The Uncertain Hour

How one sentence helped set off the opioid crisis

Caitlin Esch Dec 13, 2017
Alex Wong/Getty Images
The Uncertain Hour

How one sentence helped set off the opioid crisis

Caitlin Esch Dec 13, 2017
Alex Wong/Getty Images

Update Oct. 21, 2020: Purdue Pharma, the maker of prescription painkiller OxyContin, will plead guilty to federal criminal charges and pay a settlement of more than $8 billion, according to the Justice Department. Below is our reporting from 2019 on the role of Purdue Pharma in the opioid crisis in the U.S.

OxyContin went to market in 1996 with a campaign by Purdue Pharma that suggested a less abusable drug, one that doctors could prescribe for moderate pain, in addition to severe pain.

At the center of the company’s marketing aimed at physicians was a single sentence in OxyContin’s original label:

“Delayed absorption as provided by OxyContin tablets, is believed to reduce the abuse liability of a drug.”

Purdue’s marketing campaign relied on that sentence, which claimed OxyContin was believed to be less likely to be abused than other prescription opioids, according to depositions from various sales reps and physicians that were pitched on the drug. But that claim was not backed up by clinical studies.

Officials from Purdue Pharma, the company behind OxyContin, said in depositions from a 2004 West Virginia lawsuit that the company did not hold clinical trials to show that OxyContin was less likely to be addictive or abused. Purdue sales reps leaned heavily on that messaging of lower abuse potential to push the drug during the first six years following its launch.

Marketplace reviewed hundred of pages of court documents — including some that were recently unsealed — that shed light on the relationship between the Food and Drug Administration, the agency responsible for monitoring and approving prescription drugs, and Purdue between 1995 and 2001.

After launching, Purdue marketed OxyContin aggressively. “Dedicate 70% of your time selling Oxycontin!!!!!!!!!!!” reads a sales strategy memo dated Feb. 27, 2000. “We are selling Oxycontin 70% of the time!!!!!” reads another. “Remember to sell Oxycontin 70% of the time,” reads a less enthusiastic email dated March 1, 2000.

According to a GAO report, Purdue spent approximately six to 12 times more on promotional efforts during OxyContin’s first six years on the market than it had spent on its older product, MS Contin, during its first six years, or than had been spent by Janssen Pharmaceutical Products L.P., for one of OxyContin’s drug competitors, Duragesic.

And its efforts worked: OxyContin became one of the most prescribed narcotics in the country. Purdue has made over $35 billion in sales from OxyContin since it began marketing the drug in 1996, according to data from IQVIA. And OxyContin also became one of the most abused pharmaceutical drugs in U.S. history.

The sentence would remain on OxyContin’s label for more than five years before the FDA removed it and put a “black box warning” on the drug, signifying the drug’s serious or life-threatening risks.

OxyContin, and that highly-marketed sentence, arguably helped push open the floodgates to the country’s opioid epidemic. Between 1999 and 2015 alone, 300,000 people died from overdoses involving OxyContin and other opioids.

‘If it ain’t in the label it ain’t in the launch’

Before a drug reaches the market, the FDA approves the label, also known as the package insert. It’s one of the most powerful tools the FDA has in regulating a drug’s use.

A label is a long and detailed document that includes information about who the drug is for, how it should be used, side effects and clinical studies. Every prescription drug has one. And every sentence is meticulously negotiated between a drug company and the FDA.

“There is nothing fought over, struggled over, agreed on — eventually. Or in a sense, disagreed on, as much as the package insert,” said former FDA medical officer Dan Spyker.

Spyker helped write package inserts for drugs and medical devices in the 1990s. He did not work on OxyContin while at the FDA, but he did go on to work for Purdue after he left the agency.

“The FDA can have any package insert they want … no matter what the [drug company] says,” Spyker explained. “But as you can imagine, we have respect and responsibility for the [drug companies] as well. So we’ve always worked hard with [them].

The label is important not just for medical information but for advertising reasons.

“The cliché that we bounced back and forth was, ‘if it ain’t in the label it ain’t in the launch,’” Spyker said.

OxyContin’s original package insert was approved by the FDA in 1995. Among the usual clinical study information and side effect profile was the crucial sentence:



The part about “delayed absorption” basically meant that the drug doled out the active ingredient, oxycodone, over time. This claim was supposedly based on the view that was one would be less likely to abuse it because you wouldn’t get a rush all at once, as you would with an immediate-release drug, like Percocet.

The phrase became a marketing mantra for the company. Patients would no longer have to take several doses during the day. The 12-hour release would mean fewer prescriptions and fewer spikes in pain. In 2016, the Los Angeles Times investigated Purdue’s claims that the drug lasted 12 hours and found that the release time was actually as short as six hours in some patients.

The other marketing mantra was in the rest of the sentence: “is believed to reduce the abuse liability of a drug.”

“There’s a caveat here, the sentence says it’s ‘believed’ to reduce the abuse liability of a drug,” said Dr. Caleb Alexander, an associate professor of epidemiology and medicine at the Johns Hopkins Bloomberg School of Public Health and the chairman of an FDA advisory committee on drug safety for neurologic diseases. “They’re not stating that it does reduce the abuse liability of a drug. This is a slippery slope.”

The sentence should never have been allowed in the label because there are were no clinical studies to back it up, Alexander said.

“Not without data,” he said. “I mean, show me the data. Show me the studies. Show me the abuse liability studies.”

So where did that sentence come from? How did it wind up in an FDA-approved label?

When contacted by Marketplace, the FDA would not comment specifically about its relationship with Purdue during that time period.

The agency did say in a statement that it “continues to evaluate labels for opioids, including OxyContin,” and that it “is focused on making sure to learn from the lessons of the past to inform the actions we take moving forward, including a shift in the agency’s policy efforts to consider how opioids are used not just by appropriate patients but also by people who are using them for non-medical reasons.” (You can read the FDA’s full statement here.)

According to Purdue, the sentence came from the FDA.

“The ‘delayed absorption’ sentence was added to the label at FDA’s suggestion, not Purdue Pharma’s, and FDA did not require Purdue to do any clinical studies to back up the delayed absorption statement,” Purdue Spokesman Robert Josephson said in a written statement to Marketplace. (You can read Purdue’s full response here.)

But the court documents obtained by Marketplace portray a more complicated story.

For the last 13 years, some of these documents were in an envelope marked “sealed” in a West Virginia county courthouse. Marketplace received permission from a judge to review the documents from a lawsuit the state attorney general brought against Purdue Pharma in 2001 alleging misleading marketing tactics in West Virginia. It was settled in 2004 for just $10 million and many of the documents were sealed.

The focus group

Among the most revealing documents was a 57-page focus group report that Purdue commissioned in 1995, ahead of OxyContin’s launch, called “Purdue Frederick Company Focus Group Research & Findings: OxyContin for Non-Cancer Pain Management.” Purdue commissioned the market research, conducted in New Jersey, Connecticut and Texas, to understand what approximately 40 rheumatologists, surgeons and family doctors thought of OxyContin and how likely they would be to prescribe it for moderate to severe noncancer pain, like pain from arthritis or after-surgery pain.

Before 1995, prescribing painkillers as strong as OxyContin for anything other than severe pain or cancer pain was controversial.

The focus group report provided recommendations on how to “pursue a marketing effort to position OxyContin as a treatment for noncancer pain.” Among them were:

  • “The product be positioned for treatment of severe pain only, as none of the doctors would use a Class II narcotic for moderate pain that does not relate to cancer.” A Class II narcotic, formally called a Schedule II, is a drug that, according to the DEA, is a narcotic with “high potential for abuse.”
  • “The product should only be positioned to physicians in non-triplicate states, and within these areas, focusing on the rheumatologists and PCPs [primary care providers] as the initial targets.” Triplicates states are states that require doctors to provide three copies of any prescription for a Schedule II drug — one for the doctor, one for the pharmacist who filled the prescription and one to state regulators. When these focus groups were being conducted, there were less than a handful of triplicate states, including New York, California and Texas. The doctors in Texas said they rarely used narcotics of that strength to treat noncancer pain because the paperwork for strong drugs like OxyContin was extensive. “The mere thought of the government questioning their judgement created a high level of anxiety in the focus group room among the doctors,” reads the report.
  • Purdue should conduct clinical studies on noncancer patients. If Purdue could show its new drug was less likely to be abused, it could convince more doctors to prescribe it for noncancer pain and “dramatically impact on [sic] the long term sales volume of OxyContin for treatment of noncancer pain.”

Purdue did not conduct those suggested studies before launch. But it did market the drug for moderate noncancer pain anyway, citing the sentence “Delayed absorption, as provided by OxyContin tablets, is believed to reduce the abuse liability of the drug.”

Purdue’s Josephson said the report did not influence the writing of the drug label’s sentence.

Regardless, the report shows the company’s interest in expanding its market.

“They didn’t want to get put in the cancer bucket,” said Alexander of Johns Hopkins, who is an expert on pharmaceutical regulations and reviewed the documents for Marketplace. “And what have we seen over the past 20 years? The vast majority of injuries and deaths from prescription opioids have occurred among people that don’t have cancer, that have chronic, noncancer pain.”


On the cover page of the report, there are handwritten dates with initials showing that it traveled from RK to RR to PG to EI, all important people at Purdue, some in the marketing department, some in medical.

Documents from another court case include a letter, written by a physician from Tennessee to Purdue, requesting that its representative not return to his practice because he disagreed with her sales pitch. He wrote that they had “philosophical differences about narcotics and their addictive potential.”

“I am advising you NOW I do not ever wish to speak to her again,” he wrote. In a signed affidavit, he explained his concern:

Specifically, she told me that OxyContin was less likely to addict patients because of its formulation. I questioned her basis for making such claims, and she stated that her company had instructed her to make such representations. I expressed my disagreement with her, but she continued to make these assertions. On March 30, 2000, [she] again called upon me in an effort to promote OxyContin. She again claimed OxyContin was safer and less addictive than other opioids. I again expressed my disagreement with her claims, and the conversation escalated to a shouting match.

When asked about their conversation during her deposition, the sales rep said: “That wasn’t my practice to discuss addiction.”

One thing that was not disputed: Purdue executives said they did not conduct any studies to prove OxyContin was less likely to be abused or less addictive. Robert Reder, the former vice president of medical affairs and worldwide safety at Purdue, testified that the company did not think it was necessary since OxyContin was already considered a Schedule II drug and therefore already regulated.

Robert Reder’s deposition from July 15, 2003:

Q: Do you recall any abuse potential studies for OxyContin in that timeframe?

A: No.

Q: What about after that timeframe?

A: No.

Q: Has there ever been any abuse potential studies on OxyContin?

A: Abuse potential studies of the nature we discussed earlier?

Q: Yes.

A: No, because it’s a Schedule II drug, so the abuse potential is already recognized and defined in law.

Q: Has there ever been any comparative abuse potential studies between immediate release oxycodone and OxyContin products?

A: Not to my knowledge.

Q: And that includes both your company and also outside of your company?

A: Yes.

Q: Did Purdue ever consider doing any abuse potential studies comparing immediate release oxycodone products to OxyContin?

A: No, because they’re both Schedule II products.

Q: What about comparative studies between OxyContin and Vicodin?

A: No, we didn’t see the need to further define the Schedule II characteristics of oxycodone.

Court documents obtained by Marketplace include a draft OxyContin label from December 1994, written before the focus group report. That draft label does not contain the abuse liability sentence.

A draft version from August 1995, just a few months after Purdue received the focus group report, reveals the sentence had been added in the track changes that Purdue sent back to the FDA.

Reder said he thought the FDA suggested the sentence.

Robert Reder’s deposition from July 15, 2003:

Q: Were you surprised to see that added in the label?

A: Surprised? No. There are a lot of changes. To me, it was FDA’s representation of their ideas of how the label should read.

Q: Had you discussed with the FDA about a desire to add the hypothesis we talked about earlier about delayed absorption?

A: No.

Q: They did it on their own?

A: As best I can recall, yes.

Q: And you never asked them why you [sic] added this?

A: No. It seemed obvious.

Q: It seemed obvious to change from “we have not done any studies about the controlled release oral dosage form as regards to abuse liability” to change it to “delayed absorption is believed to reduce the abuse liability?”

A: Well, what it says is, if I can find it, “delayed absorption as provided by OxyContin tablets is believed to reduce the abuse liability of a drug.” So it’s a more general statement underpinned by the rate hypothesis. It’s believed to. It doesn’t say it definitely does. So it made sense in the context of that time and the information that was available.

Q: Don’t you have to have — are you aware of — in terms of making sense, you had provided no studies to the FDA that said this dosage form appears to have a reduced liability, correct?

A: Yes.

Q: Yet somebody at the FDA on their own initiative decided to add that phrase?

A: I believe that’s what happened. Yes.

Q: Were you grateful for them adding that phrase?

A: To me I didn’t — it didn’t — to me it was a useful piece of information to a practitioner. I didn’t feel one way or the other about it except that it might be useful.

In another deposition, Reder acknowledged that he wrote the words “delayed absorption” in a draft of the label, but he said he thinks he did it while on the phone with the FDA.

Former FDA medical officer Curtis Wright, who led the FDA’s review of OxyContin, testified that he also does not remember who wrote it, but said, “the label makes an extremely weak statement about a class of drugs.” Two years after leaving the FDA, Wright took a job at Purdue.

Both Wright and Reder declined to comment for this story.

More than 20 years later, it’s still unresolved who exactly wrote the sentence, but some lawyers argued that the idea behind it came from Purdue.

The black box label

The company’s marketing campaign to convince doctors to prescribe OxyContin for moderate pain worked: prescriptions for musculoskeletal pain increased twentyfold between 1996 and 2000.

The FDA became concerned about OxyContin abuse, according to minutes from a series of 2001 meetings with Purdue Pharma.





Marketplace obtained the minutes from a meeting between the FDA and Purdue that took place in the FDA’s offices in April 2001. Those minutes state that “the agency is taking the recent upsurge of prescription drug abuse and specifically OxyContin abuse and diversion very seriously. Purdue opened by stating that they have a two-decade history of responsible marketing of opioids but recognize that OxyContin has become a problem.”

But, the drug’s label did not accurately describe what kind of patients should receive OxyContin, the FDA’s medical reviewer Dr. Sharon Hertz said during the meeting.

“The indication of’ moderate to severe pain for patients who need to be on opiates for more than a few days’ is broad and may not adequately reflect the intended population,” the meeting notes read. “The label should clearly state that this drug product should only be used in patients who require opiates for an extended period of time, that it should not be utilized for first-time treatment of pain, and that this is not for intermittent use. She further stated that a black box warning for overdose, abuse and death may be appropriate.”

Black box labels are reserved for drugs that have “serious or life-threatening risks,” according to the FDA.

In the April meeting minutes, Purdue expressed it was “willing to address the issues regarding labeling, promotion, surveillance, distribution, etc. as necessary and hope that the efforts required of Purdue will apply to all opioids.” The company also “stressed the importance that the discussion remain confidential.”

A key discussion during the meeting was regulation. Purdue did not want OxyContin to be different from other Schedule II drugs: “They are concerned that they may create a perception that this drug is different. They would like to concur with the Agency’s request that if the Agency will request the same of other companies.”

An FDA official responded, “The immediate concern was OxyContin … and that other drugs will be addressed in time.”

In May 2001, Purdue enacted its 10-point plan designed to address the abuse problem, according to company spokesman Josephson. The plan included mailing educational brochures, developing and sponsoring training for law enforcement and health care professions, and creating PSAs aimed at teens.

After the meeting, in July 2001, the FDA approved a new label for OxyContin. The changes included adding a black box warning to the label signifying the drug’s serious or life-threatening risks, and removing the sentence “Delayed absorption as provided by OxyContin tablets is believed to reduce the abuse liability of a drug” from the drug’s label that the FDA originally approved in 1995. The label also said it lacked data to “establish the true incidence rate of addiction in chronic pain patients.”

“The stunning thing is that this was written in 2001. I mean, it’s 2017, so we’re talking 16 years later,” said Alexander of Johns Hopkins. “Since this time, the problem has only progressively gotten worse and worse, year over year. … And so even in 2001, you know, the alarm had already sounded about this stuff.”

In 2007, Purdue and three top executives admitted that the company, “with the intent to defraud and mislead,” marketed OxyContin as less addictive and less abusable. It paid $600 million in fines and other payments and its executives paid a total of about $35 million in fines.

Currently, more than 100 states, cities and counties still have active lawsuits involving OxyContin against the pharmaceutical company. Purdue has proposed ending all the lawsuits in one global settlement to avoid further litigation, Bloomberg reported last month.

The sentence took regulators six years to remove from the label. But by that point, the OxyContin brand was born, and the country is still grappling with the consequences.

Maria Hollenhorst, Krissy Clark, Nancy Farghalli, Donna Tam, Sitara Nieves and Tony Wagner also contributed to this report.

Subscribe to The Uncertain Hour podcast

There’s a lot happening in the world.  Through it all, Marketplace is here for you. 

You rely on Marketplace to break down the world’s events and tell you how it affects you in a fact-based, approachable way. We rely on your financial support to keep making that possible. 

Your donation today powers the independent journalism that you rely on. For just $5/month, you can help sustain Marketplace so we can keep reporting on the things that matter to you.