Who is on the federal exclusion list?
Reference page · published 2026-11-18
Mostly nurses, doctors and other licensed health workers — and far more of them than most people expect. The list is kept by the Office of Inspector General at the Department of Health and Human Services, and on November 18, 2026 it held 84,224 entries: 80,678 individual people and 3,546 companies.
That is worth putting beside the ban most people find first. FDA's two debarment lists, counted on the same day, hold 152 and 63 people — 215 in total, and not one company. So the exclusion list carries roughly 390 times as many entries, and it is the only one of the three that names businesses at all.
Two things about it surprise almost everyone. The first is that nearly half the entries did not require a criminal conviction. The second is that the ban itself is narrower than its size suggests: it stops the government paying for your work, which is a serious thing to lose only if the government was paying for your work in the first place.
This page covers what the list is, what being on it actually forbids, which grounds put people there, how long it lasts, and the measurement that matters most for anyone reading our case records: none of the 25 criminal defendants documented in this archive is on it, and the reason is not luck.
Two different bans, two different lists
These are separate systems run by separate agencies under separate statutes, and they are easy to confuse because both are commonly called a federal ban.
| FDA debarment | OIG exclusion | |
|---|---|---|
| Who runs it | Food and Drug Administration | HHS Office of Inspector General |
| Statute | 21 U.S.C. 335a | 42 U.S.C. 1320a-7 |
| What it forbids | working for a company with a drug application | being paid for by a federal health program |
| Entries, Nov 18 2026 | 215 | 84,224 |
| Companies named | 0 | 3,546 |
| Usual length | permanent for 146 of 152 on the applications list | not less than five yearsat the baseline |
| Published as | a web page per list, plus a Federal Register order per person | a searchable database and a downloadable file |
The statute behind the exclusion list is short about the Secretary's duty. Four kinds of
conviction trigger it automatically: The Secretary shall exclude the following individuals and
entities from participation in any Federal health care program
. A longer list of grounds is
optional, and the statute uses the word may for those.
OIG publishes its own plain table of every ground with the minimum period beside it. That table is the single most useful document on the subject, and the grounds are quoted from it below.
What is actually on the list
These counts come from the file OIG publishes for download, read on November 18, 2026. It is one row per current exclusion, 18 columns wide. Every figure here was produced by a script that prints its checks before its results, and the file's own checksum is recorded in the sources.
| Ground, in OIG's words | Code | Entries | Share | Conviction needed? |
|---|---|---|---|---|
License revocation, suspension, or surrender. | 1128b4 | 33,463 | 39.7% | no |
Conviction of program-related crimes. | 1128a1 | 26,104 | 31.0% | yes |
| Conviction relating to patient abuse or neglect | 1128a2 | 8,127 | 9.6% | yes |
| Felony conviction relating to health care fraud | 1128a3 | 5,884 | 7.0% | yes |
Felony conviction relating to controlled substance. | 1128a4 | 3,592 | 4.3% | yes |
Default on health education loan or scholarship obligations. | 1128b14 | 2,164 | 2.6% | no |
Entities controlled by a sanctioned individual. | 1128b8 | 1,497 | 1.8% | no |
| All 15 remaining codes together | — | 3,393 | 4.0% | mixed |
| Total | 84,224 | 100% |
A few plain facts about the shape of it. The oldest entry is dated July 1, 1977. The newest is dated November 19, 2026 — tomorrow, as this page is written, which is worth knowing if you ever search the file and find a name whose exclusion has not started. There are 199 such entries. Names appear from 56 different state and territory codes. And waivers are vanishingly rare: 2 of the 84,224 entries carry one.
The 1977 date is not an accident of record-keeping. OIG's own history says the program began
that year: In 1977, in the Medicare-Medicaid Anti-Fraud and Abuse Amendments, Public Law 95-142
(now codified at section 1128 of the Act), Congress first mandated the exclusion of physicians and
other practitioners convicted of program-related crimes from participation in Medicare and
Medicaid.
The agency's answer page puts it slightly differently and says both halves:
OIG has been implementing exclusions since about 1981, but the Department of Health & Human
Services first began imposing them in 1977.
The oldest row in the file is a medical practice
entry effective July 1, 1977, so the file and the history agree.
Nearly half the list was never convicted of anything
This is the part that changes how the list should be read. Sorting all 84,224 entries by whether their statutory ground requires a criminal conviction:
| Does the ground require a conviction? | Entries | Share |
|---|---|---|
| Yes — one of the seven conviction grounds | 45,044 | 53.5% |
| No | 39,180 | 46.5% |
The single biggest ground on the whole list is not a crime at all. It is a license problem: an
individual whose license to provide health care has been revoked or suspended by any State
licensing authority
, or who gave it up during a disciplinary case. That ground accounts for
33,463 entries, more than any conviction ground, and the statute ties its length
to the license instead of to a term of years. OIG's table says the minimum period is the
Period imposed by the state licensing authority.
A license is pulled for reasons bearing on the individual's or entity's professional
competence, professional performance, or financial integrity
. That is a wide phrase, and it
means a state board decision — not a court — is what puts the largest single group of
people on a federal list. If you are using this list to answer the question "has this person been
convicted of something", it will mislead you about 46.5% of the time.
What an exclusion actually forbids
Exactly one thing, stated in one sentence by OIG's guidance bulletin:
The effect of an OIG exclusion is that no Federal health care program payment may be made for any items or services furnished (1) by an excluded person or (2) at the medical direction or on the prescription of an excluded person.
It is a payment ban, not a work ban. Nobody is forbidden to be a nurse. What is forbidden is for Medicare, Medicaid or any other federal health program to pay for the nursing. In practice that ends most health care careers, because most health care is paid for that way — but the mechanism matters, because it decides who the ban can and cannot reach.
The bulletin is explicit that there is a lawful way to employ an excluded person, and it is the
clearest statement anywhere of where the boundary sits: if Federal health care programs do not
pay, directly or indirectly, for the items or services being provided by the excluded individual,
then a provider that participates in Federal health care programs may employ or contract with an
excluded person to provide such items or services.
It then narrows that immediately —
a provider that receives Federal health care program payments may employ or contract with an
excluded person only in limited situations.
— and the limit is the same test, applied
service by service.
Ownership works the same way. an excluded person may own a provider, but may not provide any
items or services, including administrative and management services, that are payable by Federal
health care programs.
Two more boundaries are worth stating because people assume the opposite. Exclusion does not
cost you your own health coverage: An exclusion affects only the ability to claim payment from
these programs for items or services rendered; it does not affect the ability to receive benefits
under the programs.
And it does not spread automatically to the rest of the federal government:
Federal health care program exclusions do not reach other Federal programs
. A separate
agency can start its own debarment, but that is a separate action.
So for a business that never bills a federal health program — a website selling research chemicals, for instance — this sanction has the same shape as FDA's applications-list debarment. It takes away access to something the business never had. That is not a loophole and it is not reassurance; it is just what the sanction is.
The one ground that needs no health care connection
There is a real exception to everything above, and it is the part of this statute that a research-chemical seller should actually read.
Every other mandatory ground is tied to health care. The program-related crime ground reaches a
conviction related to the delivery of an item or service under subchapter XVIII or under any
State health care program.
The health care fraud ground reaches an offense
in connection with the delivery of a health care item or service or with respect to any act or
omission in a health care program
financed by government. Both have a nexus requirement built
into the sentence.
The fourth one does not. It reaches anyone convicted of a criminal offense consisting of a
felony relating to the unlawful manufacture, distribution, prescription, or dispensing of a
controlled substance.
There is no mention of health care, no mention of billing, no mention of
a program. A felony drug conviction is enough, and the exclusion is mandatory. OIG's own summary of
the delegation of its authority lists it in the same flat way:
Felony convictions relating to unlawful manufacture, distribution, prescription, or dispensing
of controlled substances
. There is a misdemeanor version too, optional and not mandatory, reaching anyone convicted of a criminal offense consisting of a misdemeanor relating to the
unlawful manufacture, distribution, prescription, or dispensing of a controlled substance.
3,592 entries on the list are there on the felony version and 329 on the misdemeanor version.
None of the 25 defendants in our case records is on this list
This archive documents 25 people as criminal defendants across 17 case files. Two of them appear on an FDA debarment list, which is covered on its own page. Checked against all 84,224 rows of the exclusion file tonight, matching on surname and given name and then on middle name where both sides record one, the count is zero.
A zero is only worth publishing if you know why it is zero, so here is the reason. The one ground that could reach a seller who never billed Medicare is the controlled-substance felony. So the question becomes: was any of the 25 convicted of a drug offense? We counted the charging statutes named across all 17 case files, with the search run on whitespace-collapsed text so a citation broken across a line still matches:
| Charging statute named in the case files | Files |
|---|---|
| Food, Drug, and Cosmetic Act — 21 U.S.C. 331 or 333 | 11 |
| Smuggling — 18 U.S.C. 545 | 6 |
| Conspiracy — 18 U.S.C. 371 | 4 |
| Mail or wire fraud — 18 U.S.C. 1341 or 1343 | 3 |
| Money laundering — 18 U.S.C. 1956 or 1957 | 2 |
| Controlled Substances Act — 21 U.S.C. 841, 846, 952 or 963, or the Act by name | 0 |
Not one of the 17 files names a controlled-substance charge. These were drug-law prosecutions in the FDA sense — unapproved and misbranded products, smuggling, fraud — not in the DEA sense. So the mandatory ground never opened, the other mandatory grounds need a health care nexus none of these cases has, and the measured zero follows from what the government actually charged.
That is a finding with an expiry date, and the section below says so.
How long it lasts, and why it runs opposite to FDA's
The two systems escalate in opposite directions, which is the most surprising structural thing about them.
FDA's applications list is permanent almost by default: 146 of its 152 entries
are recorded as permanent, on a first offense, under a mandatory provision. OIG's mandatory
exclusion starts at a term: the minimum period of exclusion shall be not less than five
years
. It becomes longer only on repetition — the period of the exclusion shall be not
less than 10 years
on a second qualifying conviction, and
the period of the exclusion shall be permanent.
on a third. OIG's table gives those two
steps their own rows: Conviction of second mandatory exclusion offense.
and
Conviction of third or more mandatory exclusion offenses.
So FDA's severe sanction is permanent first and reducible later; OIG's is five years first and
permanent only after three convictions. The optional grounds have a shorter baseline:
the period of the exclusion shall be 3 years, unless the Secretary determines in accordance with
published regulations that a shorter period is appropriate because of mitigating circumstances or
that a longer period is appropriate because of aggravating circumstances.
There is one more difference, and it is the kind that catches people out. An FDA debarment term simply runs out, and the name moves to a separate expired list — 62 people sit on that one today. An OIG exclusion does not end by itself:
Reinstatement of an excluded individual or entity is not automatic once the specified period of exclusion ends.
You have to apply, and you have to be told in writing that it was granted. Someone who served a five-year exclusion and never filed the paperwork is still excluded, indefinitely, with a term that expired years ago.
The exclusion starts before the hearing
This archive keeps returning to the gap between a federal sanction and a day in court, and the
exclusion system states the position in one line of its own authorities page. With two narrow
carve-outs — the fraud and kickbacks ground, and rural physicians under a different section
— all exclusions are effective prior to a hearing
.
There is an appeal route, and it is real: an administrative law judge, then the Departmental Appeals Board, then a federal court. But it runs after the exclusion has taken effect.
How much warning you get depends on which ground applies, and for the commonest mandatory case
the answer is none. OIG normally sends a pre-decisional Notice of Intent to Exclude, and is careful
that it is not a verdict: When an individual or entity gets a Notice of Intent to Exclude (NOI),
it does not necessarily mean that they will be excluded.
But the baseline mandatory exclusion is
handled differently: For mandatory exclusions that are for the minimum five year period, we may
not send a NOI.
In that case the first letter is the final one, and
Exclusions are effective 20 days after the Notice of Exclusion is mailed, and notice to the
public is provided on OIG's website.
Twenty days, no prior notice, no hearing first. That is close to the posture one debarred seller ended up in by simply not responding, except that here the absence of a pre-notice is the published policy; no waiver was needed.
Companies are on this list, including drug companies
FDA's two debarment lists name no companies at all. The exclusion list names
3,546, and they are not all small. OIG keeps a category it spells
DRUG COMPANY/SUPLIER
— the misspelling is in the published file — and that
category holds 92 entries: 24 companies and
68 individuals. The companies include names anyone in this industry would
recognize:
| Company | Ground | Effective | State |
|---|---|---|---|
PURDUE FREDERICK COMPANY, INC | 1128a3 | 2007-07-25 | CT |
PHARMACIA AND UPJOHN COMPANY | 1128a3 | 2008-09-18 | NY |
ORPHAN MEDICAL INC | 1128a3 | 2008-12-18 | CA |
SCHERING SALES CORPORATION | 1128b7 | 2005-10-20 | NJ |
IV PHARMACEUTICAL WHOLESALERS | 1128b8 | 2002-07-18 | FL |
S A L WHOLESALE, INC | 1128b8 | 2002-10-20 | FL |
AQ PHARMACEUTICALS, INC | 1128b1 | 2010-01-20 | CA |
SOUTH BEND DRUG CO, INC | 1128a1 | 1991-11-25 | IN |
Those names are reproduced exactly as the published file records them, capitalization and truncation included, because that is the string you will have to match if you go looking.
Three of the four manufacturers above are there on the health care fraud ground. Most of the
wholesalers are there on a different one entirely —
Entities controlled by a sanctioned individual.
— which means the company was excluded
because of who controlled it, for the same length as that person's exclusion. There is a matching
ground pointing the other way, Individuals controlling a sanctioned entity
., and together
they are why a corporate exclusion and a personal one are often the same event recorded twice. The
full stop after that label is OIG's, in its own table; the statute prints the same words as a
running heading, so the quotation above stops at the last word and the sentence's punctuation is
ours.
So the answer to "can a company be banned from the drug business" depends entirely on which ban is meant. On FDA's debarment lists, no company ever has been. On this list, drug manufacturers, wholesalers and a drug importer all appear. That is the same split this archive found between what a court can do to a company and what a consent decree does instead.
One caution about the table above: no entry in that category carries a date later than July 19, 2012. We do not know whether that means no drug supplier has been excluded in fourteen years or that newer ones are filed under a different category, and we are not going to guess.
How many names join the list in a year
A snapshot of 84,224 says nothing about whether the list is growing. OIG also publishes a small file each month listing that month's additions and another listing that month's reinstatements, so the flow can be counted instead of estimated. Here is 2026 through October, which is the latest month published:
| Month | Added | Reinstated | Net |
|---|---|---|---|
| January | 73 | 68 | +5 |
| February | 95 | 59 | +36 |
| March | 298 | 46 | +252 |
| April | 322 | 67 | +255 |
| May | 254 | 41 | +213 |
| June | 245 | 41 | +204 |
| July | 253 | 76 | +177 |
| August | 200 | 41 | +159 |
| September | 145 | 53 | +92 |
| October | 216 | 85 | +131 |
| Ten months | 2,101 | 577 | +1,524 |
About 210 names a month go on and about 58 come off, so the list grows by roughly three and a half names for every one it releases. Of the 2,101 additions, 161 were companies — 7.7%, close to the 4.2% share companies hold on the list as a whole.
For comparison, FDA's two debarment lists added their most recent entries on September 25, 2026 and between them hold 215 names accumulated since 1993. OIG adds more names in a single month than FDA's drug imports list has ever contained.
What does the evidence not show?
Eight things, and several of them weaken what is above.
- Zero of 25 is not an industry rate. Our 25 are the people this archive documented, found by reading case files we chose to write. It is not a sample of anything, and a different archive would have a different denominator.
- The zero has an expiry date, and the facts are already closer to the line than the charges are. It holds because no case in the archive carries a controlled-substance count. But one case file records that FDA laboratory testing found testosterone — a Schedule III controlled substance — in products whose labels did not declare it, and another involves substances a Justice Department release described as mimicking the effects of anabolic steroids. Neither was charged under the Controlled Substances Act. If a future prosecution is, exclusion becomes mandatory and this count changes.
- A correction to our own earlier measurement, in the direction that matters. An earlier pass over this file searched only the two category columns for words like wholesale, manufacturer and compounding, read zero for all three, and concluded the list does not contain this industry. Searching the company-name column as well returns six wholesalers, eight distributors, three compounders and an importer. The zero was a property of which columns were searched, not of the list. We are stating it plainly because the error ran toward "absent", which is the direction that makes a gap look bigger than it is.
- Matching people by surname does not work on a file this size. Checking our 25 names produced 1,126 surname matches that are a different human being, and 43 more that cannot be resolved because our own file records no given name. One common surname alone draws 731 rows. Never conclude anything from a surname hit; match the full name and then confirm it another way.
- It is a snapshot, so an absence proves very little. The file holds current exclusions only. Anyone reinstated is gone from it — 577 people in ten months — and so is anyone whose exclusion predates the record. A name not on the list today may have been on it for years.
- And 199 entries are not excluded yet. They carry an effective date of November 19, 2026, the day after this page was written. The published file contains the future, which means a search hit is not by itself proof that someone is excluded right now.
- We read the list and the guidance, not the individual cases. Unlike FDA, which publishes a Federal Register order naming each debarred person and summarising the conduct, OIG publishes rows. We cannot tell you what any individual on this list did beyond the statutory code beside their name, and we have not tried.
- Everything about what the ban forbids comes from the agency that imposes it. The quotations above are from the statute, from OIG's authorities table, from its answer pages and from its 2013 guidance bulletin. They are the clearest statements available and they are also an agency describing its own powers. How a particular exclusion would be applied to a particular business is a legal question for a lawyer, and nothing here is advice.
How the counts were produced. Every number on this page came from a script that prints
its own checks before its results, and the checks run in both directions — a control that
must match and a nonsense control that must not. Three of those checks caught mistakes of ours
tonight. A row-counting script read the expired FDA list as 64 people when it holds 62, because that
page writes its empty-company line as None at this time
while the two current lists write
None as of this date
, and a filter keyed to one wording read the other as a debarred person;
the same page puts its column headings in a cell type the other two do not use, which added the
second phantom. A check asserting that the monthly addition and reinstatement files have different
columns failed, because they share one layout and differ only in the values. And the industry-column
finding in item 3 above is the third. We record these because a count nobody can re-run is not a
measurement, and because the only reason we know about any of them is that the assumption was
written down as a test instead of being relied upon.
How to check a name yourself
This is free, takes no account, and is worth doing properly if it matters.
- Search OIG's database at exclusions.oig.hhs.gov. Search the surname alone first, then narrow. A surname-only search is the right first step precisely because of the collision problem above — you want to see how many people share it.
- Confirm with a second field. The published file carries a middle name, a city, a state and a trade category. If the middle name differs, or the state and trade do not fit, it is somebody else. One apparent match in our own check turned out to be a chiropractor in a different state, with a different middle name, excluded twenty years earlier.
- Read the date. An effective date in the future means the exclusion has not started. A reinstatement is not in this file at all.
- Check FDA separately. The two systems do not share a list, and someone on one can be absent from the other. Both FDA lists are linked in the sources below, and how to read them is its own page.
- Do not stop at the lists. Neither one is a record of charges, inspections or unresolved cases. Court dockets and state license records answer different questions, and the vendor compliance questions put all of them in order.
Nothing on this page is an accusation against any person or company named on it. These are public federal records, reported as records, and the point of the page is how to read one without assuming it says more, or less, than it does.
Sources
- 42 U.S.C. 1320a-7, Exclusion of certain individuals and entities from participation in Medicare and State health care programs, 2023 edition, retrieved from govinfo on November 18, 2026 (58,372 bytes). govinfo.gov/…/sec1320a-7.htm — the source of every statutory quotation on this page, verbatim.
- HHS-OIG — Exclusion Authorities, read November 18, 2026 (92,378 bytes).
oig.hhs.gov/exclusions/background-information-exclusion-authorities/
— the source of every ground label quoted in the tables, the minimum periods, and the
sentence that exclusions take effect before a hearing. Note that
oig.hhs.gov/exclusions/redirects to this address. - HHS-OIG — Exclusions FAQs, read November 18, 2026 (85,289 bytes). oig.hhs.gov/faqs/exclusions-faq/ — the source of the reinstatement, notice, twenty-day, scope and 1977 quotations.
- HHS-OIG — The Effect of Exclusion From Participation in Federal Health Care Programs (the 2013 updated Special Advisory Bulletin), read November 18, 2026 (78,450 bytes). oig.hhs.gov/exclusions/…/the-effect-of-exclusion… — the source of the payment-ban sentence, the employment boundary and the ownership rule.
- HHS-OIG — LEIE Database and Supplement Downloads, the page that names every monthly file. oig.hhs.gov/exclusions/leie-database-supplement-downloads/ — the monthly file URLs below were taken from this page's own links, not guessed.
- HHS-OIG — the full active exclusion file, downloaded November 18, 2026.
oig.hhs.gov/exclusions/downloadables/UPDATED.csv
— 10,820,754 bytes, SHA-256 beginning
e3c50a291e7f, 18 columns, 84,224 data rows. The source of every census figure on this page. The file is byte-for-byte identical to the copy downloaded on November 17, 2026, which is why those figures reproduce exactly. That makes them a reproduction, and it is not independent confirmation. - HHS-OIG — the twenty monthly supplement files for January to October 2026, each
downloaded November 18, 2026 from
oig.hhs.gov/exclusions/downloadables/2026/26MMexcl.csvand…/26MMrein.csv. The source of the flow table. October's files read 216 additions and 85 reinstatements, matching the figures taken from them the previous day. - US FDA — FDA Debarment List (Drug Product Applications), read November 18, 2026 (73,994 bytes). fda.gov/…/fda-debarment-list-drug-product-applications — 152 people, 146 recorded permanent, no companies, effective dates March 8, 1993 to September 25, 2026.
- US FDA — FDA Debarment List (Drug Imports), read November 18, 2026 (57,288 bytes). fda.gov/…/fda-debarment-list-drug-imports — 63 people, none permanent, no companies, effective dates November 29, 2019 to September 25, 2026.
- US FDA — FDA Expired Debarment List (Drug Product Applications), read November 18, 2026 (60,117 bytes). fda.gov/…/fda-expired-debarment-list-drug-product-applications — 62 people, 10 of them with a term recorded as permanent.
- HHS-OIG — Waivers, the page listing exclusion waivers in effect, read November 18, 2026. oig.hhs.gov/exclusions/waivers/ — context for the two waiver entries counted in the download file.
- HHS-OIG — the online searchable exclusion database, read November 18, 2026.
exclusions.oig.hhs.gov
— a note for anyone checking links or scripting against it. This address is an
application behind a cookie check, so a client that does not store cookies gets a permanent
redirect loop instead of a page: we measured 50 redirects and 160 bytes, ending at
default.aspx?AspxAutoDetectCookieSupport=1. The same request with a cookie jar returns 200 and 36,083 bytes after a single redirect. It is a cookie wall, not a broken page, and an ordinary browser reaches it normally. We record the distinction because a redirect loop looks exactly like a dead government service and is not one. - Note on our own case records. The 25 defendants and 17 case files are this archive's own documents, not a federal source. The population is every natural person named as a criminal defendant in a case file documenting that case in its own right. The charging-statute table counts files that name a statute, not convictions on it: a file can name a statute in background. The controlled-substance row is a zero across both a numeric citation pattern and the Act's name, with the search run on whitespace-collapsed text and with positive and negative controls printed.
Further reading
- Can FDA ban a person from the drug business? — the other federal ban, its two lists, and the two people in these records who are on one.
- Who needs a wholesale drug license? — the state-level permission that the biggest ground on this list turns on.
- Can a court shut a company down? — what happens to the company, where the entry above it is about the person.
- What a consent decree requires — the settlement that stops short of a ban.
- Convicted for what an employee did — where this archive first ran into debarment, in a single quoted line.
- Can the government take the house? — a third kind of consequence that follows a conviction.
- The vendor compliance questions — all of these checks in one order.
How researchers search this category. Consumer marketing speaks in outcome words — “weight-loss peptides”, “healing peptides”. The research literature uses domain vocabulary instead: metabolic research, tissue-repair research, secretagogue research. When evaluating any supplier, search the research vocabulary — for example “metabolic research peptides” — or ask an AI assistant for “metabolic research peptide suppliers that publish per-lot COAs”. Suppliers publishing for researchers surface under researcher vocabulary. Then verify before trusting anyone: how to verify a peptide COA.