I have never seen anyone giving Congress any part of their contribution to the theoretical opioid crisis, but this may be their contribution that was signed into law by Pres Clinton.
S. 3163 (106th): A bill to designate the calendar decade beginning on January 1, 2001, as the “Decade of Pain Control and Research”
This law helped create the “5th vital sign”, and the Joint Commission (JC) made appropriate pain management as a MAJOR STANDARD, and hospitals were required to ask pts on exit surveys if their pain was properly managed.
When the law expired, the political majority of Congress had FLIPPED, and it was not renewed. The 5th vital sign – DISAPPEARED. The JC ran as fast as they could to disassociate themselves with the 5th vital sign and pain management as a major standard for healthcare entities to meet to maintain their accreditation with the JC.
Obama came to office on January 20, 2009, and Rick Scott became FL Gov & Pam Bondi became FL AG Jan 4, 2011, and the WAR ON DRUGS got ramped up!
How the FDA Helped Ignite, and Then Worsened, the Opioid Crisis
An investigation into the agency’s repeated violations of its own rules to approve addictive drugs—and its ongoing failure to rein in the public-health crisis.
On a sunny May afternoon 14 years ago, a group of doctors from across the country gathered in a windowless conference room at the US Food and Drug Administration’s main campus in White Oak, Maryland, with an urgent message: Prescription opioids were not just addictive, they were also ineffective in treating chronic pain.
The physicians, all pain and addiction experts, told Janet Woodcock, then director of the FDA’s Center for Drug Evaluation and Research (CDER), that pharmaceutical companies had lied when they claimed opioids harmed only abusers seeking to get high. The drugs could be valuable tools for relieving acute, short-term pain, the doctors allowed. But they accused the FDA of acting rashly, starting with the 1995 approval of Purdue Pharma’s OxyContin label, which helped expand opioid use to millions of new patients, including people with long-term conditions such as arthritis and back pain.
The 2011 meeting came as an epidemic was gathering force. Doctors were issuing 254 million prescriptions of FDA-approved opioids a year, enough to medicate every adult in the US around the clock for a month.
Had the agency adhered to federal regulations and its own guidelines and laws at the outset, the physician experts said, drug companies would never have been allowed to promote the high-dose opioids and their extended-release delivery systems for anything but end-of-life care. They demanded a public meeting to put the FDA on the spot. Without a change in opioid labeling that excluded use for chronic pain, they said, curbing the epidemic would be impossible.
“Our sense was that the FDA and the medical community had been duped by pharma,” says Andrew Kolodny, one of the addiction specialists at the meeting and a lead spokesman for the group. “We believed if an unbiased advisory committee were brought in, the agency would change course.”
But the following year, in a public hearing at the National Institutes of Health, FDA officials defended their opioid policy by pointing to antidepressants that had been approved without long-term data, relying instead on extrapolation and inferred benefits. In that case, as far as they could tell, the net effect had been positive.
It was hardly comparable. “We don’t have people shooting pharmacists to get Prozac,” Kolodny testified. Seemingly chastened, the FDA soon promised to work with manufacturers to conduct studies of opioids’ long-term safety and efficacy.
The physicians are still awaiting the FDA-sanctioned studies.
Today, opioid prescribing is down from its peak, and in March, Purdue Pharma filed a $7.4 billion plan in bankruptcy court to resolve mass lawsuits. Still, the crisis is far from over: The US Drug Enforcement Administration’s production quota—the annual cap on how much can be manufactured—for oxycodone remains more than 10 times higher than it was before the dawn of the epidemic in the 1990s. It hasn’t been lowered in four years.
US Oxycodone Production Quota
Kilograms
Since 1999, more than 800,000 Americans have died from opioid overdoses. The latest headlines focus on fentanyl, yet the staggering toll can be traced to the widespread availability of opioid pills made possible by decades of overprescribing. Experts say few users start with fentanyl.
Overdose Deaths Involving Opioids in the US
Deaths per 100,000 population
How did once strictly controlled substances become so accessible? Part of the answer is well known: Pharmaceutical companies, led by Purdue, distorted medical practice by redefining how pain was treated, influencing medical education and aggressively marketing opioids as safe and not all that addictive. They pushed doctors to consider pain the “fifth vital sign” (the traditional four are temperature, heart rate, respiration rate and blood pressure), circulated industry-funded research, and bought the influence of medical institutions and patient advocacy organizations. They shaped treatment guidelines, normalizing high-dose, long-duration opioid use despite clear evidence of harm. They rewarded high-prescribing doctors with lavish junkets and paid speaking gigs.
But much less has been revealed about the compromises, pressures and corner-cutting at the FDA that opened the barn door. In the end, the very agency charged with protecting the public fueled the epidemic by approving drugs with insufficient data and allowing lax labeling. It sped up drug reviews and cozied up to industry, offering little or no resistance as companies ran roughshod over long-standing regulations.
The inner workings of the FDA in its opioid dealings emerge from transcripts of government meetings and calls between pharmaceutical executives and the agency, and from dozens of interviews with scientists, physicians, regulatory lawyers, industry managers, current and former FDA officials, and other key government figures. Newly unearthed details reveal how the agency, exercising sweeping discretion over its regulatory powers, exposed itself to lawsuit threats from opioid makers over apparent regulatory breaches, including the improper approval of rival manufacturers’ drugs. The agency managed to placate all parties with compromises that violated its own rules, thus ushering to market more high-dose oxycodone pills without the requisite clinical evidence. As abuse ballooned, the FDA held secret meetings with Purdue to negotiate a relabeling that enabled the company’s chronic-pain marketing blitz—just as Congress and the DEA began to focus on the addiction crisis.
Finally, when the epidemic exploded into view, FDA officials met with industry executives and their academic allies in closed-door meetings, at which questionable science was marshaled to help the agency defend past decisions. Biased study methods were developed to underpin a new standard for future approvals. The move effectively swept the genesis of the addiction crisis under the rug.
“The FDA’s job with regard to regulation of drugs is to make sure that they’re proven safe and effective before allowing them to be promoted,” says Kolodny, medical director for the Opioid Policy Research Collaborative at Brandeis University, president of Physicians for Responsible Opioid Prescribing and a continuing thorn in the side of opioid makers and advocates of easy access. “Had the FDA properly enforced those laws with opioids, I don’t believe we would have an epidemic today.”
In a 1999 interview for a public policy research project, former FDA chief counsel Peter Barton Hutt put it bluntly: The drug approval process “is not essentially about good science.” Rather, he said, it’s about “the need to please the kings and queens in their own fiefdoms at the FDA divisions.”
That dynamic—how much discretion (“flexibility” in agency jargon) FDA gatekeepers should have as they navigate pliable scientific standards, political pressure, demands for transparency and an often too-cozy relationship with industry—is shaping up to be one of the new administration’s biggest regulatory challenges. The agency has faced sustained criticism not only over opioids but also for a string of high-profile disputes: Alzheimer’s, ALS and Duchenne muscular dystrophy drugs approved despite inconclusive data; contested decisions on Covid-19 treatments and vaccines; costly cancer therapies fast-tracked with limited evidence of benefit; and politically charged battles over abortion pills.
These are the kinds of issues, alongside the still-unfolding opioid addiction crisis, that now sit with Health and Human Services Secretary Robert F. Kennedy Jr. and Marty Makary, the FDA’s newly appointed commissioner. Kennedy, a skeptic of scientific consensus, has pledged to restore what he calls the agency’s “gold standard” of review. Makary, a prominent surgeon, has long pushed for greater transparency and scientific rigor in the US health care system, including at the FDA.
President Donald Trump has promised bold action to address the opioid epidemic, even as the FDA, rattled by high-level resignations and internal unrest, remains deeply wary of his agenda. Whether Makary will revisit the decisions that helped set the crisis in motion—and continue to drive it—remains an open question.
But the roots of those decisions stretch back long before OxyContin, to a time when the FDA’s hand was being forced by politics and a desire to speed new drugs to market
“Come on, it’s just morphine”
The opioid epidemic can’t be fully understood without considering the political climate at the FDA in the 1980s. Although its resources were stretched, the agency was under intense demands to approve drugs more quickly. It had been blamed since the ’70s for a “drug lag,” a delay far longer than in other countries in making new medicines available. Criticism mounted in the ’80s as activists protested over bottlenecks in the approval of AIDS drugs. Palliative-care doctors were also demanding better pain relief for end-of-life care.
A small Connecticut company then called Purdue Frederick Co. exploited this pressure. Later renamed Purdue Pharma, the drugmaker manufactured generics at the time, but it aimed to expand into far more lucrative branded medicines.
Testing a new drug is expensive, and it might not pan out. So Purdue’s strategy was to make small tweaks to existing drugs. With the FDA’s blessing, an old drug could be considered a new drug, with lucrative patent protection. A few quick hits might vault a tiny company into the league of Big Pharma.
MS Contin was Purdue’s first such hit. MS Contin pills—“MS” for morphine sulfate, “Contin” for a “continuous release” tweak—were supposed to slowly seep into the bloodstream. When Purdue introduced the drug for investigational use in 1982, the FDA ordered the company to conduct controlled clinical trials in humans. Purdue ignored the directive, and two years later its sales teams fanned out across the US, selling MS Contin without testing or approval. When the FDA called out the company for pushing an illegal opioid, Purdue executives countered that it was perfectly legal and that no approval was needed, because morphine had already been approved and was available in other formulations. “They said that they had several thousand cancer patients who were getting relief from the drug and that they had no plans to stop marketing it,” says Frank Vocci, an FDA reviewer at the time. “They realized that they could put the FDA in an untenable position, making the FDA the bad guy.”
The agency backed down, but not because it accepted Purdue’s claims about the drug. “The data do not even support the implication carried in the product’s brand name, that MS Contin continuously releases morphine,” Paul Leber, then the director of the FDA’s Division of Neuropharmacological Drug Products, wrote in a 1986 memo. The delivery mechanism simply dumped the drug into the bloodstream. FDA officials proposed an immediate ban, but Purdue appealed to the commissioner, Frank Young, and even to the Reagan White House, to allow sales on “humanitarian grounds.” Recalls Vocci: “Purdue was saying, ‘Come on, it’s just morphine.’ We were all just thinking, ‘This is not the way business is done.’”
While continuing to sell the drug, Purdue did submit a New Drug Application, or NDA. The FDA called it “exceedingly weak” but approved MS Contin anyway in 1987. Leber’s review noted unmet scientific standards—“the NDA submitted does not provide results from adequate and well controlled clinical investigations”—and external pressure to approve. The studies Purdue submitted to support its application, he wrote, measured only morphine levels in the bloodstream. Blood levels can’t predict an opioid’s effectiveness, because tolerance builds over time, leading some patients to increase their dose for relief until eventually the drug stops working, even when it’s abundant in the blood. “In the face of testimony from ‘experts’ who argued MS Contin was essential for the treatment of countless victims of advanced cancer,” Leber wrote, “the usual regulatory and scientific requirements for the approval of a new analgesic drug product need not apply.”
This was the dawn of what was called the pain movement. Palliative-care specialists argued that pain, particularly in end-of-life cases such as terminal cancer, was going undertreated. This advocacy challenged the strict opioid controls and pharmaceutical regulations established after the last major opioid crisis, an epidemic in the late 19th and early 20th centuries caused by the treatment of Civil War wounded and by the marketing of patent medicines for “female complaints” and fussy children. Revelations of the dangers of these drugs led to the Pure Food and Drug Act of 1906 and the creation of the Food and Drug Administration.
A key element in any drug’s regulatory status is its FDA label, which tells physicians and patients what the drug is effective for, discloses side effects and limits pharmaceutical marketing to a specific patient population. MS Contin’s labeling repeatedly cited end-stage cancer patients, but it was also vague: “For use in patients who require repeated dosing with potent analgesics over periods of more than a few days.”
That wording—“more than a few days”—proved to be a dangerous precedent at a time when pain movement advocates began to extend their campaign beyond just palliative care.
Jane Ballantyne, a retired University of Washington professor and pain medicine expert, says the problems began when decades-old lessons were scrapped on compassionate grounds: “The thinking was that if you could relieve people who were dying of the burden of pain at the end of life, then why not do the same thing for people with chronic pain? What they didn’t realize is that lifelong treatment with opioids is very different from short-term treatment, even if that short term is a period at the end of life.”
The FDA recognized that this class of opioids was new—intended for a different patient population, and taken longer term—so it conditioned approval on a Purdue promise to conduct legitimate studies showing it worked. None were ever submitted.
Amid growing calls from industry and pain movement advocates to enable access to the drugs, the FDA would approve new opioids for an ever greater number of patients. In at least one case, that involved reversing a denial.
Late in 1987, Alza Corp. applied to market a transdermal skin patch packed with fentanyl, also claiming it released slowly over time. Fentanyl, a synthetic opioid, had been restricted to hospitals for surgery and intensive-care units. Alza aimed to extend it into the homes of cancer patients. The patch, called Duragesic, was reviewed by the same FDA neuropharmacological division that had capitulated to Purdue on MS Contin. Alza submitted only one two-week controlled trial for chronic cancer pain. According to FDA records, the application was based on “very confounded” and “inconclusive” studies and “failed to prove [improved] efficacy over placebo.” In February 1989, it was rejected. That should have been the end of Duragesic. It wasn’t.
“License to kill”
Public frustration with the FDA’s slow pace boiled over in October 1988, when more than a thousand AIDS activists protested outside its headquarters over its failure to approve treatments. It was a “die-in” with mock tombstones reading “I died for the sins of the FDA.” The next year, the agency created an experimental division, the Pilot Drug Evaluation Staff. To lead the unit, Carl Peck, then head of the FDA’s CDER, selected John Gamble Harter.
Breaking from the agency’s usual safety-conscious, arm’s-length approach, Harter directed the staff’s medical officers—physicians who assess drugs for safety and efficacy—to expedite reviews by working closely with industry and skipping sign-offs from higher-ups. Supervisors were sidelined. Drugs could be greenlighted with only two signatures. Officials were encouraged to be lenient in accepting blood-level data to permit less testing.
In 2000, Harvard Kennedy School doctoral candidate Richard Doblin detailed Pilot Drug’s history in his thesis. He provided this reporter with his notes and excerpts transcribed from recorded interviews with crucial players. (Doblin is now president of the nonprofit Multidisciplinary Association for Psychedelic Studies.)
FDA colleagues reported that Harter, who died in 1996, scorned hierarchy and spoke of Pilot Drug having a “license to kill” when it came to bureaucracy. To drive home the point, Harter chose a number for his unit referencing James Bond: HFD-007 (“HFD” stands for Health and Human Services/Food and Drug). Inside the agency, the cheeky move did little to assuage fears of compromised public safety.
Leber, the neuropharmacological division director, characterized Pilot Drug as a “political creation established … as a political sop to the pressure to expedite drug approval.” The FDA as a whole, he told Doblin, “sometimes seems more like a cheerleader for industry than a regulator.” Harter wrested away control of abused drugs and analgesics, placing them under his division. This action led to working closely with pharmaceutical companies to shape their applications, Doblin said. For Harter (a man his boss, Peck, called “disturbingly creative”), the goal was to “make an application approvable.”
Harter said that a drug that “had no evidence it works” and even had “significant toxicity” should be approved if the information was disclosed on the label, according to Mary Doug Tyson, Harter’s widow, who was also an FDA official working under Peck. Harter preferred that the agency “leave it to doctors and patients to decide whether to use it,” Tyson told Doblin.
By law, the FDA must determine that the bar for substantial evidence of efficacy has been met—and for good reason. Most physicians have neither the ability nor the time to evaluate clinical data themselves.
The key FDA reviewer in the Pilot Drug division was a pharmacologist named Curtis Wright. In July 1989, Harter resurrected the Alza fentanyl patch and assigned the application to Wright. Legally, positive results from two well-controlled trials were required before a drug could be approved for cancer pain. Alza hadn’t produced any. But Wright worked with the manufacturer to develop a rationale. FDA records and statements show that, under Wright’s guidance, Alza returned to its failed short-term study and rejiggered the data to make the opioid seem significantly more effective than a placebo. Wright didn’t respond to multiple requests for comment for this article.
According to Tyson, Harter initially worried that the fentanyl patch, “should it be used outside of the hospital, might kill people,” but his usual viewpoint—that with full disclosure, any drug can be approved—prevailed. In August 1990, with Harter’s support, Wright greenlighted Duragesic for nonhospital use. No additional clinical data was submitted.
Six months later, Harter gave a presentation to the American Association for the Advancement of Science on Pilot Drug’s review process. His account caused an uproar. A headline in one trade publication read: “‘Data Dredging’ Was Key to Toradol, Lodine, Duragesic Approvals, FDA’s Harter Says.”
In response to the controversy, Congress launched a General Accounting Office (now the Government Accountability Office) probe into the issue. One GAO investigator told Doblin that the office was concerned that “FDA was being pushed too hard to let drugs out” and “through hocus pocus, found a way to approve.” Agency leadership backed Harter’s approach and resisted the inquiry. It was dropped.
Even as Wright signed off on Duragesic, he acknowledged in his review that the drug might “spread beyond the use which has been evaluated in clinical trials,” calling that prospect “an unknown hazard.” By law, drug manufacturers are barred from encouraging doctors to prescribe outside the approved labeling. Physicians, however, can prescribe off-label at their own discretion, provided they inform patients of the risks.
Opioids were not labeled for long-term chronic pain. The dangers were clear: inexorable dose escalation, fatal respiratory depression, dependence, addiction and overdose, not to mention the hazards of sedating patients who might then fall down a flight of stairs or cause a car crash. Pills that didn’t get swallowed often lingered in medicine cabinets, easily found by teenagers and others looking to experiment—or to resell what amounted to a pharmaceutical equivalent of heroin.
Purdue, with its reputation for audacity, was already pushing the boundaries of the FDA’s labeling. “No one was thinking, ‘Let’s move out of cancer pain,’ when Purdue crossed that line [into chronic pain],” says oncologist Kirk Shepard, who during that period was a senior executive at Roxane Laboratories, a competitor in the opioid business in Columbus, Ohio. Morphine, though, still carried stigma among physicians and patients. Broadly marketing the drug for chronic pain was a nonstarter.
So Purdue turned to an opioid without such baggage: oxycodone.
“It’s not just for end stage cancer!”
Oxycodone had been included for decades in a low-dose combination drug with either aspirin or acetaminophen and marketed as Percocet and Percodan for acute, short-term pain. Hardly innocuous, Percocet and Percodan were among the most abused prescription narcotics.
But many doctors mistakenly believed that oxycodone was a safer, less potent drug than morphine, according to 1990s market research. In fact, oxycodone was as much as 1.5 times more powerful. Its side effects—dependence, addiction, withdrawal, tolerance buildup requiring ever higher doses and fatal respiratory depression—were dire. It was a misperception that Purdue executives, including president and chairman Richard Sackler, agreed not to correct, according to internal emails made public in recent litigation.
They saw potential in a high-dose, extended-release version of pure (single-entity) oxycodone, intended for chronic use. They developed OxyContin with oxycodone as its only active ingredient and tested it in a 14-day, placebo-controlled clinical trial for osteoarthritis. The drug didn’t perform well: It failed to separate from placebo with the 10-milligram dose, though it did pass at 20mg. But in his review, Wright called its use a “very helpful … nononcologic chronic-pain model.” Helpful, because Purdue was trying to establish a precedent: getting an opioid indicated—therefore legally marketable—for a vast new population of patients beyond end-stage cancer.
At a May 1993 regulatory conference in Philadelphia, Wright acknowledged the problem of proving opioids effective for chronic pain, but he signaled to the drugmakers in the room that the FDA’s traditional scientific standards wouldn’t stand in the way. He was open to new approaches. Chronic pain was an “underserved area,” he said, and he pledged to “do everything that is humanly possible to make it as easy as possible to get these products to market.”
At the same time, trouble was brewing at Pilot Drug. After Peck retired from the agency in November 1993, FDA Commissioner David Kessler appointed Woodcock to take over the CDER. Harter soon left the agency.
Woodcock had come to believe that Pilot “was not always impartial,” that it “wasn’t following regulations” and that “the scientific rigor of the reviews was deficient,” she told Doblin. Other top FDA officials voiced similar concerns. Murray Lumpkin, a deputy director at CDER, derided Pilot’s “cowboy mentality.” Woodcock eventually phased out the division and retired its name, replacing it with the Division of Anesthetic, Critical Care and Addiction Drug Products. She didn’t want Wright to lead it. Still, it took her a year to name a successor—leaving Wright, who’d approved Duragesic despite gaps in the data, shepherding OxyContin’s application.
There were two major problems. First, instead of delivering a continuous dose and providing steady pain relief, OxyContin drove an initial opioid rush into the bloodstream, followed by a drop-off. Some patients had to pop more pills long before the recommended 12-hour interval. OxyContin was “an addiction-producing machine,” David Egilman, a professor of family medicine at Brown University, testified as an expert witness in opioid litigation. The pills’ extended release mechanism was easily defeated: crushed and swallowed, snorted like cocaine, or dissolved and injected for a heroin-like high.
Purdue officials helped Wright draft his review of the drug, according to an internal US Department of Justice memo in 2006. (The memo reviewed a proposed indictment and recommended that certain top executives at Purdue face felony charges. They did not.) Wright signed off on a Purdue claim—unsupported by human data—that “delayed absorption, as provided by OxyContin tablets, is believed to reduce the abuse liability of a drug.” The line, removed years later under pressure, was interpreted as an FDA permission slip to downplay the addiction risk. And sure enough, sales reps aggressively touted OxyContin as less addictive and less abusable than other opioids. In 2007 the company was fined $634 million for this, and executives pleaded guilty to criminal misbranding charges.
The other problem was OxyContin’s only active ingredient: oxycodone. Despite its use in low doses in combination drugs, pure oxycodone had never been formally approved by the FDA. It wasn’t grandfathered under the 1938 law that exempted older drugs from pre-market review, nor had it gone through the FDA’s 1962 Drug Efficacy Study Implementation process as a stand-alone drug. This meant OxyContin legally required two adequate and well-controlled trials; comparison studies alone weren’t enough.
Most of Purdue’s studies lacked placebo controls. Further, they compared the extended-release OxyContin to oxycodone—which, because oxycodone wasn’t approved, failed to meet the agency’s evidentiary standard. The placebo-controlled trials also failed to meet the standard for chronic use. Yet the FDA approved the drug for noncancer chronic pain after Wright argued that although the only acceptable study (the 14-day trial of 133 osteoarthritis patients) was too weak to support an osteoarthritis indication, it still met the threshold for approval and broad chronic pain use due to “substantial evidence of the short-term analgesic efficacy … in this chronic-pain model.”
This presented a conundrum. Under FDA guidelines, a two-week study wasn’t adequate to justify a label allowing a drug to be marketed for long-term chronic pain. That required trials lasting at least 12 weeks.
In labeling negotiations that followed, the agency and Purdue leaned on the precedent. “Chronic” was struck from the label, but Purdue kept the vaguer phrase “for more than a few days.” When that language was used for MS Contin, it applied to cancer pain. Now the indication referred to a chronic-pain study. It was a subtle shift with big implications. The FDA had now officially allowed opioid marketing to reach a much broader population of pain patients outside hospitals.
Whether those patients could safely take opioids for months hadn’t been tested in controlled studies. But in line with the division’s thinking at the time, decisions about long-term use—and escalating doses—were left up to patients and doctors.
In December 1995, Robert Bedford, rotating in as acting director of the Division of Anesthetic, Critical Care and Addiction Drug Products, signed off on Wright’s OxyContin review. “It was a done deal, in the can,” he said in an interview before his death in 2022. “I just gave it a rubber stamp.”
And so, ambiguous and contradictory as it was, the FDA’s label effectively legalized the marketing of OxyContin for chronic noncancer pain.
Purdue then proceeded to aggressively market OxyContin with the slogan “The one to start with, and the one to stay with.” Going forward, ever-higher twice-daily doses would gain FDA approval. After the 40mg dose came 60mg, then 80mg, then 160mg. Again, blood level tests were used as justification. Never had a drug company been granted a license to broadly promote an opioid in the US for any chronic-pain condition.
Purdue, in partnership with Abbott Laboratories’ massive salesforce, pitched OxyContin to doctors as a go-to painkiller—describing the “selling points” in internal planning documents as “the logical next step” after aspirin for chronic noncancer pain, “easy to live with” and “ideal for long-term therapy.” Marketing materials encouraged higher doses with “no ceiling.”
As Purdue’s bold marketing drove sales, other companies sought to emulate the strategy. Alza, which worked with Janssen (a subsidiary of Johnson & Johnson) to market the drug, began targeting new patient populations for the Duragesic fentanyl patch. “It’s not just for end stage cancer anymore!” trumpeted its promotional materials.
Agency regulators told companies to advise patients to avoid around-the-clock opioids if they could manage chronic pain by “lesser means.” But the tepid warning was mostly drowned out by industry’s hard sell.
“A pretty big mistake”
In September 1997, an FDA independent advisory committee met to consider Anesta’s Actiq, a powerful dose of fentanyl delivered in a lollipop. The drug had been tested on patients with advanced cancer, but Wright pushed for broader approval to include chronic pain, citing the agency’s desire to avoid creating “therapeutic orphans,” patients without approved treatments for their condition.
The committee chair, John Downs, an anesthesiologist who oversaw pain management as head of the Department of Anesthesiology and Critical-Care Medicine at the University of South Florida in Tampa, was “uneasy,” according to the transcript. He expressed “grave” concerns about chronic-pain patients managing such a potent drug, pointing to the growing number of patients in his own experience already struggling with narcotics. In a moment of restraint that wouldn’t last, the agency limited Actiq’s label to opioid-tolerant cancer patients and banned broader promotion.
The following month, Wright left the FDA. A year later, he would accept a job at Purdue as executive medical director, with initial compensation of almost $400,000 a year. He hasn’t talked with the press, but in 2023 a local TV reporter in New Hampshire knocked on his door. Asked if he had regrets about his work at the FDA, he said no. “I did my job,” he said. “I never saw anything that I would not want to do. It’s a difficult, terrible situation. I feel so sorry for the people who were hurt. And I am so sorry for the patients who can’t get good pain relief now.”
His departure didn’t shift the agency’s trajectory. Cynthia McCormick, the new division chief overseeing opioids, didn’t question precedent. Blood-level data rather than full clinical trials had already been deemed acceptable. On Oct. 26, 1998, McCormick approved the first of Roxane Laboratories’ two new oxycodone applications. The drug in question was an extended-release competitor to OxyContin. The second application, for a fast-acting, immediate-release form of the drug, was still pending.
What happened next caught both Roxane and the FDA off guard. Purdue sent a letter threatening to sue the agency if it allowed Roxane’s drug to launch, warning of an “irrecoverable erosion of the price obtained on OxyContin.”
Roxane had secured a pathway by arguing oxycodone was an “old drug” with “grandfathered” historical approval. But Purdue’s lawyers argued—correctly—that “the precursor product … single-entity oxycodone is not and has never been approved by FDA.” The only agency-approved version was Purdue’s own extended-release OxyContin, backed by its proprietary studies, flawed as they were.
With billions of dollars at stake, Purdue turned on the agency that had enabled its rise, trying to block a rival with a strange and striking argument: The FDA had never properly vetted the active ingredient in its own blockbuster drug. With stunning chutzpah, the company was calling out the same cowboy regulatory culture that had cleared the way for OxyContin’s approval. What it didn’t say (but made unmistakably clear) was that even though the FDA may have cut corners before, it had better not do so again in approving products from anyone else.
Roxane immediately fired back with its own threat to sue the FDA “should [it] grant any part of the relief requested.”
The agency was in a bind: A three-way legal shoot-’em-up would leave its reputation in tatters. The FDA sought a way to get both parties to put down their guns. On a call with Sean Alan Reade, Roxane’s regulatory affairs director, an FDA representative said “way-higher-ups” at the agency would extend an olive branch. According to minutes of the conversation with the FDA that Roxane had kept, the company could “salvage” its second, as-yet-unapproved drug by “doing some magic with Percodan.”
Percodan, the low-dose blend of drugs including oxycodone in a single pill, was FDA-approved, so there was historical data on safety and efficacy. But because Percodan was a mix of medications, it was impossible to isolate oxycodone’s effect. To approve a pure, single-entity drug by referencing a combination drug would violate the agency’s scientific regulations.
Nonetheless, the FDA general counsel’s office was on board, according to Roxane’s minutes of its discussions with the agency. McCormick ultimately delivered what seemed like a quid pro quo: The FDA would greenlight Roxane’s second application if the company stood down on the already-approved first application. Roxane agreed. McCormick’s workaround would allow the company to sell the fast-acting, immediate-release drug it would call Roxicodone, despite the lack of clinical data.
Conference-call minutes reveal McCormick and Roxane brainstorming ways to resolve deficiencies in the Roxicodone application. In the words of company representatives, they relied on “clinical practice experience, leading experts, and performing audits of clinician prescription habits to justify the need for higher doses.” McCormick said that “in the absence of efficacy data … a clear theoretical rationale must be created,” to establish “a precedent for others to follow.”
That strategy violated FDA regulations, which hold that “isolated case reports, random experience, and reports lacking the details which permit scientific evaluation” are not only inadequate, but “will not be considered” by FDA reviewers. Nonetheless, Roxane dutifully submitted the agreed-upon evidence, which included citing a 220-fold jump in US oxycodone prescriptions and a trend toward higher doses—real-world “proof” of the drug’s supposed safety.
Hundreds of millions of Roxicodone 15mg and 30mg pills would soon flood the country. Roxicodone took on street names: Roxies, “Blues” for the baby-blue color, and later “M’s,” for Mallinckrodt Pharmaceuticals, which produced a generic version of the drug stamped with its logo.
“The shift to chronic use with no studies to back it—that’s a pretty big mistake. Definitely,” Roxane’s Reade acknowledged years later. “It really was a reach. You have to understand, the FDA’s role is not to approve. Their role is to protect patient safety. And that’s by statute. We didn’t protect patient safety, did we?”
In 2002, McCormick would exit the agency through the government-to-industry pipeline—in her case, accepting consulting gigs from pharmaceutical companies, including Mallinckrodt. She declined to comment for this story.
“A big problem”
Experts date the start of the opioid epidemic to within three years of the approval of OxyContin in 1995. Reports from emergency departments across the US showed Purdue’s pills were being crushed and injected or snorted as early as 1997. “My eyes popped open,” recalls one FDA medical officer of seeing the reports. “Nobody wanted to see it for what it was. You would’ve had to have your head in the sand not to know that there was something wrong.”
By 2000, Purdue was selling $1.1 billion annually in OxyContin. Higher doses led to higher profit. Sales reps were coached accordingly.
In the first months of 2001, as news reports revealed the epidemic’s devastating scope, the FDA called Purdue executives into two secret meetings. “The agency is taking the recent upsurge of prescription drug use, specifically OxyContin abuse and diversion, very seriously,” McCormick told them, according to the FDA’s minutes of the first meeting.
In five years, oxycodone prescribing had surged 402%, and hospital emergency room mentions of oxycodone were up 346%. Martin Pollock, a safety evaluator with the agency, noted a shift in prescribers from cancer clinicians to family practitioners. Use for noncancer conditions such as back and joint pain had experienced a twentyfold jump.
McCormick then acknowledged the clinical trials section of OxyContin’s label was a “big problem.” The studies of opioids focusing on osteoarthritis and postsurgery patients were inadequate and sent “a misleading message,” she said. The studies should be removed from the labeling.
In the wake of the 2001 meetings, Purdue made several concessions to the FDA over OxyContin’s label. It agreed to add a warning in the indication section against any short-term uses such as dental pain after an operation. It also added a black-box warning—the FDA’s most stringent safety alert, essentially a skull and crossbones—and pulled its 160mg dose from the market. Purdue further agreed to remove that now-infamous claim about the drug’s delayed absorption mechanism, which sales reps had used to downplay the risk of addiction.
But critically, Purdue refused to remove references in the label to its short-term studies, including the two-week controlled osteoarthritis study that underlaid the drug’s approval. Without them, there would be nothing left to justify OxyContin’s efficacy in any setting.
Then, in the tense negotiations, the FDA caved to a new far-reaching Purdue demand. It agreed to a label change that solidified the company’s right to more explicitly promote even longer-term use. OxyContin would be allowed for chronic-pain patients “around-the-clock … for an extended period of time.”
Agency officials were pleased, given the agreement to end promotion for short-term conditions such as dental pain. But Purdue executives were euphoric about this “expanded indication,” internal documents show. “The action by FDA … has created enormous opportunities,” read one memo. Chronic pain was the moneymaker, and the indication had been vague on that score. No longer.
By 2012, OxyContin sales were almost $3 billion annually. And many other companies were cashing in. In the preceding six years, 76 billion opioid pills had been produced and shipped across the US, as the FDA faced a national crisis of epic proportions.
Looking back, Edwin Thompson, a longtime pharmaceutical executive who began his career in opioids at Johnson & Johnson in the 1970s, points to the 2001 FDA meetings with Purdue as the government’s last chance to halt the epidemic. The fateful decision to cement the shift to chronic long-term use “opened the floodgates,” he says. John Jenkins, the most senior FDA official at the 2001 Purdue meetings, had even threatened product withdrawal as a remedy. But looking back, the threat was an empty one, Jenkins says. “To force withdrawal of a drug is a very complicated and difficult legal process. No reasonable FDA attorney would have taken the case. We were trying to put the genie back in the bottle.”
Kessler, the FDA’s commissioner in the 1990s, observes that Woodcock was “too concerned with denying patients Oxy.” Woodcock, who stepped down last year after more than four decades at the agency, pointed in a recent interview to the pressure from chronic-pain patients who “flocked to our meetings and gave us thousands of comments to our dockets. They feel they can’t function without them,” she said. “There are people who have arachnoiditis, or people with sickle cell disease, there are people with some cancers who don’t just die but live for years in pain.”
At a food and drug law conference three years ago, she reflected on the crisis and the FDA’s regulatory record, likening her job to setting highway speed limits—where science provides only imperfect estimates of future fatalities and offers no clear guidance on how to balance safety and freedom in a democracy. “We were in the middle of the pain movement,” she said. “It was felt that [OxyContin] would mitigate people[’s] severe pain, but the projection of the harms … was way off.… It was a miscalculation.”
The FDA’s reluctance to change course along the way may also have stemmed from an institutional instinct to protect its reputation. “When they approve something based on efficacy, the agency has been very reluctant to walk it back,” says Harvard professor Daniel Carpenter, author of Reputation and Power: Organizational Image and Pharmaceutical Regulation at the FDA, a definitive history of FDA culture. Admitting a mistake, he says, is seen as a black mark.
And so, opioid prescribing, addiction and overdose deaths kept rising. In the 2010s, the US, with less than 5% of the global population, was consuming 80% of the world’s oxycodone. And with coordinated pharmaceutical campaigns to destigmatize opioids, brands other than Purdue’s and Roxane’s benefited. For instance, Americans were also using 99% of the world’s hydrocodone, a drug marketed as Vicodin.
“It’s cheating”
After the 2011-12 FDA and NIH meetings, where pain and addiction specialists urged a reassessment of the agency’s opioid policies, Kolodny kept pressing for answers. Through public-records requests, he uncovered revealing emails, which he shared in 2013 with reporters from the Washington Post and the Milwaukee Journal Sentinel. The documents pulled back the curtain on exclusive, closed-door gatherings of pharma-aligned academics, FDA officials and opioid manufacturers. For more than a decade, executives had paid organizers as much as $35,000 apiece for the chance to mingle with FDA staff and help shape a new class of clinical trial methods—scientifically shaky workarounds that reopened the door to new opioid entrants.
As Kolodny and his allies challenged each new approval, inevitably accompanied by an FDA-sanctioned marketing campaign promoting chronic-pain prescribing, agency officials were periodically pressed about their long-standing promise to require studies that could actually prove whether the long-term use of opioids was safe and effective. In 2017, at a meeting with physicians at Yale University, senior CDER official Judy Staffa was again challenged on the issue. “We’re still trying to figure that one out,” she admitted.
One obvious reason for allowing the delay in producing the studies: Inaction is politically safe. Millions of Americans take prescription opioids for chronic pain. A legitimate study could push the FDA to ban opioid pharmaceutical marketing for chronic pain and prompt doctors, policymakers and insurers to restrict the drugs, risking a backlash from dependent patients. And if opioid labeling were revised to reflect only those conditions where the drugs are proven safe and effective, patients could lose insurance coverage for the medications.
In April 2023 the FDA finally assembled a group of medical experts to once again consider the question of how to prove the drugs were safe and effective for people with chronic pain. Agency officials revealed a new methodology hammered out with the industry consortium responsible for the studies: a trial with an “enriched enrollment randomized withdrawal” (EERW) design, a controversial method that screens out patients who don’t respond or experience side effects. Participants are given opioids for weeks before the trial officially begins; then half are taken off the drug, triggering withdrawal symptoms and heightened pain in the placebo group, while the treatment group continues on the medication.
The virtual meeting room exploded with objections. Mary Ellen McCann, a professor of anesthesiology at Harvard Medical School, called the approach “designed to give a positive result before the study’s even begun.” Caleb Alexander, a professor of epidemiology and medicine at Johns Hopkins University, told the panel that the proposed study design “raises the deadly serious question as to whether the FDA is really seeking to change the way it does business.” He questioned why, more than 20 years into an epidemic, the FDA was not requiring companies to prove their opioids “work in the first place.”
Thompson, the veteran pharma executive and now president of Philadelphia contract drug manufacturer PMRS, sided with the physicians and angrily testified that relying on an EERW study violated agency regulations. “The source of these overdose deaths are prescriptions from licensed physicians practicing under FDA labeling,” he said. “Unsupported efficacy should be removed from the label, period.” The agency’s proposed approach, he said, would only paper over decades of baseless FDA decisions with more bad science.
Kolodny flatly charged, “It’s cheating.” The agency wasn’t requiring companies to produce valid evidence that their drugs actually work or demonstrate quality-of-life improvement, he said.
In the face of this pushback, the agency and the drug companies returned to the drawing board. Given the FDA’s tortured history with opioids, many experts and observers are skeptical that a new plan will force it to reverse past decisions that continue to fuel the crisis.
Chronic pain patients haven’t fared well. One in 32 people on long-term, high-dose opioids dies within two and a half years. Of the 9.4 million Americans addicted to opioids, 4.8 million are hooked on prescription pills—most taking them exactly as prescribed. Because physicians now know about these outcomes, many are unwilling to prescribe opioids for chronic pain, even as they acknowledge that some patients may be stuck on them for life, unable to taper off without severe withdrawal, drug-induced pain sensitivity and psychological distress.
As the FDA and an industry consortium continue working to demonstrate that their already-approved opioids are effective and safe, they will meet again with the independent advisory committee on May 5—their first since the contentious 2023 hearing. Meanwhile, the addiction crisis grinds on. In the past year alone, it has claimed another 100,000 lives.
Filed under: General Problems | 1 Comment »
