Foster care system problems: what the data shows.
The state takes custody of children to keep them safe. Even the government’s own self-reported data shows how often that promise breaks — and independent research shows it’s worse than the official numbers admit.
About 3 in 4 removals are filed as “neglect,” not abuse, and neglect is closely tied to poverty. Children in care face a higher risk of abuse than at home, roughly half of youth who age out experience homelessness, and the official federal numbers undercount the damage because agencies report on themselves.
Written for parents in every state. This guide covers federal law and how the process works nationwide. Where we give a state’s rule or deadline, it’s labeled as an example — your state’s may be different. Ask your lawyer or the court clerk for your local rules, and track your own deadlines in a free FAN case file.
Foster care facts at a glance: what the government reports.
On September 30, 2025, 331,747 children were in U.S. foster care, and 501,745 passed through the system during the year. Entries and the number in care both rose, the first increase after years of decline. These figures are preliminary and self-reported by the same agencies being measured — treat them as a floor, not the full picture.
Every state reports its foster care data to the federal government through AFCARS. The table below compares the two most recent years (AFCARS FFY 2025; AFCARS FFY 2024).
| Measure | FFY 2025 | FFY 2024 |
|---|---|---|
| In care on September 30 | 331,747 | 328,947 |
| Entered care | 175,008 | 170,943 |
| Exited care | 169,927 | 176,730 |
| Served during the year | 501,745 | 505,682 |
| Waiting for adoption | 67,249 | 70,421 |
| Parental rights terminated for all parents | 46,308 | 49,994 |
| Adopted | 43,380 | 46,935 |
| Aged out | 15,030 | 15,379 |
Federal fiscal years run October 1 to September 30. FFY 2025 data are preliminary (AFCARS Report No. 33, data as of Feb 27, 2026). In FFY 2020, 407,332 children were in care on September 30.
Behind every number is a family. In FFY 2025, 46% of children who left care went home to their parents, and 20% of all exits came after more than three years in care (AFCARS FFY 2025). The rest of this page looks at what happens in between.
The system grades its own homework.
Almost every official foster care number comes from the agencies themselves. States and counties define, count and report it differently, often late, sometimes not at all — and the gaps always run in the system’s favor.
There is no single, enforced definition of “neglect,” so the same home can be written up in one county and left alone in the next. Only about 16.7% of investigations substantiate any abuse at all, yet more than 650,000 children are labeled abused or neglected every year, the vast majority for vague neglect (FAN brochure, NCANDS/AFCARS).
When outside auditors check, the numbers move. Federal inspectors found 8 states kept no records of children placed in hotels and offices (HHS OIG, 2026), estimated 34,869 missing-child episodes were never reported to NCMEC as federal law requires (HHS OIG, 2023), and found 978 children went missing from Missouri foster care in a single year — roughly half never reported to police or NCMEC (FAN brochure (HHS OIG, 2021)).
| Measure | What the government reports | What independent research shows |
|---|---|---|
| Removals filed as “neglect” | 55% of FFY 2025 entries (AFCARS, preliminary) | About 75% of removals (FAN Research) |
| Abuse in foster care | 9.07 victimizations per 100,000 days in care (federal benchmark) | 2 to 28 times more likely than at home, from foster homes to group institutions (Johns Hopkins & state studies) |
| Deaths in care | No consistent national count by type of placement | 25–50 per 100,000 in foster care vs. about 27 for all children (FAN Research) |
| Youth aging out each year | 15,030 in FFY 2025 | Roughly 20,000 a year in widely cited counts (The Truth About the System) |
| Homelessness after aging out | Not tracked by AFCARS | About 1 in 2 within 18 months (The Truth About the System) |
| Trafficking victims with a child-welfare history | Not tracked by AFCARS | 60–70% nationally; 86% in Connecticut (FAN Research) |
Federal figures are self-reported by state agencies. Independent figures come from the studies and reports cited in FAN’s research and in our brochure, The Truth About the System.
A system that counts its own failures will always find fewer of them.
FAN Research Desk
Children removed for poverty, not abuse.
About 75% of removals are filed as “neglect,” not physical or sexual abuse. A family earning under $15,000 a year is 22 times more likely to be flagged. Being poor gets read as being dangerous.
About three-quarters of child welfare cases involve neglect reports, according to the federal Child Welfare Information Gateway. Its 2023 brief, Separating Poverty From Neglect, found children in low-income families are 7 times more likely to be found neglected (CWIG). Analyses of federal NCANDS and AFCARS data put neglect at about 75% of removals (FAN Research), and families earning under $15,000 a year are 22 times more likely to be flagged (FAN brochure). The government’s own preliminary entry data lists neglect in 55% of cases, a figure that depends on how each state chooses to code a removal.
Neglect and poverty can look the same on a caseworker’s form: an empty fridge, an unstable apartment, a parent working nights with no child care. In FFY 2025, caretaker incarceration was listed in 7% of entries (AFCARS FFY 2025).
When the problem is rent, a removal doesn’t fix it. It adds a court case to the rent.
FAN Research Desk
Some courts already draw the line. An Illinois court’s shelter care information sheet says income level or lack of access to money cannot be used to find a parent unfit (Illinois 17th Circuit). The same federal brief shows what happens when help goes to families directly instead. Read more in what works better.
Foster kids sleeping in hotels, offices and cabins.
Federal auditors found 9,045 children were placed in temporary emergency placements, including hotels and agency offices, between January 2022 and June 2023. The longest hotel stay lasted 251 days.
The HHS Office of Inspector General’s July 2026 national overview counted 15,705 stays. Of 49 responding states, 42 used these placements, and 8 kept no records of them. The 9,045 figure counts only the 30 states whose data could be matched to individual children, so the real number is likely higher (HHS OIG A-07-24-06114).
Hotels accounted for 1,922 stays in 17 states. State and county offices accounted for 2,698 stays in 22 states. Another 4,412 stays were at other unlicensed sites, such as state-park cabins. The average stay was 7 days (HHS OIG).
Arizona placed the most children this way, 3,481, according to Stateline’s reporting on the audit. Texas had 3,923 placements involving 1,977 children (Stateline).
This is the system’s own placement. These children were removed to be kept safe, then slept in places no family would be licensed to use. If your child has been placed in a hotel or office, write down the dates and ask your lawyer to raise it in court.
If your child is in foster care, document everything.
Start a free, secure case file. Track visits, placement moves, medications and every missed service, and request a FAN advocate when you’re ready.
Moved again and again: placement instability.
Children who enter foster care move an average of 4.48 times for every 1,000 days in care, by the federal government’s own national measure.
That figure comes from ACF’s Round 4 Child and Family Services Reviews data (ACF CFSR Round 4). America’s Health Rankings, using federal Child Welfare Outcomes data, reports that 18.7% of children in foster care had three or more placements within 12 months in 2024 (America’s Health Rankings).
Each move can mean a new school, a new caseworker, new rules and the loss of whatever friendships a child had left. NPR and The Marshall Project reported on one Alaska foster youth who moved 15 times in 10 years (NPR, 2021).
If your child is moved, ask in writing why, when you’ll be told the new address and how visits will continue. Log every move in your case file with the date and the reason you were given.
Overmedication: psychotropic drugs in foster care.
A federal investigation found foster children were prescribed psychotropic drugs at 2.7 to 4.5 times the rate of other children on Medicaid, based on 2008 data from five states.
The U.S. Government Accountability Office reviewed 2008 Medicaid data from Florida, Massachusetts, Michigan, Oregon and Texas and reported its findings in December 2011 (GAO-12-270T). In Massachusetts, 39.1% of foster children were prescribed psychotropics, compared with 10.2% of other Medicaid children. In Texas it was 32.2% versus 7.1%.
- Five or more drugs at once: 0.11% to 1.33% of foster children across the five states, compared with 0.01% to 0.07% of other children. That was 609 foster children. GAO’s experts said no evidence supports this practice.
- Doses above FDA-label maximums: 1.12% to 3.27% of foster children, compared with 0.16% to 0.56% of other children.
- Infants under 1: 0.3% to 2.1% of foster infants were given psychotropics, compared with 0.1% to 1.2% of other infants. GAO’s experts said there is no established mental health use for these drugs in infants.
GAO noted that higher rates do not necessarily mean inappropriate prescribing. Foster children have often been through trauma. But the gap is wide enough that parents should ask questions.
Ask your lawyer how to get your child’s medication list and who consented to each prescription. Write down every change you notice at visits.
Unsafe in state custody: abuse, missing children and trafficking.
Federal auditors found state agencies did not always ensure children missing from foster care were reported to the National Center for Missing & Exploited Children, as federal law requires. An estimated 34,869 episodes were never reported at all.
In a March 2023 audit, HHS OIG reviewed 74,353 missing-child episodes of two days or more from July 2018 to December 2020. It estimated 51,115, about 69%, were not reported to NCMEC as required (OIG A-07-21-06102). A second audit estimated 13,983 episodes, about 19%, were not reported to law enforcement for entry into the national crime database as required, and 8,324 were never reported (OIG A-07-21-06104).
When children came back, many weren’t checked for trafficking. In five states, OIG found no evidence of a sex-trafficking screening for 65% of children returning from a missing episode: 72% of boys and 59% of girls (OIG OEI-07-19-00371). An NCMEC analysis of 2013 to 2017 data found 86% of missing boys who were likely child sex trafficking victims were in the care of social services when they ran (NCMEC, 2018).
Children are also harmed while in care. The federal national performance figure for maltreatment in foster care is 9.07 victimizations per 100,000 days in care (ACF CFSR Round 4). In Georgia, a U.S. Senate subcommittee report found about 1,800 children in state custody were reported missing from 2018 to 2022, and NCMEC testified that 410 were likely sex trafficked, according to 11Alive. Georgia’s agency disputes the method (11Alive).
Racial disproportionality in foster care.
Black children were 23% of foster care entries in 2024 but only 14% of the U.S. child population. American Indian and Alaska Native children entered care at 5.9 per 1,000, compared with 1.9 nationally.
Those figures come from the Annie E. Casey Foundation’s analysis, updated September 2026 (Annie E. Casey Foundation). On September 30, 2025, Black children made up 25% of all children in care (AFCARS FFY 2025). American Indian and Alaska Native children entered at about three times the national rate.
Because neglect is so closely tied to poverty, a system that treats poverty as neglect will fall hardest on the communities with the least money. That’s a design problem, not a parenting problem.
Aging out with nothing.
Roughly 20,000 young people age out of foster care every year with no family to catch them. About half are homeless within 18 months, a quarter are incarcerated within two years, and only half have a job by 24.
Those figures, gathered in our brochure from the Annie E. Casey Foundation and other researchers, are far worse than anything in the federal dashboards, because AFCARS stops counting young people the day they leave care. The federal government reported 15,030 youth aged out in FFY 2025 (AFCARS FFY 2025). Foster youth are also twice as likely to drop out of school, carry PTSD at twice the rate of U.S. combat veterans, and each one costs taxpayers more than $300,000 over a lifetime (The Truth About the System).
The Midwest Study followed 732 young people who aged out in Illinois, Iowa and Wisconsin (Chapin Hall). By age 26, between 31% and 46% had been homeless at least once; 36% was the observed figure (AJPH, 2013).
At 26, 70% had earnings in the past year, and half of those earned $9,000 or less. Nearly 70% of the women and over 40% of the men had received food stamps in the past year (Annie E. Casey Foundation).
These young people were in the state’s legal custody right up until they became adults. The system that removed them is the system that let them go.
Keeping children connected to parents and relatives from the start is the best protection against aging out alone.
Slow permanency, and the termination clock.
Only 35.2% of children who enter foster care reach a permanent home within 12 months. Meanwhile, federal law pushes states to seek termination of parental rights after 15 of 22 months.
The 35.2% figure is the national performance measure in ACF’s Round 4 review data. The same data show 5.6% of children re-enter care within 12 months of leaving (ACF CFSR Round 4).
On September 30, 2025, 67,249 children were waiting to be adopted, and 46,308 had already had parental rights terminated for all parents. For 18% of children in care, a case goal had not yet been established (AFCARS FFY 2025).
| Exit from care, FFY 2025 | Share of exits |
|---|---|
| Reunification with parents | 46% |
| Adoption | 26% |
| Guardianship | 10% |
| Emancipation (aging out) | 9% |
| Living with a relative | 6% |
Source: AFCARS Dashboard No. 33 (preliminary).
If your child is in care, the clock matters. The federal termination trigger has an exception when the state hasn’t provided the services needed for your child to come home. Read our family and dependency court survival guide and our guide to termination of parental rights.
Follow the money: incentives that reward placement.
Agencies aren’t paid a fee per removal. But federal reimbursement is tied to children being in care and has no cap, adoption bonuses reward beating past adoption numbers, and nearly half of all child welfare spending goes to out-of-home placement.
Title IV-E, the main federal foster care program, is an open-ended entitlement. For each eligible child in a licensed placement, it reimburses 50% to 83% of room-and-board costs depending on the state, plus 50% of administrative costs, for as long as the child is in care (ACF Title IV-E). Federal child welfare funding totals $11.7 billion in FY 2026, about $10.1 billion of it Title IV-E (CRS IF10590).
States spent $34.3 billion on child welfare from all sources in state fiscal year 2022 (Child Trends). Nearly half goes to out-of-home placements, and prevention is “a much smaller proportion,” according to Child Trends (Child Trends). The question isn’t how much the system spends. It’s where the money points.
Congress continues to fund Adoption and Legal Guardianship Incentive Payments, with $75 million appropriated for FY 2026 (CRS). States earn them only for adoptions and guardianships above their own baseline (42 U.S.C. 673b).
Children’s own Social Security benefits
In 2003 the Supreme Court ruled unanimously that a state acting as a foster child’s representative payee may use the child’s Social Security benefits to repay its own foster care costs (Keffeler). NPR and The Marshall Project reported in 2023 that the practice had been found in at least 49 states plus D.C., often without telling the child or family. Some states and cities, including Arizona, have since moved to protect the money (NPR, 2023).
Agencies aren’t paid per removal. But the money follows children into care, not back home.
FAN Research Desk
What works better: kin and direct help for families.
The strongest evidence for keeping children out of foster care involves money and support going directly to families, not to agencies. And 39% of children in care already live with relatives.
The federal Child Welfare Information Gateway’s 2023 brief summarizes research on direct support to families (CWIG, 2023):
- $1,000 more family income through the EITC: 8% to 10% less CPS involvement
- Each $1 minimum wage increase: 9.6% fewer neglect reports
- Housing subsidies: 50% fewer foster placements
- Each extra month of child care subsidy: 16% lower odds of a neglect report
Relatives are already doing much of the work. On September 30, 2025, 39% of children in care lived with relatives or kin (AFCARS FFY 2025). A 2023 federal rule lets states use separate, simpler licensing standards for kin homes and requires equal maintenance payments for licensed kin and non-kin homes (ACF IM-23-07).
If your child has been removed, name every relative who could help, as early as possible, and ask the court to consider them. Then document everything.
Frequently asked questions
How many children are in foster care in the U.S.?
331,747 children were in foster care on September 30, 2025, and 501,745 were served during federal fiscal year 2025. These are preliminary federal AFCARS figures.
What are the biggest problems with the foster care system?
Removals tied to poverty, frequent placement moves, high rates of psychotropic drugs, abuse in care, missing children who go unreported, stays in hotels and offices, racial disproportionality, slow permanency and poor outcomes after aging out.
Why are most children removed from their homes?
About 75% of removals are filed as “neglect,” not physical or sexual abuse, and neglect is strongly linked to poverty. The government’s own preliminary entry data lists neglect in 55% of cases, but states code removals differently and undercount.
Are official foster care statistics accurate?
They are a floor, not the full picture. Nearly all federal figures are self-reported by the same agencies being measured, definitions vary by state, and federal auditors have repeatedly found missing records and unreported missing children.
Is poverty treated as neglect?
Often, critics say. A 2023 federal brief found children in low-income families are 7 times more likely to be found neglected, and that direct income support, housing and child care help reduce CPS involvement.
Do foster kids really sleep in hotels and offices?
Yes. HHS OIG found 9,045 children had temporary emergency placements from January 2022 to June 2023, including 2,698 stays in offices and 1,922 in hotels. The longest hotel stay lasted 251 days.
How many foster children go missing?
HHS OIG reviewed 74,353 missing episodes from July 2018 to December 2020 and estimated 34,869 were never reported to the National Center for Missing & Exploited Children.
Are foster children overmedicated?
A 2011 GAO investigation found foster children were prescribed psychotropic drugs at 2.7 to 4.5 times the rate of other Medicaid children in 2008, including some infants. GAO noted higher rates don’t necessarily mean inappropriate prescribing.
How often do foster children change placements?
The federal national measure is 4.48 placement moves per 1,000 days in care for children entering care. America’s Health Rankings reports 18.7% had three or more placements in 12 months in 2024.
Are Black children overrepresented in foster care?
Yes. Black children were 23% of entries in 2024 but 14% of the U.S. child population. American Indian and Alaska Native children entered at 5.9 per 1,000, compared with 1.9 nationally.
How many foster children age out each year?
Roughly 20,000 a year in widely cited counts. The federal government reported 15,030 in FFY 2025.
What happens to kids who age out of foster care?
About half experience homelessness within 18 months, a quarter are incarcerated within two years, and only half are employed by 24. The Midwest Study found 31% to 46% had been homeless at least once by age 26.
How many children are waiting to be adopted?
67,249 on September 30, 2025. Of all children in care, 46,308 had parental rights terminated for all parents.
Do agencies get paid for each child they remove?
No per-removal fee exists. But federal Title IV-E reimbursement is open-ended and tied to eligible children being in care, and adoption bonuses reward states for beating their past adoption numbers. Critics say this tilts incentives toward placement.
How many foster children live with relatives?
39% on September 30, 2025. A 2023 federal rule lets states use simpler licensing standards for kin homes and requires equal payments for licensed kin and non-kin homes.
Don’t miss the next story.
New investigations, court updates, families’ wins and free help for parents — every day. Follow FAN and share what the system doesn’t want seen.
Foster care abuse we’ve reported
Sources (34) +
- FAN: The Truth About the System
- FAN brochure: The Truth About the System (PDF)
- FAN Research: 75% of foster care removals start as neglect
- FAN Research: children in foster care face higher abuse risks
- FAN Research: are children in foster care dying at higher rates?
- FAN Research: sex trafficking in foster care
- ACF: AFCARS Dashboard FFY 2025 (No. 33, preliminary)
- ACF: AFCARS Dashboard FFY 2024 (No. 32)
- ACF: CFSR Round 4 Statewide Data Workbook (March 2026)
- America’s Health Rankings: foster care stability
- GAO-12-270T: psychotropic drugs for foster children (2011)
- HHS OIG A-07-21-06102: missing children and NCMEC reporting (2023)
- HHS OIG A-07-21-06104: missing children and NCIC reporting (2023)
- HHS OIG OEI-07-19-00371: sex-trafficking screening after missing episodes (2022)
- HHS OIG A-07-24-06114: temporary emergency placements (July 2026)
- Stateline: most states temporarily house foster children in hotels, offices (2026)
- NCMEC: missing male victims of child sex trafficking (2018)
- 11Alive: Senate report on Georgia foster care
- Dworsky, Napolitano & Courtney, AJPH (2013)
- Annie E. Casey Foundation: Midwest Evaluation summary
- Chapin Hall: Midwest Evaluation
- Annie E. Casey Foundation: foster care population by race and ethnicity (2026)
- CWLA: The Nation’s Children 2026
- Child Welfare Information Gateway: Separating Poverty From Neglect (2023)
- Illinois 17th Judicial Circuit: Shelter Care Hearing Information Sheet
- ACF IM-23-07: kinship licensing standards
- NPR / The Marshall Project (2021)
- NPR / The Marshall Project: foster care and Social Security benefits (2023)
- ACF: Title IV-E Foster Care
- CRS IF10590: child welfare funding (Aug 2026)
- 42 U.S.C. 673b: adoption and guardianship incentive payments
- Child Trends: how states spend child welfare funds, SFY 2022 (2025)
- Child Trends: largest share of spending is out-of-home placements (2026)
- Washington State DSHS v. Guardianship Estate of Keffeler, 537 U.S. 371 (2003)
Father’s Advocacy Network is a nonprofit newsroom and family-advocacy organization, not a law firm. This guide is general information, not legal advice. Laws differ by state and change often; talk to a family defense lawyer about your situation. Spot an error? Tell our corrections desk.
