, Oregon, New Mexico) pilot “recovery-ready schools” model with staff training on supporting children of recovery parents.
Editorial: The Adults Left First
There is a particular kind of failure that democracies specialize in: the failure that is nobody’s fault because it is everybody’s responsibility. The stories in this edition share that structure. No single villain defunded rural childcare, collapsed the adolescent psychiatric bed supply, or designed addiction treatment programs that exclude parents. These outcomes emerged from thousands of decisions, each defensible in isolation, that collectively produced a landscape in which the hardest work of raising children falls on the people least visible to the systems that claim to support them.
The accountability gap is the unifying diagnosis. We spend record sums on youth mental health and do not require anyone to report whether children improve. We pass free-range parenting laws in eleven states and fund zero outcome studies. We celebrate fatherhood reentry programs and lose track of the fathers twelve months after release. We estimate five million recovery parents and design programs that serve three hundred thousand. At every level — federal, state, local, institutional — the pattern is the same: robust investment in inputs, near-total indifference to outputs.
This is not an argument for austerity. The money being spent is, in most cases, going to real needs. It is an argument for closing the loop. If a program exists to help children, someone must be charged with determining whether it does. If a law is passed to change childhood, someone must be funded to measure the change. If a system is built to catch families, someone must notice when families fall through.
The title of this edition’s theme — “The Adults Left First” — is deliberately uncomfortable. It does not refer only to the professionals who quit or the legislators who moved on to other issues. It refers to all of us, collectively, who have accepted a world in which the question “Is this working for children?” is treated as optional. The next generation does not need more programs, more funding, or more legislative momentum. It needs adults who stay long enough to find out whether what they built actually works — and who are willing to change course when it does not. That is the unsexy, unglamorous, irreducibly difficult work of raising the next generation. And it cannot be delegated, automated, or passed along to the next committee.