"Immediately, and in No Case Later Than 24 Hours": The Law That Was Supposed to Count America's Missing Foster Children — and the Nine Years Nobody Enforced It
Investigation by Project Milk Carton
"Immediately, and in No Case Later Than 24 Hours": The Law That Was Supposed to Count America's Missing Foster Children — and the Nine Years Nobody Enforced It
A Project Milk Carton investigation — child welfare transparency, public data, and the children the federal government cannot count.
By PMC Intelligence Desk
PROLOGUE: THREE SISTERS, AND A SECRETARY WHO DIDN'T KNOW
In October 2017, Kansas legislators asked the state's Secretary for Children and Families a simple question: how many foster children are missing right now? The answer, assembled from the state's private contractors, was roughly seventy — a number the Secretary herself did not have until lawmakers forced it into the open. Among the missing were three sisters, ages 12, 14, and 15, who had been gone from their placement for weeks without the head of the agency knowing they were gone at all. They were eventually found in the company of a 48-year-old man, who was arrested.
By October 2017, federal law had required for more than a year that every one of those children be reported to law enforcement and to the National Center for Missing & Exploited Children — "immediately, and in no case later than 24 hours" after the state learned they were missing. That is not a paraphrase. It is the operative text of 42 U.S.C. § 671(a)(35)(B), written by Congress in 2014, in force since September 29, 2016.
This investigation is about what happened to that sentence. It is about the federal government's own auditors finding, years later, that roughly 69 percent of qualifying missing-from-foster-care episodes were never reported to NCMEC on time — nearly half never reported at all. It is about 110,446 missing episodes in thirty months, about 6,619 children still missing on a single winter day in 2020, and about the number of dollars any state has ever lost for ignoring the law: zero.
It is also about something stranger and quieter than noncompliance. Nine years after Congress ordered these children counted, there is still no federal system that counts them. The children the 2014 law was written to make visible remain, in the federal data of record, largely invisible. What follows is the paper trail — statute by statute, audit by audit, state by state.
The investigation proceeds in five movements. First, the law itself — what Congress actually wrote, when it took effect, and how Congress came back in 2023 to say it again. Second, the measurement — the federal government's own four-audit series quantifying the failure. Third, the void — the discovery that no federal data system counts these children at all. Fourth, the vacuum — the penalty machinery that exists on paper and has never once been used. Fifth, the ground truth — state by state, from the state that counts because a judge forces it to, to the state that has never counted at all. That last state is ours.
Everything in this article is drawn from primary sources: the U.S. Code, public laws, HHS Office of Inspector General audits, sworn congressional testimony, federal court records, state agency reports, peer-reviewed research, and Project Milk Carton's own NCMEC-fed database. Every figure is attributed. Where a widely repeated statistic could not be traced to a primary source, we did not use it — and we tell you so at the end.
PART I — THE LAW CONGRESS WROTE
A sentence with a deadline in it
The Preventing Sex Trafficking and Strengthening Families Act — Public Law 113-183 — was signed on September 29, 2014. Its Section 104 carries a heading that reads like a mission statement: "LOCATING AND RESPONDING TO CHILDREN WHO RUN AWAY FROM FOSTER CARE."
The mechanism Congress chose was Title IV-E of the Social Security Act — the roughly $9-billion-a-year federal foster care funding stream. To stay eligible for IV-E money, § 671(a)(35) requires every state plan to promise two things:
First, by September 29, 2015, each state had to "develop and implement specific protocols" for expeditiously locating any child missing from foster care; determining the primary factors that contributed to the child running away or otherwise being absent from care; determining the child's experiences while absent — "including screening the child to determine if the child is a possible sex trafficking victim" — and reporting related information as the Secretary requires.
Second, by September 29, 2016, the operative duty:
"the State agency shall report immediately, and in no case later than 24 hours after receiving, information on missing or abducted children or youth to the law enforcement authorities for entry into the National Crime Information Center (NCIC) database of the Federal Bureau of Investigation… and to the National Center for Missing and Exploited Children."
The covered population — defined through § 671(a)(9)(C)(i)(I) — is broader than children physically placed in foster homes. It includes children with open case files who were never removed from home, runaways from foster care, and older youth receiving federal transition services. Congress was not writing a narrow rule for one placement type. It was writing a floor under every child the system touches.
The 24-hour clock has a companion on the law-enforcement side: under the National Child Search Assistance Act (34 U.S.C. §§ 41307–41308), once police receive a missing-child report, the NCIC entry must be made within two hours. Congress built a relay: agency to police within 24 hours, police to the FBI's national database within two. The relay only works if the first runner leaves the blocks.
Congress came back and said it again
In January 2023, Congress strengthened the mandate. Section 137 of Public Law 117-348 — the Trafficking Victims Prevention and Protection Reauthorization Act of 2022 — added duties to maintain regular communication with law enforcement and NCMEC, and to include, "where reasonably possible," a photo, a physical description, and endangerment information in every report: pregnancy status, medications, suicidal tendencies, trafficking vulnerability.
Section 136 of the same law is a sense-of-Congress provision with a barb in it: it pointedly notes that these reports "were required to be provided… beginning on September 30, 2016." Legislators do not usually restate a deadline that is being met.
The implementing guidance existed from the start. The Children's Bureau issued program instruction ACYF-CB-PI-14-06 in November 2014, telling states exactly what the new plan requirements were. Eight years later, in December 2022, ACF issued another guidance document — ACF-IM-22-01, "Responding to Human Trafficking among Children and Youth in Foster Care and Missing from Foster Care." Hold that second date. It matters later, because a guidance memo turns out to be the federal government's entire enforcement response.
PART II — THE FAILURE, QUANTIFIED BY THE GOVERNMENT ITSELF
Four audits, one verdict
Between 2022 and 2023, the HHS Office of Inspector General published a four-part body of work on children missing from foster care. It remains the only serious federal measurement of § 671(a)(35) compliance ever performed. Its design tells its own story: OIG could not simply pull the numbers from a federal database, because — in the auditors' own words —"there is not a single data system that has missing children's data available for all States." OIG had to survey all fifty states by hand to build its universe. Four states could not produce usable data at all, leaving a 46-state frame, from which the auditors drew a stratified national sample of 100 missing episodes.
Here is what the federal government found about the federal law:
A-07-20-06095 (May 2022) — the national snapshot. Between July 1, 2018 and December 31, 2020 — thirty months —1,016,895 children passed through foster care, and states logged 110,446 missing-from-care episodes. On December 31, 2020,6,619 foster children were still missing. Sixty-five percent of the missing were aged 15–17 — the age band NCMEC's data associates with the highest trafficking risk.
Investigation data visualization
A-07-21-06102 (March 2023) — the NCMEC reporting audit. Of the 74,353 episodes lasting two days or more, OIG estimated that 51,115 — roughly 69 percent — were never or late reported to NCMEC as the statute requires. An estimated 34,869 episodes (47 percent) were never reported at all; another 16,246 were reported late. In the 100-episode sample itself: 45 episodes never reported, 22 late, and only 33 reported within the statutory 24 hours. The 90-percent confidence interval on the noncompliance estimate ran from 43,609 to 58,622 episodes — meaning even the rosiest statistical read leaves a majority of episodes unreported or late.
A-07-21-06104 (May 2023) — the law-enforcement side. An estimated 13,983 episodes (18.8 percent) were noncompliant on the police-reporting duty —8,324 never entered into NCIC at all, 5,659 entered late. Better than the NCMEC number but consider what it means: more than eight thousand times in thirty months, a child disappeared from state custody and no record of it ever reached the FBI's national crime database. To every officer in America running a query, those children did not exist.
OEI-07-19-00371 (June 2022) — the return-screening audit. The same 2014 statute requires that when a missing child returns, the state screen them to determine whether they were a sex-trafficking victim while gone. In five states examined,65 percent of returned children — 268 of 413 — had no documented screening at all (72 percent of boys; 59 percent of girls). Of the screenings that were documented, one-third recorded no conclusion. The children came back, and in two cases out of three, nobody wrote down whether anything had happened to them.
Why it failed
Both 2023 audits land on the same systemic cause, stated almost verbatim in each:" State agencies generally lacked adequate systems to readily identify whether or not they had reported missing children's episodes." Nearly all sampled states could not tell federal auditors whether NCMEC reports had been made at all. This is worth sitting with. The failure is not merely that states missed a deadline. It is that most states had no mechanism to know whether they had complied with federal law — and therefore neither did anyone else.
Translate the estimates back into children and the scale becomes hard to hold. Fifty-one thousand episodes never or late reported to NCMEC in thirty months is roughly fifty-six failures per day, every day, for two and a half years. The 34,869 never-reported episodes mean that, on average, about thirty-eight times each day, a child vanished from a state's custody and the one national organization built to look for missing children was never told. Each of those episodes had a first hour, a first night. The statute exists because Congress decided those hours belong to searchers, not to silence. The audits establish that, most of the time, the silence won.
One more feature of the audit series deserves emphasis: its restraint. OIG's estimates deliberately exclude episodes shorter than two days from the NCMEC analysis — the 24-hour statute has no such exclusion. The auditors measured a subset of the legal duty and still found a 69 percent failure rate. The full statutory failure rate, if anyone ever measures it, can only be worse.
Inside the sample: the states that failed every time
Appendix C of the NCMEC audit names names. Among sampled episodes:
Nevada: 4 of 4 episodes never reported to NCMEC — a 100 percent failure rate in the federal sample.
Nebraska: 3 of 3 never reported.
Alaska, Connecticut, Illinois, Kentucky: every sampled episode (2 of 2 each) never reported.
Mississippi: 3 never and 1 late out of 4. California: 3 never of 5. Missouri: 3 never of 5.
Texas, for contrast: 1 never, 2 proper, of 3 sampled.
A sample of two or four episodes is not a state report card — OIG designed a national estimate, not fifty state audits. But when every episode the federal government happened to pull from your state was never reported, it is a data point with a direction. Nevada's direction returns in Part VIII.
PART III — THE STATES THAT COULDN'T ANSWER
The snapshot audit's state-level appendix (July 2018–December 2020) reads like a map of the problem's scale:
The rates and repeat patterns matter as much as the raw counts. Kansas and Maryland tied for the worst missing rate at 7 percent of children in care. Nevada ran 6 percent — with an average of five missing episodes per missing child, among the worst repeat rates in the country. Connecticut was worse still: an average of seven episodes per missing child. Maryland's missing children stayed gone an average of 52 days. Georgia's, 46 days. Missouri's, 37.
And then there are the states that answered the federal government's only census of missing foster children with shrugs:
Rhode Island's data was unusable.
Pennsylvania and West Virginia submitted incomplete data.
New York provided information only for IV-E-eligible children — meaning its true count is higher than anything in the federal record.
Texas could not provide dates of birth for 614 of its own missing episodes.
Remember the frame: this was not a routine reporting exercise the states fumbled. This was the federal government's one attempt, ever, to count children missing from its own funded foster care system — and four states could not produce usable numbers, while others filtered or truncated theirs. The census itself became evidence.
PART IV — THE DATA VOID
The federal foster care database does not track missing children
AFCARS — the Adoption and Foster Care Analysis and Reporting System — is the federal data system of record for foster care. If you ask it about missing children, here is everything it can say: on September 30, 2022,3,742 children (1 percent of the foster population)had "Runaway" recorded as their placement setting. That is a point-in-time snapshot of a single day. AFCARS does not collect missing episodes, dates, durations, recoveries — or whether the legally required reports to NCMEC and NCIC were ever made.
Now line up the three numbers the federal ecosystem produces for the same phenomenon:
AFCARS snapshot: ~3,700children "currently runaway" on one day.
NCMEC intake: ~23,000missing-from-care reports per year.
OIG episode census: ~44,000missing episodes per year (110,446 over thirty months).
Each number is a floor under the next, and no federal agency reconciles them. And since OIG found that roughly47 percent of episodes are never reported to NCMEC at all, even NCMEC's 23,000 is a floor, not a count. The honest answer to "how many children go missing from American foster care each year" is: the federal government does not know, and has no system designed to know.
The count that was never repeated
Investigation data visualization
The OIG snapshot was a one-time exercise. It has not been repeated. There is no recurring federal count of children missing from foster care — not annual, not biennial, not at all. The Government Accountability Office has never issued a report specifically on missing-from-care counts or on AFCARS's undercounting of missing children; the nearest GAO work is a 2011 report (GAO-11-444) on law-enforcement NCIC timeliness — which noted, tellingly, that police had trouble getting information from child-welfare agencies— and a 2003 report on AFCARS reliability generally. We searched for the missing GAO report on missing foster children. Its absence is itself a finding, and we document it as one.
PART V — THE ENFORCEMENT VACUUM
The penalty machinery that exists on paper
Section 671(a)(35) is a Title IV-E state-plan requirement. In theory, two enforcement paths hang over it.
Path one: the Child and Family Services Reviews. Under 45 C.F.R. §§ 1355.32–1355.36, states found out of "substantial conformity" with federal child-welfare requirements enter a Program Improvement Plan; failure can trigger withholding of between 1 and 3 percent of a defined funding pool — composed of the state's Title IV-B allotments plus 10 percent of its Title IV-E administrative claims — with the percentage escalating by review round to a theoretical annual maximum of 42 percent. But walk into the machinery and you find the gap that swallows the whole statute: the CFSR conformity criteria at 45 C.F.R. § 1355.34 contain no measure of missing-child reporting, runaway episodes, or NCMEC/NCIC compliance. The 24-hour rule is simply not a scored item. A state can fail every missing-child report it ever owes and remain, for CFSR purposes, in perfect standing on that subject — because the subject is not on the test. And withholding is suspended during any active improvement plan in any case.
Path two: state-plan disapproval. Under 45 C.F.R. § 1355.30(n)(2), the Secretary can move against a state plan that does not meet requirements. But this reaches the plan document. A state whose written plan contains the § 671(a)(35) assurance — and every state's does — faces nothing here, no matter what its practice looks like. The separate Title IV-E eligibility reviews audit per-child payment eligibility, not missing-child reporting.
Has any state ever been penalized? No.
We ran the question to ground systematically: ACF's penalty and policy pages, both nationwide OIG audits, the HHS Departmental Appeals Board's published decisions, the Congressional Research Service's Title IV-E reports (R42794 and R43757), the Federal Register's CFSR rulemakings, and open-web searches pairing the statute's citations with every enforcement term in the vocabulary — penalty, disallowance, withholding, compliance action.
The result: no instance exists — anywhere in the public record — of any state losing Title IV-E funds or facing any formal federal penalty for violating the missing-child reporting requirement. Nine years of the mandate. A federally measured 69 percent failure rate. Zero dollars.
For perspective on how rare federal child-welfare penalties are in general: the only CFSR withholding ever litigated to a published decision is Idaho's —$935,712, upheld by the Departmental Appeals Board in 2023 after a failed Round 3 improvement plan. It had nothing to do with missing children. CRS's review of the first two CFSR rounds (2001–2010) found that no jurisdiction in the country was in full conformity with federal child-welfare policy — and records no state-plan penalties. The machinery is not merely unused against § 671(a)(35); it is barely used at all.
What the federal government did instead
Confronted with its own Inspector General's finding that states failed to make roughly seven of every ten required NCMEC reports, the Administration for Children and Families responded with: an information memorandum (ACF-IM-22-01, December 1, 2022) and webinars— one in November 2022, one in January 2023.
Then the file was closed. OIG's recommendation on NCMEC reporting was marked closed as "implemented" on March 20, 2024. The NCIC recommendation followed on August 23, 2024. In the federal accountability ledger, the matter of the unreported missing children is resolved.
Congress has never added teeth
Since 2014, exactly one legislative vehicle has addressed the subject: the Find and Protect Foster Youth Act (S. 1146 / H.R. 2426 in the 118th Congress), a bipartisan bill — Grassley, Cornyn, Stabenow, Capito in the Senate; Brown and Gonzales in the House — that would have required HHS to provide technical assistance and evaluate state compliance with the missing-child protocols. Note what it did not contain: penalties. Even the gentlest version of accountability — a study — moved slowly. The Senate passed it October 31, 2023. The House passed an amended version 408–7 on September 18, 2024. The two chambers never reconciled the versions before the 118th Congress ended. It did not become law. A bill to evaluate compliance with a ten-year-old mandate, passing one chamber 408 to 7, still died of inattention.
In twelve years, no bill has ever proposed attaching Title IV-E funding penalties to § 671(a)(35) violations. And no class action has ever pleaded the statute as a standalone claim — missing-from-care facts surface inside broader foster-care litigation (Texas's M.D. v. Abbott, West Virginia's Jonathan R. v. Morrisey, revived by the Fourth Circuit in June 2026, Oregon's Wyatt B., settled 2024), but the reporting mandate itself is enforced, if at all, only by the administrative machinery described above. Which is to say: not at all.
A mandate without a mechanism is a suggestion. That is the structural finding of this investigation, and everything else in it is the evidence.
PART VI — THE TRAFFICKING PIPELINE
Why the 24 hours matter
The 2014 law was not a paperwork statute. Its title names its purpose: Preventing Sex Trafficking. The reporting clock exists because the first hours a child is missing are the hours a NCMEC analyst can push a photo out, a poster can circulate, an NCIC flag can catch a traffic stop. Every unreported episode spends those hours in silence.
What follows is NCMEC's own accounting, restricted to figures we traced to primary sources — sworn testimony and NCMEC's own publications. (One widely circulated statistic did not survive that screen; see the editorial note at the end.)
2016 data, sworn: NCMEC President and CEO John F. Clark testified to the House Committee on Education and the Workforce on March 16, 2017 that of endangered runaways flagged as likely sex-trafficking victims,86 percent were in the care of social services or foster care when they went missing.
2021: NCMEC reported that19 percent of children reported missing from social-services care were likely sex-trafficking victims.
2023, sworn: NCMEC President and CEO Michelle DeLaune's written testimony (February 6, 2024) put the risk in ratio form: for missing children generally, roughly 1 in 6 is a likely sex-trafficking victim; for children missing from care, the likelihood rises to1 in 5. That year NCMEC assisted on 28,886 missing-child cases;22,442 — 78 percent — were children who ran from a social-services facility.
2024: Per NCMEC's "2024 in Numbers" and Senate-side testimony in December 2025:29,568 missing-child cases, of which 23,160 — again 78 percent — were children missing from foster or state care. One in seven of all missing children reported were likely sex-trafficking victims; 91 percent of missing children were recovered.
2025:NCMEC's live impact reporting:23,348 missing-from-care reports; 17 percent flagged as likely sex-trafficking victims; 4,215 instances of victim support for likely trafficking victims missing from child welfare.
Set NCMEC's figures beside one from the federal snapshot:65 percent of children who went missing from foster care were aged 15 to 17.The population going missing and the population NCMEC classifies at elevated trafficking risk are substantially the same teenagers. And recall that the same 2014 statute that created the 24-hour reporting duty also created the return-screening duty — the one OIG found unmet in 65 percent of cases. Congress built the statute as a loop: report the child out, screen the child back. The audits show the loop broken at both ends — the disappearance unreported going out, the experience undocumented coming back. A child can transit the entire episode, whatever happened during it, without generating a single record anywhere.
Two disciplines in reading these numbers. First, "likely sex-trafficking victim" is NCMEC's own official case-classification term, drawn from its analysts' assessment — we quote it verbatim and do not harden it into a definitive count of trafficked children. Second, the statistic changed direction over the years — early figures describe what share of likely trafficking victims came from care; later figures describe what share of missing-from-care children are likely victims. The denominators are different and must never be mixed. We haven't.
And one sworn sentence that frames the whole statute. DeLaune, February 2024, on the world before the 2014 law:" Before this law was enacted, most children who went missing from child welfare placements were not reported to NCMEC or law enforcement." The OIG audits establish the epilogue: after the law, most still weren't.
Investigation data visualization
The peer-reviewed corroboration
The strongest academic evidence comes from Florida. Latzman, Gibbs and colleagues, publishing in Children and Youth Services Reviewin 2019 (vol. 98, pp. 113–124), followed roughly 37,000 youth who had a foster placement at age 10 or older, 2011–2017, through state administrative data. Nineteen percent had at least one runaway episode from foster care. Of those runaways, 7 percent — 542 children — had a human-trafficking allegation while on runaway status. Most striking: for 70 percent of youth who ever had a trafficking allegation, the first identified allegation occurred during a foster-care runaway episode. The runaway episode is not incidental to the trafficking risk. It is, in the majority of identified cases, the doorway. (The study is Florida-only and counts allegations, not confirmed victimization; a federally funded companion analysis appears as OPRE Report 2020-143.)
The cost that has never been measured
Here is a question no federal document answers: what happens, in outcomes, when the 24-hour report is missed? No published federal data compares cases where the report was made against cases where it wasn't. The failure has been measured; its cost has not. The nearest documented consequences are the 65 percent of returned children never screened for trafficking — and the Texas numbers in the next section, where the state's own reports show children leaving state custody while still missing. The absence of a cost accounting is not an accident of research priorities. You cannot measure the cost of unreported cases in systems that do not record them. The void protects itself.
PART VII — TEXAS: TRANSPARENCY BY COURT ORDER
Texas has the nation's most detailed public accounting of children missing from foster care. It did not volunteer it. Texas reports because Texas was sued.
M.D. v. Abbott, filed in 2011 in the Southern District of Texas, produced a 2018 Fifth Circuit affirmance that the state had been "deliberately indifferent to a substantial risk of serious harm" to children in its permanent managing conservatorship. U.S. District Judge Janis Jack held the state in contempt three times over the following years; the third contempt order, in April 2024, carried a $100,000-per-day fine — stayed within days and later reversed by the Fifth Circuit, which also removed Judge Jack from the case in 2024. But a decade under a federal judge and court-appointed monitors forced into existence a statutorily mandated annual missing-from-care report (the Texas Family Code § 264.017/264.123 regime) that no other large state matches.
Read what mandated transparency looks like — Texas DFPS's own published numbers:
Fiscal year 2021: 45,870 children in state conservatorship;1,767 unique children went missing at some point in the year. By year's end, 1,495 (85 percent) had been located; 185 were still missing. And then the figures that only exist because a court made Texas count: 87 children exited state custody while still missing. Forty-three aged out at 18 while missing — the state's legal responsibility ended before the child was found. Forty-two had custody terminated by a court while still missing and under 18. Two children died while missing. Of the 1,632 children recovered that year, 119 (7.3 percent) were victimized while missing; 43 were identified as sex-trafficking victims.
Fiscal year 2023: 31,475 children in conservatorship; 1,164 unique children missing across 2,714 total episodes —41 percent went missing more than once, averaging 2.5 episodes each. At year's end, 109 were still missing; 39 exited custody while missing (15 aged out, 23 court-terminated, 1 died). Of 1,095 recovered children, 106 — 10 percent — were victimized while missing; 53 were identified as sex-trafficking victims. Separately, 386 children in substitute care were flagged as suspected or confirmed trafficking victims that year.
The federal auditors' view of the same state: 11,042 missing episodes over thirty months, an average of 21 days per episode and three episodes per missing child, 432 Texas foster children still missing on December 31, 2020 — and, in the detail that says everything about data infrastructure, Texas could not produce dates of birth for 614 of its own missing-episode records.
And the monitors' record shows what the numbers can't. In 2022, at The Refuge in Bastrop — a state-contracted shelter specifically for child sex-trafficking survivors— staff were alleged to have trafficked and abused the residents themselves: seven girls, ages 11 to 17, nine alleged perpetrators. Court monitors found "ample evidence" of sexual exploitation and documented girls repeatedly escaping the facility and "returning to trafficking multiple times." The state suspended the license on an emergency basis on March 11, 2022; Judge Jack sought a federal criminal probe weeks later. The same monitors documented Texas's "children without placement" — children the system had nowhere to put, housed in offices, hotels, and churches, where monitors found children becoming pregnant and being trafficked, concluding DFPS frequently failed to protect children from serious risk in those settings.
Sit with the category Texas's court-ordered reports created: children who exit state custody while still missing. In two reported fiscal years alone, 126 Texas children left the state's legal responsibility without the state ever having found them — 58 of them by aging out, meaning the system's obligation expired on a child whose whereabouts it did not know. Three died while missing. No federal database has a field for any of this. It is knowable in Texas for exactly one reason: a court required the state to write it down. Whatever the equivalent numbers are in the other forty-nine states, they exist as facts. They simply do not exist as records.
Texas is not the worst state in this story. Texas is the measured state. Its numbers are grim because they exist. Which brings us to the state where they don't.
PART VIII — NEVADA: THE INVERSE CASE
Project Milk Carton is a Nevada organization, and we hold our own state to the light first. Nevada is Texas's photographic negative: no court order, no mandated report, no numbers.
Start with what the federal auditors logged:3,328 missing episodes among 11,601 Nevada foster children in thirty months — a 6 percent missing rate, with an average of five missing episodes per missing child, among the worst repeat rates in the nation. Forty-four Nevada foster children were still missing on December 31, 2020. And in the federal NCMEC-reporting sample, Nevada failed four out of four — every sampled Nevada episode was never reported to NCMEC. No state drew a worse hand in that appendix.
Now look for Nevada's own accounting. The state Division of Child and Family Services publishes a Data Book running past 200 pages, current through the third quarter of state fiscal year 2024. We searched it. "Runaway" appears as a single point-in-time placement row:34 children statewide on June 30, 2023(27 of them in Clark County); 39 on March 31, 2024. That is the entirety of Nevada's public reporting on missing foster children. No annual missing counts. No episode data. No recovery data. No trafficking-screening data. No 24-hour-compliance data. No Nevada-specific OIG audit exists. No state statute mandates a Texas-style annual report. A state whose federal sample failure rate was 100 percent publishes, on this subject, effectively nothing.
The context that does exist is not reassuring. A 2022 Nevada legislative audit of DCFS foster homes found 33 percent with health or safety deficiencies and roughly 79 percent of placements carrying at least one regulatory violation — and contained nothing at all on missing children. The controlling federal case, Henry A. v. Willden (9th Cir. 2012), established Nevada foster children's constitutional right to safety and produced a $2.075 million settlement — a failure-to-protect precedent for Clark County, though not a missing-children case. On the specific subject of children who vanish from Nevada's foster system, there has never been an audit, a lawsuit, or a report.
Our own data completes the picture. Project Milk Carton's NCMEC-fed database carries136 active missing-child cases in Nevada— average age 9.3 years, 33 of them under 13. Our federal funding correlation shows$685.4 million in Title IV-E-correlated foster funding flowing through the state, alongside 269 recorded trafficking offenses, 6,496 child-maltreatment victims, and 21 child fatalities in 2023. Two-thirds of a billion federal dollars running through a foster system that publishes zero recurring public accounting of the children who go missing from it.
Put the Nevada pieces in one frame: a federal sample in which every pulled episode went unreported; a repeat rate of five episodes per missing child, meaning the same children cycle out of placements again and again; a public data book whose only answer to "how many children went missing this year" is a one-day headcount in the thirties; and no external mechanism — audit, statute, or lawsuit — that has ever asked the question. Under the OIG's thirty-month count, Nevada averaged roughly110 missing episodes a month. Under Nevada's own publishing practice, the public sees a number smaller than that for the entire state on a single day, once a quarter. The gap between those two numbers is not a rounding error. It is the difference between a system being watched and a system reporting on itself.
Transparency, this pair of case studies shows, tracks litigation — not law. Texas counts because a judge made it count. Nevada doesn't because no one ever has.
PART IX — GEORGIA, CONNECTICUT, KANSAS, MARYLAND: THE PATTERN REPEATS
Georgia: the Senate found what the state didn't report. A bipartisan, 13-month probe by the Senate Human Rights Subcommittee under Chairman Jon Ossoff obtained NCMEC's analysis of Georgia:1,790 children in DFCS care were reported missing between 2018 and 2022 — more than 2,400 reports — and 410 of them were likely victims of child sex trafficking. At the November 6, 2023, hearing, NCMEC's Dr. Samantha Sahl told the subcommittee: "We know we have an urgent issue when children feel better on the streets or with a trafficker than they do in their foster care placements. "The final report landed in April 2024. No federal penalty followed. It took a Senate subcommittee to produce a count one statute had required and another audit had shown missing.
Investigation data visualization
Connecticut: audited into the open. The state's own Auditors of Public Accounts, in findings released June 25, 2025, reported that runaway incidents rose94 percent from 2020 to 2023; those 606 children went missing a combined 3,736 times; that96 percent of incidents lacked documentation of police-notification decisions; and that94 percent of cases had no documented sex-trafficking screening upon return. In one case, the auditors wrote the sentence that this entire investigation orbits: "DCF does not know where the child is." The federal sample had already flagged Connecticut — its missing children averaged seven episodes each, the worst repeat rate in the OIG table, and both sampled Connecticut episodes were never reported to NCMEC.
Kansas: the disclosure that took a hearing. The October 2017 episode that opens this article — roughly seventy children missing statewide, three sisters gone for weeks without the agency head's knowledge, found with a 48-year-old man who was arrested — unfolded a full year after the 24-hour rule took effect. Kansas's 7 percent missing rate later tied for worst in the nation in the federal snapshot. (The 2017 figures come from contemporaneous, concordant news coverage of the legislative hearings; no federal audit reaches back to them.)
Maryland: the refusal, present tense. In May 2026 — this year — Maryland officials declined three separate requests to disclose how many foster children are currently missing from care, amid active trafficking concerns reported by WJLA. Maryland's line in the federal table: a 7 percent missing rate, tied worst, with missing children gone an average of 52 days. Ten years after the reporting mandate, a state's answer to "how many of your foster children are missing" can still be: we won't say.
Four states, four mechanisms of visibility — a Senate probe, a state audit, a legislative hearing, a press inquiry refused. What none of them are: the federal enforcement system working. Every count in this section exists because someone outside the child-welfare agency forced it out.
PART X — WHAT THE PATTERN MEANS
Five findings, from the documents:
1. A mandate without a mechanism is a suggestion. The 24-hour rule has no scored CFSR item, no eligibility-review hook, no litigation track record, and no assessed penalty in nine years — against a federally measured 69 percent failure rate. The only state ever to lose a dollar under the entire CFSR machinery lost $935,712 over something else entirely.
2. The data void is load bearing. No federal system counts these children. Four states couldn't produce data when the federal government asked; Rhode Island's was unusable; New York filtered its own; Maryland now declines to answer at all. Non-reporting is not a paperwork failure at the edge of the system. It is the thing that keeps the failure unmeasurable — and therefore unactionable.
3. Transparency tracks litigation, not law. Texas, a decade under a federal judge and court monitors, publishes the nation's most detailed missing-from-care accounting — down to the children who die while missing. Nevada, never audited and never sued on this subject, publishes a single point-in-time row. Both states operate under the same federal statute. The statute is not what makes the difference.
4. The stakes are the ones sworn to under oath. Seventy-eight percent of the missing-child cases NCMEC works are children missing from foster or state care, and by NCMEC's own sworn testimony roughly one in five of those children is a likely sex-trafficking victim. Every unreported episode is a child invisible to NCIC, to NCMEC's analysts, and to the public — during the hours that matter most.
5. The invisibility reaches all the way to the public record. Project Milk Carton operates one of the larger public-facing missing-children databases in the country: 2,851 active cases, fed from NCMEC's public data. Foster-care status is discernible in effectively none of them — seven free-text mentions across the entire database — because the public national data does not carry it. We checked our own house and found the same void the auditors found. The children the 2014 law was written to make visible cannot be seen in the very data built to make children visible. That is the finding, end to end.
What the record itself points to
This is a transparency organization; we present what the documents show. And what they show is that every piece of a functioning system already exists somewhere — just never in the same place. The statute exists (federal law, since 2014). The measurement method exists (OIG built it in 2022 — then used it once). The penalty machinery exists (the CFSR — which simply does not score this statute). The reporting model exists (Texas publishes it annually — under court order). The reconciliation is straightforward to describe: a recurring federal count; missing-episode data in AFCARS; the 24-hour rule as a scored compliance item; state-level public reporting on the Texas model, without requiring a decade of litigation first. Congress came within a conference committee of even studying the problem in 2024, on a 408–7 House vote. The record does not show any of it happening. It shows guidance, webinars, and a closed file.
The pending NEICE mandate is often presented as a solution. NEICE, the National Electronic Interstate Compact Enterprise, is designed to modernize ICPC transmission and reduce paper-based delays. Digital transmission matters, and full implementation is better than fax machines, mail bins, and scattered local processes. But digitizing a broken process is not the same as fixing it. If the underlying compact still lacks enforcement, penalties, staffing requirements, family remedies, and real-time accountability, then NEICE may simply move files more efficiently into the same queues. A digital bottleneck is still a bottleneck.
The ICPC crisis should force a much bigger question: why does the system make it so hard to place children with family while making it so easy to keep them in paid care? The answer is not one villain. It is design failure. It is legacy law. It is federalism without accountability. It is underfunded administration. It is risk avoidance. It is provider economics. It is the absence of a private right of action. It is decades of reform language without enforcement teeth. It is the oldest story in child welfare: the state says family matters, then builds a process that treats family as a paperwork problem.
A serious reform agenda would begin with enforceable deadlines. The 60-day home study requirement should not be a polite suggestion. If a receiving state misses the deadline without documented good cause, there should be automatic escalation, public reporting, and financial consequence. If the delay continues, relatives should have a direct administrative appeal and a fast-track court remedy. A compact that can stop a child from moving should also give the child and relative a way to force the government to act.
Second, the ICPC should include a child-centered emergency placement pathway for vetted relatives. When a relative has already passed background checks, has an existing relationship with the child, and meets basic safety criteria, the law should allow provisional placement with supervision while the full home study is completed. The system already takes risks by placing children with strangers in emergency settings. It should not pretend that relatives are uniquely dangerous simply because they live across a state line.
Third, Congress and the states must fund ICPC offices like critical infrastructure. A state with thousands of foster care cases cannot run interstate placement through one or two overburdened staff members and then act shocked when children wait. Staffing ratios should be published. Caseloads should be published. Average processing times should be published. Returned packets should be tracked. Clock restarts should be tracked. Every state should have a public ICPC dashboard showing how many children are waiting, how long they have waited, and where the file is stuck.
Fourth, the law must close the clock manipulation loophole. The 60-day period should not begin only after the receiving state accepts a completed referral while the sending state can take weeks or months to prepare it. There should be deadlines on both sides. The sending state should have a fixed period to identify relatives, prepare the referral, and transmit the packet. The receiving state should have a fixed period to accept, reject, or request specific missing information. If a packet is returned as incomplete, the reason should be documented, time-limited, and reviewable.
Fifth, ICPC delays should trigger placement cost review. If a child remains in congregate care or high-cost foster care solely because of interstate paperwork delay, the state responsible for the delay should bear financial consequence. Right now, the cost of delay is diffused across systems, while the child pays emotionally. Accountability should follow the bottleneck. If a state’s failure to process a home study causes months of unnecessary paid placement, that failure should show up in the budget.
Sixth, federal child welfare reviews should include ICPC compliance as a core performance measure. It is absurd to review state child welfare performance while allowing interstate family placement delays to remain a side issue. If a state cannot move children to safe relatives across state lines within legal timelines, that is a permanency failure. It should affect federal oversight, corrective action plans, and funding scrutiny.
Finally, families need rights, not just promises. Relatives who step forward should receive notice, status updates, reasons for delay, and access to review. Children should have counsel empowered to challenge ICPC delay when it harms permanency. Parents should be able to argue that the state cannot move toward termination while kinship placement is delayed by government inaction. No child should lose family because two state agencies could not move paperwork on time.
The ICPC began as a protective reform. That origin matters because it explains why the system has been so hard to challenge. Nobody wants to be accused of weakening safeguards for children. But after sixty-six years, the question is no longer whether children need protection across state lines. Of course they do. The question is whether the existing compact protects them from harm or creates a new kind of harm through delay, drift, and family separation. A child left in congregate care for six months while a safe relative waits is not being protected by paperwork. That child is being failed by it.
The phrase “fall through the cracks” is too soft for what is happening here. Children are not falling through cracks. They are being held in a pipeline with known failure points, known delays, known staffing shortages, known enforcement gaps, and known financial consequences. When a system knows where children get stuck and still does not build a release valve, the problem is not accidental. It is tolerated.
This is why the ICPC crisis deserves national attention. It is not as visible as a courtroom termination. It is not as emotionally packaged as an adoption ceremony. It is not as politically explosive as a removal caught on video. But it may be one of the most consequential hidden mechanisms in the child welfare system because it decides whether family can actually function as family when state lines are involved. In a mobile country where relatives often live in different states, a slow interstate placement system is a family-separation engine.
The children caught in this engine do not need another conference panel. They do not need another best-practices memo. They do not need another voluntary compact reform stalled for twenty years. They need enforceable timelines, funded staffing, transparent dashboards, provisional kinship pathways, legal remedies, and consequences for states that ignore deadlines. They need a system that treats family placement as urgent, not optional.
The ICPC was built to stop children from being sent across state lines without protection. But in 2026, the greater danger is that children are being kept from family without accountability. A compact designed to protect children has become a maze that delays them. A law built for paper files is governing digital-age families. A process created to ensure safety is too often producing separation.
And the most damning fact remains the simplest: when states miss the deadline, nothing happens to the states. Something happens only to the child.
If you enjoyed this work and feel encouraged by independent investigations like this, please consider becoming a paid subscriber to The Constitutional Republic on Substack or donate at
Project Milk Carton Inc. Your gift keeps our site alive, spotlighting high-risk children!
Your support helps us continue:
investigative reporting
public records research
educational transparency projects
community outreach
and future investigations like this series
Most importantly, please share this article.
Independent journalism survives when ordinary people help spread information powerful institutions would prefer remain ignored, misunderstood, or buried beneath complexity.
Because silence protects systems.
Informed communities protect children.
ALL paid subscriptions are tax deductible Project Milk Carton | 501(c)(3) | EIN: 33-1323547
A NOTE ON SOURCES AND METHOD
This investigation was researched on July 19, 2026, through four parallel verified-source sweeps — statute and enforcement history; federal audit records; NCMEC and trafficking data; state records, litigation, and oversight — plus live queries of Project Milk Carton's own civicops database (NCMEC-fed missing-children data; federal foster-funding correlations). Every figure above carries its source in the text. The full primary-source ledger:
Statute and regulation:42 U.S.C. § 671(a)(35) and (a)(9)(C) (uscode.house.gov; law.cornell.edu) · Public Law 113-183, §§ 101 and 104 (congress.gov) · Public Law 117-348, §§ 136–137 (congress.gov) · 45 C.F.R. §§ 1355.30, 1355.34, 1355.36, 1356.71 · 34 U.S.C. §§ 41307–41308.
HHS Office of Inspector General:A-07-20-06095 (May 19, 2022) · A-07-21-06102 (March 2, 2023) · A-07-21-06104 (May 12, 2023) · OEI-07-19-00371 (June 30, 2022) — all at oig.hhs.gov; figures verified against the report PDFs.
ACF / federal data:ACYF-CB-PI-14-06 (Nov. 21, 2014) · ACF-IM-22-01 (Dec. 1, 2022) · AFCARS Report #30.
GAO: GAO-11-444 · GAO-03-809 · documented absence of any GAO report on missing-from-care counts.
NCMEC: Impact page (2025 data, fetched live July 19, 2026) · "2024 in Numbers" · sworn testimony of John F. Clark (Mar. 16, 2017, gov info transcript), Michelle DeLaune (written, Feb. 6, 2024), and John Snow (written, Dec. 2025).
Congress / CRS:S. 1146 / H.R. 2426 (118th Congress) · CRS R42794 and R43757 · Senate Human Rights Subcommittee releases (Nov. 6, 2023; Apr. 2024).
Courts: M.D. v. Abbott, No. 2:11-cv-00084 (S.D. Tex.); 5th Cir. No. 18-40057 · Idaho, HHS DAB No. 3110 (2023) ·Henry A. v. Willden, 678 F.3d 991 (9th Cir. 2012) ·Jonathan R. v. Morrisey, 4th Cir. (June 4, 2026) ·Wyatt B. v. Kotek(D. Or., settled 2024).
States: Texas DFPS FY2021 and FY2023 missing-from-conservatorship reports (dfps.texas.gov) · Nevada DCFS Data Book, SFY2024 Q3 · Connecticut Auditors of Public Accounts (June 2025) · WJLA reporting on Maryland (May 2026) · contemporaneous Kansas hearing coverage (Oct. 2017).
Peer-reviewed: Latzman et al., Children and Youth Services Review98 (2019): 113–124 · OPRE Report 2020-143.
Project Milk Carton house data: civicops missing-children database (2,851 active cases; queried July 19, 2026) and federal foster-funding correlation tools.
Editorial integrity note
One statistic you may have seen elsewhere does not appear in this article: the claim that "88 percent" of trafficked children were in the care of social services. We could not trace it to NCMEC or to sworn testimony — only 67 percent (2013 data, testimony summary) and 86 percent (2016 data, sworn) survive primary-source verification — so we did not use it. NCMEC's "likely sex-trafficking victim" classification is quoted as NCMEC defines it, and we have not mixed the two different denominators NCMEC has used over the years. The Kansas 2017 figures rest on strong concordant contemporaneous reporting rather than a federal document, and are labeled accordingly. Where we assert that something does not exist — a penalty, a federal count, a GAO report — we conducted and documented the search, and we state the finding as an absence, not an assumption.
Project Milk Carton is a Nevada 501(c)(3) dedicated to child welfare transparency and missing children awareness through public data, research, and education. Our missing-children database, state data hubs, and investigation archive are available at projectmilkcarton.org.
Project Milk Carton
501(c)(3) Nonprofit • projectmilkcarton.org • t.me/ProjectMilkCarton