The System Didn't Get Fooled by Purdue. It Helped Write the Label.

On December 12, 1995, the FDA approved a new long-acting painkiller and let its label carry a single, soothing sentence: the pill's delayed absorption "is believed to reduce the abuse liability of a drug."⁷ No clinical trial said so. It was a theory, dressed in the calm authority of a federal label, and it became the most lucrative sentence in the history of American pharmaceuticals. OxyContin went on sale in 1996.¹ By the time the sentence was quietly struck from the label, the country had a body count.

The story everyone tells is that a rogue company lied its way past honest regulators. That story is comforting and mostly wrong. The regulators didn't get fooled into the abuse claim — they printed it. The truth is harder to sit with: the system didn't fail to catch Purdue. It co-authored the document Purdue then weaponized, and spent the next two decades making sure no actual person paid for it.

A label built on a belief, not a trial

Read the approved language again — "is believed to reduce the abuse liability." Belief is not data. The reduced-abuse claim rested on the mechanical theory that a slow-release coating would discourage misuse, with no clinical findings behind it.⁷ That distinction matters because a sales rep cannot say in a doctor's office what the FDA has not let the label say. The label is the permission slip. Once "reduced abuse liability" sat inside the federally approved text, Purdue's marketing wasn't freelancing — it was amplifying a claim the government had already blessed. The sentence stayed on the label until 2001, when it was finally removed and a black-box warning added.⁷ Five years of selling on a claim that should never have been printed.

Beneath the label sat an even flimsier foundation. The famous "addiction is rare" reassurance traces back to a 1980 letter to the editor in the New England Journal of Medicine by Porter and Jick — five sentences, reporting just four addiction cases among roughly 11,882 hospitalized patients given at least one narcotic dose.⁸ It studied people in hospital beds, monitored, dosed once. It said nothing about chronic outpatient use, and nothing about OxyContin. Yet by 2017 that letter had been cited 608 times, and 72% of the citing articles invoked it as proof that long-term opioid therapy rarely caused addiction — a claim the letter never made.⁸ A five-sentence note about hospital patients became the load-bearing wall under a national prescribing habit. The NEJM eventually appended an editor's note flagging the misuse.⁸ The wall held only because nobody checked it.

When the felony has no fingerprints

Here is the move that defines the reckoning, and it is a structural one. When accountability finally arrived in 2007, it landed on a shell. The entity that pleaded guilty to a federal felony for misbranding OxyContin as less addictive and less subject to abuse was the Purdue Frederick Company — an affiliate, not the operating Purdue Pharma L.P. — and the combined payments totaled roughly $600 million.² A corporate person took the felony. The human beings did not. A felony plea you can pay your way out of, attached to a subsidiary the public has never heard of, is not accountability. It is a receipt. The crisis kept accelerating for more than a decade after.

Compare the 2007 outcome with what came later and the asymmetry is the whole point. In 2020, Purdue Pharma L.P. itself pleaded guilty to three federal felonies, including conspiracy to defraud the United States, carrying a $3.544 billion criminal fine plus $2 billion in forfeiture.³ Bigger numbers, same shape: the company confessed; the family did not. And the dollar figures only became collectible after the company had filed for bankruptcy — Purdue Pharma L.P. and 23 affiliates entered Chapter 11 in September 2019.⁴ You cannot extract billions from an estate that has already been emptied.

The shield that arrived eleven billion dollars too late

The bankruptcy was not a surrender. It was a fortress. The proposed settlement would have handed the Sackler family a release from future civil claims — a permanent shield — in exchange for a contribution to the estate. On June 27, 2024, the Supreme Court struck that deal down 5-4 in Harrington v. Purdue Pharma L.P., holding that the Bankruptcy Code does not permit nonconsensual third-party releases.⁵ The decision reads like justice. The timing reads like a heist completed before the alarm rang. In the same record, the Court confirmed the Sacklers had withdrawn roughly $11 billion — about 75% of Purdue's total assets — before the bankruptcy was ever filed.⁵ The shield was struck down only after the thing it was protecting had already been carried out the back door.

After the ruling, the parties returned to the table. Purdue filed a new $7.4 billion plan in March 2025, and the bankruptcy court confirmed it in November 2025.⁶ A larger number than the deal the Court rejected. But the structural fact never moved: the Sacklers were never criminally charged. Every reckoning in this story is a transaction. None is a conviction.

Isn't this just a company that lied?

The fair objection is that this analysis lets Purdue off the hook. Purdue ran the aggressive sales machine. Purdue exploited the Porter-Jick letter in training material the letter could never support.⁸ Purdue is the villain — why blame the referee? The answer is that we are not absolving the company; we are explaining why the company succeeded. A drugmaker can market all it likes, but it cannot put words on a federal label, and it was the label that gave the abuse claim its authority.⁷ The honest counter cuts deeper still: even granting full corporate guilt, the system's job was to catch and punish that guilt — and it generated a felony with no fingerprints in 2007², let the estate be drained for years⁵, and produced its sharpest correction only after the money was gone. A villain is a story about one company. A crisis that ran for nearly thirty years is a story about the machinery that was supposed to stop it and instead kept signing the paperwork.

The deepest fiction in the OxyContin story isn't a marketing claim. It's the idea that the system was outwitted. It wasn't. It approved the sentence, accepted the affiliate's plea, watched the estate empty, and produced its boldest stroke — voiding the shield — only once $11 billion had already crossed the line where the law could reach it.⁵ Purdue didn't beat the referee. The referee was holding the pen the whole time. And the reckoning, when it came, was always a price — never a person.