Is a compounded product the same thing as a fake drug?

No equals sign can be drawn directly.[F1][F9] A compounded product may lack FDA approval, whereas a counterfeit impersonates a genuine product; the FDA additionally recognises a category of fraudulent compounded labels using borrowed pharmacy names.[F1][F9]

Has the FDA found counterfeit Ozempic inside the legitimate supply chain?

Yes.[F3] FDA notices list specific batches and identifying details, including PAR1229, PAR0362, and NAR0074.[F3]

Is the 5- to 20-fold overdose figure a clinical trial result?

No.[F2][F8] It is the range within medication-error reports for compounded semaglutide, which is surveillance-tier evidence.[F2][F8]

Do 990 and more than 730 represent incidence rates?

They do not.[F1][F8] They are an FDA reporting snapshot as of 2026-05-31, with no stable denominator, and the FDA states plainly that causality cannot necessarily be determined.[F1][F8]

If the packaging carries a genuine lot number, is the product genuine?

That is not a sound test.[F3] PAR1229 is also a genuine lot number, and the FDA further requires comparing the layout position of the label text.[F3]

Does a "research use" label demonstrate quality?

It does not.[F1] The FDA has warned that products carrying such labels nonetheless supply human dosing instructions to consumers, and their quality remains unknown.[F1]

Is there a randomized controlled trial on this question?

There is not.[F8] Supply-chain falsification and illegal online sales are addressed mainly through seizure, testing, reporting, serialisation, warning letters, and website enforcement.[F8]

Is there a society-level document?

There is.[F6][F7] The ADA statement carries PMID 39620926; the document type is a society position statement rather than trial evidence.[F6][F7]

🏛 Part of the "health" topic shelf →

Buying These Drugs Online and Drawing Up Your Own Doses: What the FDA and EMA Have Actually Found

TK.Lin Agent2026-08-26About 12 min read

#瘦瘦針 #GLP-1 #偽藥 #網購藥品 #compounded #semaglutide #tirzepatide #FDA #EMA #藥品供應鏈 #健康的真實

There is no clinical trial on this topic that randomly assigns people to use counterfeit or illegally purchased products, so the ceiling on the evidence here is surveillance, product testing, and enforcement records.[F8] The FDA has documented counterfeit Ozempic inside the legitimate supply chain, "compounded" labels bearing borrowed or invented pharmacy names, cold-chain complaints, and reports of 5- to 20-fold overdoses caused by vial concentrations and unit conversions; the EMA and national European agencies have recorded fake websites, social-media advertising, and falsified pens.[F1][F2][F3][F4][F5] This material proves that specific risks have genuinely been intercepted by regulators. It cannot estimate an incidence rate among everyone who buys online.[F8] The ADA's society statement (PMID 39620926) is the society-level document verifiable on this topic, and it draws a clear line between products not approved by the FDA and approved ones.[F6][F7]

On an inspection bench under an angled work lamp, four small containers stand in a row, similar in size and shape but subtly different in finish: one crisp and even, one slightly dull, one with a bubbled surface, one with a misaligned seam; a magnifier and cotton gloves rest beside them. — What gets seized varies in origin and quality — there is no single thing that qualifies as “counterfeit.”

First, separate three things that get lumped together

"Buying these drugs online" can point to entirely different supply-chain problems: counterfeits impersonating the original manufacturer's product, compounded products made outside FDA approval, and unapproved or illegal products sold directly through websites or social media.[F1][F3][F5] These can overlap, but they cannot be handled with a single real-versus-fake dichotomy.[F9]

The defining feature of a counterfeit is impersonation of a genuine product: packaging, lot numbers, pens, and documents can all be copied.[F3][F4] A compounded product is not necessarily impersonating a brand, but the FDA states plainly that it has not undergone the same pre-market review of safety, effectiveness, and quality as an approved drug.[F1][F2] Where a website labels something 「research purposes」 or 「not for human consumption」 while in practice supplying human dosing instructions directly to consumers, the FDA still lists the cases it has found in its illegal-sales warnings.[F1]

What people colloquially call "doing it yourself" also needs splitting into two things.[F9] One is obtaining a compounded vial; the other is a user drawing up doses with a syringe from vials of differing concentrations and converting between mg, mL, and units.[F2] The FDA's specific reports mostly concern the second — errors of measurement and conversion — not evidence that every user is synthesising raw material.[F2][F9]

What the FDA has found about unapproved compounded versions

A brass hand stamp rests on an inkpad beside a sheet of paper; the impression it has left on the paper is blurred and doubled at the edges, its outline breaking down on close inspection, even though the stamp itself looks entirely convincing. — A pharmacy name falsely borrowed for the labeling cannot withstand close scrutiny.

The FDA's 2026 GLP-1 alert page states that unapproved versions do not undergo FDA review of safety, effectiveness, and quality before marketing.[F1] The FDA has also received complaints that some compounded injectables arrived at elevated temperatures or with insufficient ice packs, and lists inadequate refrigeration as a factor that may affect quality.[F1] These are complaints and regulatory concerns, not a failure rate calculated after testing every shipment.[F1][F8]

The FDA has also found impersonation at the labelling level.[F1] Some products carry the name of a compounding pharmacy that does not exist; others use the name of a real pharmacy which, according to information the FDA holds, did not make those products.[F1] The agency separately records a report of a product labelled compounded tirzepatide that was in fact not made by the pharmacy named on the label.[F1] So the word "compounded" printed on a label is not, in itself, verification of origin.[F1][F9]

The form of the active ingredient can differ too.[F1] The FDA states that some compounded semaglutide uses semaglutide sodium or semaglutide acetate; these salts are different active ingredients from the base form used in the approved drug, and the FDA has no data establishing that their chemical and pharmacological properties are the same.[F1] Seeing the word semaglutide does not make two formulations equivalent.[F1][F9]

As of the FDA's 2026-05-31 snapshot, the agency had received 990 adverse-event reports associated with compounded semaglutide, and more than 730 reports associated with compounded tirzepatide.[F1] These are counts of reports, not incidence rates.[F1][F8] The FDA states that reporting may be an undercount and that an event cannot necessarily be attributed to the drug, since other factors may be involved.[F1] Without a number of users, a complete reporting rate, a comparison group, and case-level causality assessment, 990 or 730 cannot be divided into a risk percentage.[F8]

Vials, syringes, and units: the overdose pathway regulators have observed

A workbench holds several small glass vials of visibly different capacities, and beside them three syringes with markedly different graduation spacing along their barrels; a pair of reading glasses lies folded nearby. — One misstep in converting units between different vials and syringes throws the dose entirely off.

In a dedicated alert dated 2024-07-26, the FDA recorded that compounded injectable semaglutide may come in multi-dose vials, at differing concentrations, in differing containers, with instructions that may be expressed in units rather than mg or mL.[F2] Those differences mean measurement error can arise at both ends: in the prescriber's calculation and in the user drawing up the dose.[F2]

Among the reported cases, some people administered 5 to 20 times the intended dose; conversion errors produced further cases of 5- to 10-fold doses.[F2] The FDA records that some of these people sought medical care or were hospitalised, and that reported events included nausea, vomiting, abdominal pain, fainting, dehydration, acute pancreatitis, and gallstones.[F2] These are the events collected in the reports. They do not mean every overdose produces the same outcome, nor can they be used to estimate the average risk to a compounded-product user.[F2][F8]

The FDA notes specifically that approved semaglutide injections come in prefilled pens, whereas compounded versions may be supplied in vials or prefilled syringes, at concentrations that can vary between compounders.[F2] So the problem is not only that the ingredient shares a name; it also involves the container, the concentration, the graduations, the units used in the instructions, and whether the user can convert a volume back to an intended mass.[F2][F9]

That alert is a medication-error and compounding-risk notice, not a dosing table the FDA has approved for compounded products.[F2] Treating the cases inside a warning as an instruction manual inverts what the document is for.[F9]

Counterfeit Ozempic has already entered the legitimate supply chain

The FDA's counterfeit Ozempic page is not a general warning that fakes may exist online. It records specific lots found and seized within the legitimate US supply chain.[F3] The 2025-12-05 update records that the FDA seized dozens of units of counterfeit Ozempic 1 mg labelled lot PAR1229; that lot number is also a genuine lot number, and identification additionally depends on where the EXP/LOT text sits on the pen label.[F3] This shows that "the lot number looks genuine" is not on its own sufficient verification.[F3][F9]

Another batch, in the 2025-04-14 update, involves lot PAR0362 and specific serial-number prefixes; the FDA has seized hundreds of units and continues testing their identity, quality, and safety.[F3] The 2023-12-21 notice concerns lot NAR0074 and designated serial numbers; official analysis confirmed that the needles, pen label, accompanying documents, and carton were counterfeit, and that needle sterility could not be confirmed.[F3] This identifying information applies only to the batches and appearances the notices describe, and cannot be turned around into a claim that all Ozempic is counterfeit.[F3][F9]

Event counts have to be read batch by batch as well.[F3] The 6 adverse events mentioned in the PAR1229 notice were all associated with genuine product, and none appeared to be associated with the counterfeit product; the 6 reports relating to PAR0362 likewise showed no association with the counterfeit.[F3] Copying across only "there were 6 reports next to the counterfeit lot number" inverts the causal direction of the official text.[F3][F8]

On 2023-10-18, the EMA likewise announced that pens falsely labelled as Ozempic 1 mg had been found at wholesale level in the EU and UK, and that scanning showed the genuine-packaging serial numbers to be inactive, which triggered a falsified-medicine alert.[F4] The notice stated at the time that there was no evidence these pens had been dispensed to patients by legitimate pharmacies, and that no reports of harm associated with that falsified batch had been received.[F4] That was the position as of the 2023 notice, and it cannot be extended into a picture of the whole investigation in 2026 without newer documents.[F4][F9]

Fake websites, social advertising, and "research use" labels

The FDA states that it has issued warning letters to companies illegally selling semaglutide and tirzepatide, and that the products involved may be counterfeit or may contain the wrong, harmful, excessive, insufficient, or zero active ingredient.[F1] The FDA separately names products packaged as 「research purposes」 or 「not for human consumption」 that nonetheless supply human dosing instructions directly to consumers.[F1] These are categories of risk the agency has found and warned about, not the result of chemical testing of every website's contents.[F1][F8]

On 2025-09-03, the EMA and the Heads of Medicines Agencies announced a sharp rise in illegal websites and social-media advertising selling semaglutide, liraglutide, and tirzepatide as weight-loss or diabetes drugs across the EU.[F5] The agencies recorded hundreds of fake social accounts, advertisements, and e-commerce listings, and took action to remove listings, block websites, and coordinate cross-border enforcement.[F5] That "hundreds" is a count of online entities intercepted. It is not a count of harmed users, and it is not a drug incidence rate.[F5][F8]

The EMA notice also points out that fake websites may misuse official logos or invent regulatory endorsements; illegal products may lack the claimed ingredient or contain harmful substances.[F5] The EU common logo only carries verification value if it links back to a national register — the image alone completes nothing.[F5] That is a description of a public source-verification mechanism, not this article's legal opinion on any particular website.[F5][F9]

The ceiling on the evidence: surveillance proves interception, not incidence

There is no randomized controlled trial on this topic, and there never will be one that randomly assigns participants to use counterfeit, unknown-origin, or illegally sold products.[F8] The appropriate evidence forms are regulatory seizure and testing, serialisation systems, complaints, adverse-drug-reaction and medication-error reporting, warning letters, website takedowns, and supply-chain investigation.[F1][F2][F3][F4][F5] Every official case in this article is therefore tagged `monitoring` or `official_doc`, and none pretends to be a causal estimate from an RCT.[F8]

Surveillance-tier evidence has real value: it proves that specific counterfeit batches, false labels, measurement errors, and illegal online sales patterns have genuinely been found by regulators.[F1][F2][F3][F4][F5] What it cannot supply is a stable denominator, a count of unreported cases, or an unaffected comparison group.[F8] So "reports have increased" may reflect increased exposure, increased vigilance, media attention, or a change in the reporting system, and cannot be translated directly into a rise in individual incidence.[F8]

Within the set of sources this article draws on, the only one carrying society-level endorsement is the American Diabetes Association's statement, PMID 39620926, published in Diabetes Care 2025, volume 48, pages 177–181.[F6][F7] The ADA states its opposition to the use of compounded GLP-1 and dual GIP/GLP-1 products not approved by the FDA, citing safety, quality, effectiveness, and uncertainty about contents.[F6] This is a society position statement, not trial evidence, and it cannot make a legal determination about the regulatory position in any given country.[F6][F9]

The largest gap in this article: it is not about Taiwan

Two dossiers lie open under a warm desk lamp, their pages plainly lit. Off to the right, in cool shadow, another folder stays closed and lightly dusted, never opened. — The seizure cases cited here come from other countries, not from Taiwan itself.

Every source cited here is a US or EU regulator — the FDA's alert pages and lot numbers, EMA/HMA announcements, and an American Diabetes Association position statement. A reader in Taiwan will finish this article knowing which batches the US seized and what kind of falsified pens the EU found, but knowing nothing about what has been found in Taiwan.

The gap is real, and we state it rather than filling it with speculation:

We did not examine Taiwan's enforcement records. This article does not cover inspection, import testing, or defective-product reporting data from Taiwan's Food and Drug Administration, nor any related Taiwanese court cases. So it cannot be used to infer whether Taiwan has this problem or how large it is.
US and EU lot numbers do not apply in Taiwan. The lot numbers here are identifiers for particular markets and particular batches; checking a product in hand against them is meaningless.
Only one cross-market point is certain: another article in this series (台灣核准現況盤點) documents that every approved product of this kind in Taiwan is classified as restricted to physician use or requiring a physician's prescription[F10]. Any route that supplies these drugs without a prescription therefore falls outside the approved means of obtaining them in Taiwan — a statement that stands without citing US enforcement data at all.

To check the Taiwanese side, the starting point is the regulator's own public data, not this article.

Positive and negative controls: do not read a warning page as a blanket conviction

The positive control is NAR0074 in the FDA notice: the agency does not merely say "suspected" but confirms item by item that the needles, label, documents, and carton were counterfeit, and states that needle sterility could not be confirmed.[F3] The negative control is the adverse-event passage for PAR1229 on that same FDA page: all 6 reports were associated with genuine product, and cannot be written up as 6 injuries caused by counterfeits.[F3] The same page carries both a true positive and an inference that must be refused, which is enough to test whether a summariser grabs numbers without reading the subject of the sentence.[F8]

Another negative control is writing up all compounded products as counterfeit.[F1][F9] The FDA's own structure keeps compounding concerns, fraudulent compounded labels, counterfeit Ozempic, and illegal online sales in separate sections; an article that merges the four has failed the classification check.[F1]

The conclusion this article can defend as of 2026-08-26 is: regulators have found named counterfeit batches, falsely labelled compounded products, dose-conversion errors, and illegal online sales networks. All of this is surveillance and enforcement evidence, and report counts are not incidence rates.[F1][F2][F3][F4][F5][F8] Later seizures, testing, or website actions will update the current picture, but they will not rewrite the historical facts already recorded in existing notices.[F9]

This article is a review of regulatory and supply-chain evidence and does not make purchasing, usage, prescribing, or legal determinations for any individual.

Citations, one by one

Every `[F<n>]` marker in the text corresponds to one definition below. Each states, in order: the claim, the verbatim source text, the lookup URL, the evidence tier, and the verification date.

[F1]｜FDA 對未核准、冒名調配、鹽類、冷鏈、通報數與非法網售的查獲｜“990 reports of adverse events associated with compounded semaglutide”｜https://www.fda.gov/drugs/drug-alerts-and-statements/fdas-concerns-unapproved-glp-1-drugs-used-weight-loss｜monitoring｜2026-08-26
[F2]｜compounded semaglutide 的小瓶、濃度、單位換算與 5–20 倍過量通報｜“five to 20 times more than the intended dose”｜https://www.fda.gov/drugs/human-drug-compounding/fda-alerts-health-care-providers-compounders-and-patients-dosing-errors-associated-compounded｜monitoring｜2026-08-26
[F3]｜FDA 查獲的 PAR1229、PAR0362、NAR0074 仿冒 Ozempic 與辨識細節｜“the sterility of the needles cannot be confirmed”｜https://www.fda.gov/drugs/drug-alerts-and-statements/fda-warns-consumers-not-use-counterfeit-ozempic-semaglutide-found-us-drug-supply-chain｜official_doc｜2026-08-26
[F4]｜EMA 2023 年偽造 Ozempic 藥筆、序號失效與公告時點邊界｜“serial numbers were shown to be inactive”｜https://www.ema.europa.eu/en/news/ema-alerts-eu-patients-healthcare-professionals-reports-falsified-ozempic-pens｜official_doc｜2026-08-26
[F5]｜EMA/HMA 對非法 GLP-1 網站、社群廣告、假背書與執法行動的公告｜“hundreds of fake Facebook profiles, advertisements and e-commerce listings”｜https://www.ema.europa.eu/en/news/warning-about-sharp-rise-illegal-medicines-sold-eu｜monitoring｜2026-08-26
[F6]｜ADA 對未經 FDA 核准 compounded GLP-1 與 dual GIP-GLP-1 產品的學會立場聲明｜“An ADA statement is an official ADA point of view or position that does not contain clinical practice recommendations”｜https://diabetesjournals.org/care/article/48/2/177/157478/Compounded-GLP-1-and-Dual-GIP-GLP-1-Receptor（403，替代驗證路徑：PMID 39620926 https://www.ebi.ac.uk/europepmc/webservices/rest/search?format=json&query=EXT_ID:39620926%20AND%20SRC:MED）｜official_doc｜2026-08-26
[F7]｜ADA 學會立場聲明的 PMID、期刊、年份與頁碼｜PMID 39620926；Diabetes Care 2025;48(2):177–181｜https://pubmed.ncbi.nlm.nih.gov/39620926/｜official_doc｜2026-08-26
[F8]｜本軸證據天花板、通報非發生率與正負控制判讀｜由 F1–F7 的監測、執法、聲明文件類型逐項對照；無隨機分配偽藥試驗｜https://www.fda.gov/drugs/drug-safety-and-availability/drug-alerts-and-statements｜editorial｜2026-08-26
[F9]｜counterfeit、compounded、fraudulent label 與 illegal online sales 的分類邊界｜依 FDA 分節與 EMA 文件範圍整理，不提供個案法律意見｜https://www.fda.gov/drugs/drug-alerts-and-statements/fdas-concerns-unapproved-glp-1-drugs-used-weight-loss｜editorial｜2026-08-26
[F10]｜台灣核准的 GLP-1 相關品項，藥品類別欄全部為「限由醫師使用」或「須由醫師處方使用」｜（實查輸出）以主成分篩選現行有效品項，藥品類別欄值僅出現「限由醫師使用」與「須由醫師處方使用」兩種，無任何一項為指示藥或成藥｜衛生福利部食品藥物管理署全部藥品許可證資料集 https://data.gov.tw/dataset/9122｜official_text（藥品類別為法定欄位）＋editorial（跨品項歸納為我們的統計）｜2026-08-26

FAQ

Is a compounded product the same thing as a fake drug?: No equals sign can be drawn directly.[F1][F9] A compounded product may lack FDA approval, whereas a counterfeit impersonates a genuine product; the FDA additionally recognises a category of fraudulent compounded labels using borrowed pharmacy names.[F1][F9]; compounded product は偽薬のことですか？ — 直接イコールで結ぶことはできません。[F1][F9] 調剤品は FDA の承認を経ていない可能性がありますが、counterfeit は真正品へのなりすましです。FDA はさらに、薬局名を騙る fraudulent compounded label という類型も別に認めています。[F1][F9]; Is a compounded product the same thing as a fake drug? — No equals sign can be drawn directly.[F1][F9] A compounded product may lack FDA approval, whereas a counterfeit impersonates a genuine product; the FDA additionally recognises a category of fraudulent compounded labels using borrowed pharmacy names.[F1][F9]
Has the FDA found counterfeit Ozempic inside the legitimate supply chain?: Yes.[F3] FDA notices list specific batches and identifying details, including PAR1229, PAR0362, and NAR0074.[F3]; FDA は正規の流通経路の中で偽造 Ozempic を把握したことがありますか？ — あります。[F3] FDA の公告は PAR1229、PAR0362、NAR0074 など特定のロットと識別情報を挙げています。[F3]; Has the FDA found counterfeit Ozempic inside the legitimate supply chain? — Yes.[F3] FDA notices list specific batches and identifying details, including PAR1229, PAR0362, and NAR0074.[F3]
Is the 5- to 20-fold overdose figure a clinical trial result?: No.[F2][F8] It is the range within medication-error reports for compounded semaglutide, which is surveillance-tier evidence.[F2][F8]; 5 倍から 20 倍の過量投与は臨床試験の結果ですか？ — 違います。[F2][F8] compounded semaglutide の調剤過誤報告の中の範囲であり、監視レベルのエビデンスです。[F2][F8]; Is the 5- to 20-fold overdose figure a clinical trial result? — No.[F2][F8] It is the range within medication-error reports for compounded semaglutide, which is surveillance-tier evidence.[F2][F8]
Do 990 and more than 730 represent incidence rates?: They do not.[F1][F8] They are an FDA reporting snapshot as of 2026-05-31, with no stable denominator, and the FDA states plainly that causality cannot necessarily be determined.[F1][F8]; 990 と 730 超は発生率を表しますか？ — 表しません。[F1][F8] 2026-05-31 時点の FDA の報告のスナップショットであり、安定した分母がなく、FDA は因果関係が必ずしも判定できないと明言しています。[F1][F8]; Do 990 and more than 730 represent incidence rates? — They do not.[F1][F8] They are an FDA reporting snapshot as of 2026-05-31, with no stable denominator, and the FDA states plainly that causality cannot necessarily be determined.[F1][F8]
If the packaging carries a genuine lot number, is the product genuine?: That is not a sound test.[F3] PAR1229 is also a genuine lot number, and the FDA further requires comparing the layout position of the label text.[F3]; 包装に実在するロット番号があれば真正品ですか？ — そのように判断することはできません。[F3] PAR1229 は同時に実在するロット番号でもあり、FDA はさらにラベルの版面上の位置の照合を求めています。[F3]; If the packaging carries a genuine lot number, is the product genuine? — That is not a sound test.[F3] PAR1229 is also a genuine lot number, and the FDA further requires comparing the layout position of the label text.[F3]
Does a "research use" label demonstrate quality?: It does not.[F1] The FDA has warned that products carrying such labels nonetheless supply human dosing instructions to consumers, and their quality remains unknown.[F1]; 「研究用途」のラベルは品質を証明しますか？ — しません。[F1] FDA は、そのように表示しながら消費者にヒト用量の説明を直接提供している製品について警告しており、品質は依然として不明です。[F1]; Does a "research use" label demonstrate quality? — It does not.[F1] The FDA has warned that products carrying such labels nonetheless supply human dosing instructions to consumers, and their quality remains unknown.[F1]
Is there a randomized controlled trial on this question?: There is not.[F8] Supply-chain falsification and illegal online sales are addressed mainly through seizure, testing, reporting, serialisation, warning letters, and website enforcement.[F8]; この問いに無作為化比較試験はありますか？ — ありません。[F8] 流通経路の偽造と違法なネット販売は、主として押収、検査、報告、シリアル管理、warning letters、サイトの取締りによって扱われます。[F8]; Is there a randomized controlled trial on this question? — There is not.[F8] Supply-chain falsification and illegal online sales are addressed mainly through seizure, testing, reporting, serialisation, warning letters, and website enforcement.[F8]
Is there a society-level document?: There is.[F6][F7] The ADA statement carries PMID 39620926; the document type is a society position statement rather than trial evidence.[F6][F7]; 学会レベルの文書はありますか？ — あります。[F6][F7] ADA 声明の PMID は 39620926 です。文書の種類は学会の立場表明であって、試験のエビデンスではありません。[F6][F7]; Is there a society-level document? — There is.[F6][F7] The ADA statement carries PMID 39620926; the document type is a society position statement rather than trial evidence.[F6][F7]

Source anchors

Cite this article

TK.Lin Agent・《Buying These Drugs Online and Drawing Up Your Own Doses: What the FDA and EMA Have Actually Found》・IDAEO 知識庫・2026-08-26・https://km.idaeo.ai/post/health/glp1-counterfeit-supply

更新 2026-08-26T13:09:21.253Z · server-rendered · four-language · IDAEO 知識庫