The Boondoggle in the Basement
Do you know what the word boondoggle means?
It comes from the Boy Scouts. A boondoggle is a braided leather cord — a small useless craft project meant to keep your hands busy. In the 1930s, critics of the New Deal grabbed the word and used it as a slur for federal make-work programs: taxpayer money going to people doing nothing important, looking busy because someone had to look busy.
Ninety years later, in the basement of the CDC's Division of Reproductive Health, we have what is arguably the purest boondoggle in the modern history of American public administration. Except this time, it isn't being run by liberals defending public investment from conservative critics. It's being run by conservatives. And the administration that built it would prefer you never noticed it exists.
Let me describe it for you.
Imagine you are one of the roughly forty people still employed at the Division of Reproductive Health — not one of the 110 who got the Reduction in Force notice on April 1, 2025. You are one of the ones who survived. Officially, your job still exists. Officially, you still have a portfolio. Your badge still works. Your paycheck still hits your account every two weeks.
But the team you ran point with — the epidemiologists who cleaned the data, the program officers who managed the state grants, the senior staff who knew where the bodies were buried (sometimes literally) — they are on administrative leave or have moved on. The work you were trained to do, that you came to government to do, has been quietly liquidated around you.
You still get emails. You still attend meetings. Sometimes the meetings are about the work that no longer exists. Sometimes the meetings are about how to describe the work that no longer exists in language that won't trigger the new restrictions on what federal employees are allowed to say. You cannot use the word "women" in certain documents now. You cannot use the word "Black." "Disability" is on the list. "Tribal" is on the list. "Trauma" is on the list. There are nearly two hundred words on the list.
You can technically log into the data systems. You can technically file the paperwork. In November 2025, somebody at the agency filed an OMB extension request for PRAMS — the surveillance system whose entire operational team had been on leave for seven months. The paperwork went through. In March 2026, somebody filed a revision cutting 1,495 hours of data-collection burden. That paperwork also went through. Nobody on your floor seems to know who is going to do the work the paperwork now authorizes.
This is not the kind of boondoggle the New Deal critics meant. They imagined waste in the form of activity — federal money going to people doing pointless work. This is the inverse. This is federal money going to people who have been deliberately prevented from doing the work they were hired to do, while the bureaucratic shell that documents the work continues to file the paperwork claiming it is being done.
It is The Office, if Michael Scott had been disappeared by executive order and Toby was the only one left in the branch — expected to keep doing HR for a company that no longer manufactures paper but is still legally registered to manufacture paper, and there are still trucks pulling up to the loading dock, and the trucks are empty.
The Timeline: How Thirteen Months Disappeared a Data System
The destruction of PRAMS did not arrive in a single announcement. It arrived as a sequence of small actions, each one defensible on its own terms, each one moving the system one notch closer to silence. Read the calendar straight through and the pattern emerges — not a budget cut, not a reorganization, but a methodical decoupling of paperwork from people, of authority from capacity, of the apparatus that asks the questions from the apparatus that answers them.
The opening move came on January 20, 2025, the first day of the second Trump administration, in two executive orders. Executive Order 14168 — Defending Women From Gender Ideology Extremism and Restoring Biological Truth to the Federal Government — defined sex as "an individual's immutable biological classification as either male or female" and stated that "'Sex' is not a synonym for and does not include the concept of 'gender identity.'" Executive Order 14151 — Ending Radical and Wasteful Government DEI Programs and Preferencing — set a March 20, 2025 deadline for agency action against any program touching diversity, equity, inclusion, or accessibility. Neither order named PRAMS. Neither had to. The Office of Personnel Management memorandum that followed on January 29 gave federal agencies until 5 p.m. on January 31 to "take down all outward facing media" of DEIA offices and to "end all agency programs that use taxpayer money to promote or reflect gender ideology." Programs that asked women about their pregnancies, their depression, their experiences with partner violence, their access to contraception — programs, in other words, that produced the structural evidence the executive orders were designed to suppress — became targets without ever being named.
By early February, the CDC.gov pages began disappearing in batches. The PRAMS portal went dark. On March 3, 2025, STAT News reporter Anil Oza published "Paused CDC program rattles maternal, infant health researchers." Marian Jarlenski, a health policy researcher at the University of Pittsburgh, told STAT: "If we can't see this information, we cannot develop public health strategies to improve our maternal morbidity and mortality in the United States. I don't know how else to say it, this data system is needed." Researchers reported that they had "suddenly discovered they were locked out of the database, without any notice."
Congress noticed. On March 12, 2025, Senator Cory Booker and Representative Robin Kelly led a letter to Acting CDC Director Susan Monarez signed by seventeen additional lawmakers — Senators Tammy Duckworth, Elizabeth Warren, Amy Klobuchar, and Ron Wyden, alongside Representatives Alma Adams, Yvette Clarke, Bonnie Watson Coleman, Mike Quigley, Terri Sewell, Jake Auchincloss, Diana DeGette, Nydia Velázquez, LaMonica McIver, Kelly Morrison, and Eleanor Holmes Norton. The cosigner roster, concentrated in the Black Maternal Health Caucus, asked CDC to explain the pause and restore the data. CDC did not respond. The non-response — itself an action — would become the dominant grammar of the next thirteen months.
Then came April 1. On that afternoon, CDC epidemiologist Jennifer Bombard sent an email to state PRAMS coordinators: "the entire CDC PRAMS team, including myself, has received the Reduction in Force (RIF) notice from HHS today." The same day, HHS executed a roughly 2,400-person CDC RIF that eliminated about two-thirds of the Division of Reproductive Health's roughly 165 employees and contractors. The MEC contraceptive guidelines team in the Women's Health and Fertility Branch — authors of the third-most-viewed MMWR publication of the prior twelve months — was eliminated. The six-person Assisted Reproductive Technology surveillance team was eliminated, even though the Westat data-collection contract supporting it runs through October 14, 2027. The infrastructure supporting ERASE MM was eliminated. The directorship of Wanda D. Barfield, MD, MPH — newly elected to the National Academy of Medicine in 2025 — was disrupted. Lee Warner, PhD, chief of the Women's Health and Fertility Branch, retired the same month, his branch dissolved beneath him. CDC's statement to STAT named the cause: "Data collection for PRAMS was previously paused, a move the CDC said was temporary, to ensure compliance with President Trump's executive orders." On April 16, Booker and Kelly sent a second letter, this time directly to Secretary Robert F. Kennedy Jr. The second letter, like the first, went unanswered.
Through the spring and summer of 2025, the state-level fracture began. Mississippi paused data collection for most of 2025. By the time Mississippi rejoined in early 2026, the mothers who had given birth in February and March of 2025 were no longer eligible to be surveyed. That cohort — the women whose pregnancies overlapped with the federal RIF, in the state with the highest infant mortality rate in the country — is gone. Permanently. Arkansas, Kentucky, Georgia, Alaska, New Mexico, Oklahoma, and Kansas continued data collection. Ohio and North Carolina, which had built their own pregnancy surveys in 2016 and 2021 respectively, kept running them. California's CMQCC continued its parallel state surveillance. The well-resourced states stayed standing. The under-resourced states began to lose entire birth-year cohorts.
On June 12, 2025, HHS reversed itself on more than 400 of the April 1 RIFs, restoring positions largely concentrated in the National Center for HIV, Viral Hepatitis, STD, and TB Prevention. The PRAMS team was not among those reinstated. The selective reinstatement was its own form of statement: the executive branch had decided which categories of public health surveillance could return and which could not.
On July 31, 2025, the Senate Appropriations Committee marked up the FY26 Labor-HHS bill by a 26-3 vote. The bill funded PRAMS. It included a $53 million increase across CDC and NIH for maternal health. It required a report within 90 days on "barriers to effective and consistent data collection," and it included statutory language requiring HHS to "support staffing levels necessary to fulfill its statutory responsibilities" and to submit a reorganization plan "not less than 60 days prior to initiating the execution of any reorganization moving functions… carried out by the Centers for Disease Control and Prevention to another component of the Department of Health and Human Services." Congress, in other words, said it expected PRAMS to be funded, staffed, and protected from administrative dismantling. The administration had no statutory reply, because none was required. The administration simply did not rehire.
The longest full government shutdown in U.S. history ran from October 1 to November 11, 2025 — 42 days. When the lights came back on, the PRAMS team remained absent. On November 21, 2025, the CDC published a Federal Register notice — FR Doc. 2025-20583, Docket CDC-2025-0750 — requesting a three-year OMB extension of PRAMS authority. The comment period closed January 20, 2026, with 419 comments received. The Association of Maternal & Child Health Programs filed formal comments on January 7, 2026, observing that PRAMS "covers approximately 81 percent of U.S. live births. For many states and jurisdictions, PRAMS is the only source of data on critical maternal and infant health indicators." The Commonwealth Fund's explainer by Jamie Daw, Heidi L. Allen, Jess Maksut, and Laurie C. Zephyrin dropped the same day the comment period closed. Kristin Rankin at the University of Illinois Chicago submitted comments documenting that 76 percent of PRAMS site administrators reported they did not have the time or resources to navigate the PRAMS weighting process for 2024 data without CDC support, and 97 percent reported requiring CDC technical assistance for cleaning and weighting future data files. The agency was asking permission to operate a program no one inside the agency remained to operate.
On February 3, 2026, the Consolidated Appropriations Act, 2026 — P.L. 119-75 — was signed into law. PRAMS was funded. The 90-day report on data collection barriers was required by statute. As of late May 2026, the PRAMS team has not been rehired.
On March 9, 2026, CDC published the 30-day follow-on Federal Register notice — FR Doc. 2026-04565, Agency Docket 30Day-26-1273. The category quietly changed. What had been labeled an "Extension" in November was reclassified as a "Revision." The 1,495-hour annual burden reduction was reaffirmed: 1,395 hours attributed to "call back surveys" and 100 hours to "cognitive and field testing," because "no new questions or supplemental modules are anticipated during the approval window." The agency did not narratively explain the reclassification. The category change is not cosmetic. A "Revision" reserves administrative authority to alter the instrument — to drop modules, to remove race and ethnicity questions, to eliminate the pregnancy intention or intimate partner violence items — without triggering a fresh 60-day public comment period. The callback survey is not a completed component. It is the methodological mechanism that brings low-income, less-educated, and non-white respondents into the sample. Removing it, whether by line item or by attrition, produces a dataset that systematically over-represents the women whose outcomes were already best understood.
On April 17, 2026, during Black Maternal Health Week, Representative Summer Lee of Pennsylvania confronted Secretary Kennedy at the House Education and Workforce hearing titled "Examining the Policies and Priorities of the Department of Health and Human Services." Lee: "Your agency told programs to remove a list of nearly 200 words and phrases from their funding applications, including the word 'Black.' How are we going to solve the Black maternal mortality crisis if we cannot say 'Black'?" Kennedy's response, in the Philadelphia Inquirer's characterization, was "a mumbo jumbo rabbit hole of claiming that DEI is to blame for racial divisiveness."
On April 30, 2026, the CDC PRAMS state grants expired. State coordinators describe ad hoc extensions; HHS has issued no public timeline for renewal.
On May 8, 2026, Anna Hu and Sophia Paffenroth published "Mississippi leads the nation on infant deaths. But gaps in state data collection may make those deaths harder to track" in Mississippi Today. An HHS spokesperson provided the now-familiar non-answer: "HHS is committed to optimizing maternal and infant health outcomes and will share PRAMS data at a later time." HHS would not elaborate. The same construction — "at a later time" — had been used to defer release of the 2023 ERASE MM aggregate data and Maternal Mortality Review Information Application outputs. The phrase is its own genre.
That is the timeline. Read straight through, the cadence becomes legible. The executive orders defined the predicate. The website takedowns established the practice. The April 1 RIF eliminated the people. The congressional letters went unanswered. The appropriations bill funded a workforce that did not exist. The Federal Register notices kept the paperwork in motion while the operational team had been gone for a year. The category quietly changed from "Extension" to "Revision." The state grants expired. The 2023 data set, finished and ready, has not been released. Every step was lawful on its surface. Every step required an active choice by an identifiable office. And every step moved in the same direction.
Part IIIThe Parallel Erasures: The Rest of the Building
PRAMS is one room. The building is larger, and most of it has been emptied by the same method.
Begin in the same corridor. On April 1, 2025, the same RIF that eliminated the PRAMS team eliminated the MEC contraceptive guidelines team within CDC's Women's Health and Fertility Branch. The U.S. Medical Eligibility Criteria for Contraceptive Use is the document obstetrician-gynecologists, family physicians, and nurse practitioners consult to decide which contraceptive method is safe for a patient with hypertension, with a history of stroke, with active lupus, with a recent venous thromboembolism. Dr. Stella Dantas, the president of the American College of Obstetricians and Gynecologists, characterized the MEC as a guideline ob-gyns "turn to every day." The CDC's Morbidity and Mortality Weekly Report ranked the MEC as the third-most-viewed MMWR publication of the prior twelve months, with more than half a million readers. The November 2024 eighth edition may be the last federally maintained version. The team that updated it for evolving clinical evidence does not exist. The same April 1 RIF also eliminated the six-person Assisted Reproductive Technology surveillance team, the federal program that tracks every IVF cycle initiated in the United States. The Westat data-collection contract runs through October 14, 2027. The data continues to be collected. There is no federal analytic team to clean, weight, or publish it. The pattern is now visible: the input remains, the output stops, the system is functional on paper and inert in practice.
Walk further down the corridor. ERASE MM — Enhancing Reviews and Surveillance to Eliminate Maternal Mortality — is the CDC's five-year, $118 million investment in the Maternal Mortality Review Committees that operate in 46 states and 6 territories. The MMRCs are the bodies that adjudicate, case by case, whether a pregnancy-related death was preventable and what contributed to it. Their findings drive every modern recommendation about hemorrhage protocols, hypertensive crisis management, postpartum mental health screening, and intimate partner violence intervention. The president's FY26 budget proposed eliminating the entire National Center for Chronic Disease Prevention and Health Promotion — a $1.4 billion cut that would have ended PRAMS, ERASE MM, and the Perinatal Quality Collaboratives in a single line item. Congress rejected that proposal. The administration has continued to act as though it had not been rejected. Texas and Florida had already opted out of sharing data with ERASE MM before the cuts; Texas stopped sharing maternal mortality data with the CDC after restricting abortion access, as ProPublica documented. The states whose maternal mortality rates rose fastest after Dobbs are the states whose data is now most opaque.
Now leave reproductive health and walk to the equity offices. The CDC Office of Health Equity, the HHS Office of Minority Health, the CMS Office of Minority Health, the FDA Office of Minority Health and Health Equity, and the HRSA Office of Health Equity — five federal offices that produced the demographic analyses that made structural racism in clinical care visible at all — were functionally eliminated through the April 1, 2025 RIF and the subsequent website takedowns. The CMS and HHS Office of Minority Health eliminations may be illegal under the Affordable Care Act, which statutorily authorized them; that legal question has not been adjudicated. The CDC "Racism and Health" portal, launched by then-Director Rochelle Walensky in April 2021, is archived only on RestoredCDC.org with a January 6, 2025 snapshot date. The portal collected the peer-reviewed evidence that structural racism is itself a determinant of health. It is no longer publicly accessible through any federal source.
Up the next floor, the NIH. By April 10, 2026, the Congressional Research Service product IF13131 documented 1,392 NIH grant terminations totaling approximately $1.7 billion. The peer-reviewed analysis by Liu, Kadakia, Patel, and Krumholz in JAMA on May 8, 2025 had captured an earlier window: 694 grants worth $1.81 billion terminated between February 28 and April 8, 2025. The same paper documented that the National Institute on Minority Health and Health Disparities lost 29.6 percent of its previously active funding — $223.5 million across 77 grants. NIMHD had the highest proportional cut of any NIH institute. Justice Ketanji Brown Jackson, in her dissent from the Supreme Court's August 21, 2025 stay in the Massachusetts v. NIH litigation, observed that NIH had previously terminated fewer than six grants midstream between 2012 and January 20, 2025. The terminations were a category change in how the federal government related to its own grantmaking — a shift from administrative continuity to ideological enforcement. Morehouse School of Medicine, a historically Black institution and one of ten partners in NIH's maternal health project on pregnancy and postpartum care for Black women, lost a $2.9 million grant with $1.6 million remaining when it was cut. Morehouse President Dr. Valerie Montgomery Rice told WABE: "diversity, equity and inclusion, as it relates to health, is not a political term."
And the language itself. In March 2026, Representative Lauren Underwood reintroduced the Black Maternal Health Momnibus Act as the "Momnibus Act." The word "Black" appears once across the package. The bill's lead sponsor described the change as reflecting common usage; Angela Doyinsola Aina of the Black Mamas Matter Alliance and other advocates declined to endorse this version. The retitling did not happen in a vacuum. On December 5, 2025, a court filing in the Head Start litigation surfaced HHS's six-page list of nearly 200 words and phrases banned from federal grant applications. The list includes "accessible," "belong," "Black," "disability," "female," "Hispanic," "Latinx," "minority," "mental health," "Native American," "race," "racial," "racism," "trauma," "tribal," "transgendered," and "women." Senators Patty Murray and Bernie Sanders, in a December 18, 2025 letter to Secretary Kennedy, noted that many of the banned terms are explicitly required by the 2007 Head Start Act. The administration's reply was that the prohibition reflected the executive orders.
Stand back from the corridor map and the architecture is clear. The MEC team — eliminated. The ART team — eliminated. The ERASE MM-supporting staff — eliminated. The five federal equity offices — functionally eliminated. The NIH grants that funded Black maternal health research, structural racism research, and minority health disparities research — terminated by the thousands. The legislation that named the population most likely to die in childbirth — renamed to remove the name. The vocabulary that allowed federal grant applications to describe what the federal government had documented for half a century — proscribed in writing. Black women remain the only racial group in the United States whose pregnancy-related mortality rate did not decline in the most recent CDC analysis.
How Labora Is Surveilling Without PRAMS
What does maternal health surveillance look like when the gold-standard instrument is gone? It looks like a stack of proxies, each weaker than the lost original, each chosen because the data still flows. The point is not to replace PRAMS. The point is to refuse the silence the administration has manufactured, and to assemble, from what remains, a picture clear enough to keep clinical practice and policy moving while the federal apparatus is rebuilt or replaced.
The methodological pivot rests on a single validated finding. In July 2025, the Commonwealth Fund published Maternal Mortality in the United States, 2025 by Edward Berchick, David Radley, and Laurie Zephyrin. Their analysis demonstrated that state-level pregnancy-related mortality ratios between 2019 and 2023 broadly mirror state-level all-cause death rates for women ages 15 to 44, with a Pearson correlation of r = +0.59. Alabama, Mississippi, and Tennessee — the states with the highest pregnancy-related death ratios — also carry the highest overall death rates among reproductive-age women. California, Massachusetts, Minnesota, and Utah cluster at the low end on both. That correlation is the methodological license. When the pregnancy-specific signal is suppressed, the all-cause signal among women of reproductive age still tracks it closely enough to detect direction, magnitude, and geographic concentration of harm.
Labora's surveillance frame builds outward from there in six layers.
Layer 1 — All-cause mortality, women 15–44
All-cause mortality among women aged 15 to 44, stratified by state and by race and ethnicity, drawn from CDC WONDER's underlying-cause-of-death files. WONDER remains public as of May 2026. The data run through 2023. Deaths per 100,000 women in this age band, watched state by state and year over year, will detect the increase that miscoded or undercaptured pregnancy-related deaths produce — because a woman who dies of a pregnancy-related cardiomyopathy that is coded as I42 instead of O90.3 still shows up in the all-cause denominator. Building these dashboards now, while access is stable, is the immediate priority. The dashboard architecture itself is the insurance policy.
Layer 2 — Cause-specific spikes in non-O-coded categories
The cause-specific stratification of those all-cause deaths through WONDER's multiple-cause-of-death files. The categories to watch are cardiovascular disease (ICD-10 I00–I99), septicemia (A40–A41), suicide and self-harm (X60–X84), unintentional injury (V01–X59), and substance-related death (F10–F19, X40–X44). These are the bins that absorb pregnancy-related deaths when O-codes are not assigned. The signature of miscoding is mechanical: in a state with restricted abortion access, the O-code maternal mortality rate declines while one or more of these cause-specific rates among reproductive-age women rises in tandem. Linkage to birth certificate records via the multiple-cause-of-death file allows the analyst to flag deaths that occurred within 42 days or one year of a live birth even when the underlying cause is coded outside Chapter XV.
Layer 3 — Maternal ICU admission
Maternal ICU admission during delivery, captured on the U.S. Standard Certificate of Live Birth and reported through the National Vital Statistics System natality file. NCHS Data Brief 485 (December 2023) established the baseline: 1.4 admissions per 1,000 live births for mothers under 25 years, 6.5 per 1,000 for mothers 45 and older. Birth certificates are state-administered and cannot be eliminated without ending the legal registration of births itself. The maternal ICU admission rate is therefore one of the few maternal-acuity signals that the federal apparatus cannot quietly suppress. A year-over-year state-level tracker of this rate, stratified by maternal age and race, surfaces the deterioration that precedes mortality.
Layer 4 — NICU admission as sentinel
NICU admission as a sentinel of upstream maternal health. NCHS Data Brief 525 (March 2025) documented a 13 percent increase in NICU admission from 2016, when the rate stood at 8.7 percent of live births, to 2023, when it reached 9.8 percent. The largest increases — 11 percent among mothers aged 20 to 29 and 12 percent among mothers aged 30 to 39 — occurred in the age bands where the absolute denominator is largest, meaning the trend affects the majority of U.S. births. NICU admission is downstream of poor prenatal care, untreated preeclampsia, undetected gestational diabetes, and preterm labor — the conditions PRAMS measured upstream. The NICU rate is the bedside echo of the surveillance that has been turned off.
Layer 5 — Hospital utilization and severe maternal morbidity
Hospital utilization captured through the Healthcare Cost and Utilization Project — the HCUP Nationwide Inpatient Sample covering approximately 20 percent of U.S. discharges, the Kids' Inpatient Database, and HCUP Fast Stats on maternal hospitalizations — combined with Medicaid Transformed Medicaid Statistical Information System Analytic Files for the roughly 42 percent of births covered by Medicaid. Severe maternal morbidity is identified through ICD-10 codes capturing hemorrhage requiring transfusion, eclampsia, acute kidney injury, disseminated intravascular coagulation, and the other categories that constitute the CDC's SMM index. State-level workarounds layer on top of the federal data. The California Maternal Quality Care Collaborative — CMQCC, the model "post-PRAMS" system — has run continuous quality improvement and surveillance for two decades. Ohio's own pregnancy survey, fielded since 2016, and North Carolina's, fielded since 2021, demonstrate that state-built PRAMS-equivalents are operationally feasible when state political will exists. New York, Massachusetts, Illinois, and Washington can build versions of the same. Mississippi, Alabama, Louisiana, and Arkansas, given current state fiscal and political conditions, cannot. The bifurcation is not theoretical — Mississippi's February and March 2025 birth cohort has already disappeared.
Layer 6 — The money trail
CMS Medicare and Medicaid spending dashboards, state Medicaid expenditure reports, and hospital community benefit filings on Internal Revenue Service Form 990 Schedule H all generate financial traces of maternal complications. Pre-mortality maternal events — antepartum hemorrhage, severe preeclampsia, postpartum infection, peripartum cardiomyopathy — generate intense hospital spending. If the O-coded pregnancy-related mortality rate in a state declines while pregnancy-related hospital spending per delivery rises, the divergence is the flag. The money trail is harder to manipulate than the cause-of-death certificate, because the money was already paid.
The scorecard
Stacked together, the layers produce a quarterly state scorecard with six columns: change in all-cause mortality for women aged 15 to 44, change in maternal ICU admission rate, change in NICU admission rate, change in severe maternal morbidity rate, change in O-coded maternal mortality rate, and change in spending per delivery. A state where O-coded maternal mortality is declining while the other five columns are rising is a state where deaths are being miscoded or undercaptured. That divergence — the negative correlation where positive correlation should sit — is the Labora finding. It is reproducible. It is defensible. It is built entirely from data that remains public as of this writing. The first state-level scorecard release, covering California, New York, Mississippi, Texas, and Ohio as the pilot, is targeted for the third quarter of 2026, scaled to all fifty states by year-end. A methodology paper, with Eugene Declercq at Boston University, Marie Thoma at Maryland, and Jamie Daw at Columbia as candidate co-authors, will document the proxy stack and the validation logic in the peer-reviewed literature.
Three Scenarios for What Comes Next
Where does this end? Three scenarios bracket the plausible range. None is good. Two are recoverable. One is not.
The slow thaw
The OMB approves the three-year extension. Congress holds the line on appropriations through FY27 and FY28. A subset of the PRAMS team — perhaps eight to twelve former staff, perhaps recruited from outside CDC entirely — is rebuilt. The 2023 data set is released in late 2026 or early 2027, in stripped form, with some race-stratified analyses degraded by sample-size suppression. The 2024 cohort is partially reconstructed from state-held raw responses; the 2025 cohort is partially recovered, with permanent gaps where state collection paused. The cadence does not return to pre-2025 norms — annual data releases slip to biennial, supplemental modules contract from the Phase 8 catalog of opioid, disability, COVID, and SDOH instruments to a thinner standard core. The time series survives with documented breaks.
Field implication: PRAMS continues as a useful but methodologically wounded instrument, and the 2023–2025 window is treated as a single discontinuous segment requiring footnote in every downstream analysis.
Leading indicators that would confirm: a named PRAMS team lead hired by January 31, 2027; a public release schedule for the 2023 and 2024 data sets published by the same date; restoration of the public PRAMStat dashboard; reinstatement of the callback survey budget line that the March 9, 2026 reclassification eliminated.
What the field can do now: maintain the academic user network in publishable readiness so that the rehired team has external partners on day one; preserve the methodological documentation for Phases 1 through 9 in citable form so the institutional memory survives outside CDC.
The formal sunset
PRAMS is officially discontinued, or — more likely given the political costs of an explicit cancellation — replaced with a substantially shorter and less probing instrument under a new acronym. The OMB authority continues; the infrastructure of state-CDC data flow continues; the questions worth asking do not. The replacement instrument drops the pregnancy intention items, the intimate partner violence module, the postpartum depression screen, and the discrimination questions. It keeps smoking, breastfeeding initiation, and prenatal care timing — the items least politically costly to retain. The PRAMS shell becomes a vehicle for tracking the maternal behaviors the administration considers permissible to measure, and stops tracking the maternal conditions the administration considers permissible to ignore.
Field implication: the surveillance frame collapses from population-level maternal experience to a narrow set of preventive-care indicators that already overlap with existing HRSA Title V and Healthy People 2030 measures. The structural inequities PRAMS revealed become formally invisible at the federal level.
Leading indicators: a new Federal Register notice naming module removals; an OMB-approved instrument that does not include the standard PRAMS race and ethnicity stratification; a renamed program in CDC budget documents.
What the field can do now: build out the state-by-state independent surveillance coalition described in Part IV so that the questions PRAMS asked continue to be asked somewhere, even if not at the federal level; advance the legislative campaign to statutorily codify PRAMS — analogous to the codification of MMRC support under the Preventing Maternal Deaths Act of 2018, Public Law 115-344 — so that future instrument changes require an act of Congress rather than a Federal Register footnote.
Permanent capability loss
No team is rebuilt. The state collection apparatus, deprived of CDC technical assistance, decays past the threshold where restoration is technically feasible. The 76 percent of PRAMS site administrators who told Rankin they could not navigate the weighting process without CDC support stop trying. State Institutional Review Board approvals lapse. State data use agreements expire. The vendor relationships that supported mail, web, and telephone callback administration dissolve. The mid- to late-career epidemiologists with the methodological memory of Phases 1 through 9 take other positions, retire, or both. When a future administration attempts restoration, it confronts a starting point closer to 1987 than to 2024 — not a paused system but a dismantled one. A realistic rebuild timeline becomes 24 to 36 months from the restoration decision, and the analytic dark zone covers 2024 through the restoration year, with the time series broken in ways that affect every clinical guideline derived from PRAMS data for a generation.
Field implication: a generation of maternal health policy is built on data that does not exist, or built on proxy data with all of the bias and none of the depth of the original instrument.
Leading indicators: silence from state PRAMS coordinator positions as incumbents leave and are not replaced; explicit state withdrawals from the PRAMS network; OMB extension lapse without renewal; private contractor announcement consistent with the Project 2025 "public-private partnership" framework for CDC data infrastructure.
What the field can do now: treat the next twelve months as the window in which institutional memory can still be captured through oral history with former PRAMS team members; press for the statutory protections and the legal challenges that prevent infrastructure decay from hardening into infrastructure absence; build the parallel independent maternal surveillance entity, housed in an academic consortium of Columbia, Harvard, Minnesota, and Pittsburgh, that can hold the methodological standard until federal capacity returns.