Purdue Pharmaceuticals used to cite three major studies to argue that in prescribing OxyContin, addiction-risk was not significant. The most influential of those studies did not even mention OxyContin, because it was completed twenty-five years before OxyContin was sold. After more than a decade, the company had to admit that “data are not available to establish the true incidence of addiction.”
Editor’s note: This is the first of a series of three articles on how Purdue’s market strategies promoted one of the most well-known opioids, OxyContin. The series is based on edited excerpts from: “The marketing of OxyContin®: A cautionary tale” by David S. Egilman, Gregory B. Collins, Julie Falender, Naomi Shembo, Ciara Keegan and Sunil Tohan, published on the Indian Journal of Medical Ethics, Vol IV No 3 July-September 2019.
The development and marketing of new drugs inevitably involve striking a balance between efficacy and safety for patients, and profit for the pharmaceutical company. Unfortunately, such a balance is not always achieved, and aggressive, even misleading, marketing by a drug company can put the health and safety of patients at undue risk.
In the early 1990s, Purdue Pharma was faced with the imminent expiry of its patent on MS Contin, a long-acting, (“q12 hour”: every 12 hours.) morphine-containing pain medication. Threatened with the specter of competition from generics, the company developed OxyContin, a slow-release version of oxycodone, as a new q12 hour pain reliever. The US Federal and Drug Administration (FDA) approved OxyContin in 1995. Purdue launched sales in 1996, marketing OxyContin as the longest-acting narcotic pain reliever available. It promoted OxyContin as a superior alternative to other pain medications such as Percocet and Percodan, which afford shorter periods of pain relief (4 to 6 hours).
This marketing was incredibly successful. By 2000, US health professionals had written approximately 5.8 million OxyContin prescriptions, and OxyContin had become the number-one prescribed Schedule II narcotic in the United States.
Unfortunately, this commercial success was at least in part the result of a marketing campaign that included off-label sales, promoted questionably unique benefits of the product, and minimized the very real dangers of OxyContin, including abuse, diversion, addiction, overdose, and death. Since the drug’s introduction in 1996, opioid-related deaths have increased dramatically. From 1990 to 2017, over 210,000 people have died in the US due to prescription opioid overdoses alone.
A confidential Justice Department (DOJ) report from 2006 shows that Purdue Pharma was aware of the widespread abuse of OxyContin and proceeded to conceal this information until it was leaked in 2017 to the journalist Barry Meier, who published some parts of the investigatory record. At the conclusion of a four-year-long investigation, the DOJ recommended that three Purdue Pharma executives be charged with felonies. This investigation provided an opportunity to expose the evidence against Purdue.
Unfortunately, the George W. Bush administration at the time did not support this indictment, and the DOJ allowed the Purdue executives to plead guilty to misdemeanors. More importantly, the DOJ sealed the investigatory record and Purdue’s misleading branding and marketing tactics were not made public. Instead, aggressive marketing led to increased prescription of Oxycontin. From 2002 to 2009, the number of extended-release opioid prescriptions increased 146 percent, from 9.3 to 22.9 million.
Unsupported Claims of Low Addiction Risk in Chronic Pain Patients
As early as 1997, Purdue’s OxyContin physician-directed promotional pieces, including advertisements, contained statements that “less than 1 percent of patients taking opioids actually become addicted”; that addiction to opioid medication is “rare”; and that the notion that “opioid addiction (psychological dependence) is an important clinical problem in patients with moderate to severe pain treated with opioids” was a “myth.”
Purdue relied on three irrelevant documents to support its assertions. In one study, Medina and Diamond examined drug dependency among patients taking various medications for chronic headaches. They studied patients prescribed short-acting pain medications on an as-needed basis for relief of headaches. Only 62 of the 2,639 patients reviewed took narcotics, or a combination of analgesics and barbiturates, at least four times a week for at least six months. Of these 62 patients, 13 (19 percent) were dependent or abusers, based on definitions from the Diagnostics and Statistical Manual of Disorders (DSM). Only two of the 62 had taken an oxycodone product—Percodan (5mg); neither took OxyContin. This 19 percent rate of dependence and abuse is hardly “rare” and is certainly not “less than 1 percent” as Purdue asserted.
Moreover, this abuse/addiction rate occurred in patients taking PRN (as needed) narcotics, which are much weaker (Schedule III) than OxyContin (Schedule II). In fact, the original authors stated that “there is a danger of dependency and abuse in patients with chronic headaches.” When Diamond was asked about Purdue’s use of his study in a 2003 New York Times interview he stated that “[Purdue’s characterization of addiction risk] distorts the picture and it clearly underplays the risk.”
In the second paper Purdue cited, Perry and Heidrich studied patients in pain undergoing burn debridement. They sent questionnaires to 151 burn facilities in the United States to review how burn pain was assessed and managed. Perry and Heidrich did not use any diagnostic criteria for addiction; nor did they collect patient data. Instead, the study was retrospective, relying entirely on the memory of staff members, and none of the surveyed doctors used OxyContin , which was not on the market at the time of the study. Thus, the study provided no basis for comparison to current OxyContin usage; nor could it support Purdue’s contention that pain patients treated with narcotics experience a low rate of addiction.
Finally, to support its low risk of addiction claims for OxyContin , Purdue relied on a 110-word letter to the editor of the New England Journal of Medicine written by Hershel Jick, a doctor at Boston University Medical Center, and Jane Porter, a graduate student, in 1980 – 16 years before the introduction of OxyContin. The letter, in its entirety, is as follows:
“To the Editor: Recently, we examined our current files to determine the incidence of narcotic addiction in 39,946 hospitalized medical patients who were monitored consecutively. Although there were 11,882 patients who received at least one narcotic preparation, there were only four cases of reasonably well documented addiction in patients who had no history of addiction. The addiction was considered major in only one instance.
The drugs implicated were meperidine in two patients, Percodan in one, and hydromorphone in one. We conclude that despite widespread use of narcotic drugs in hospitals, the development of addiction is rare in medical patients with no history of addiction.”
There are several problems with Purdue’s reliance on this letter. First, this retrospective analysis does not include a single patient who was prescribed a controlled-release oxycodone product like OxyContin. Later, in sworn testimony, Dr. Jick stated that “We don’t have any information in this letter which relates to these “[people taking OxyContin ].” Only 450 of the nearly 12,000 patients reviewed took an oxycodone product, and at least one of the four addiction cases identified involved a patient who took such a product: Percodan. The largest dose of oxycodone in Percodan is equivalent to the lowest recommended dose of OxyContin.
As is the case with the other studies Purdue cited, the patients Porter and Jick studied bear little resemblance to those likely to be prescribed OxyContin. Porter and Jick studied hospitalized patients who may have received as little as one narcotic pill or shot, generally used the drug for a brief time, and had the drug administered by hospital personnel. Porter and Jick did not follow patients after discharge. Thus, the study could not reliably detect addiction, which develops over time after repeated use; nor could it address the arguably higher addiction risk in unsupervised ambulatory outpatients, who are most likely to use a controlled-release opiate on an extended basis.
The Porter and Jick study did not mention OxyContin at all because it was completed twenty-five years before OxyContin was sold. Most OxyContin patients self medicate at home, outside of the health care setting, where they can alter both the prescribed dose and dose schedule—alterations that increase addiction potential and rarely occur in the hospital setting. Much like Perry and Heidrich, Porter and Jick did not rely on standard DSM criteria to diagnose addiction, but instead based their results on the presence or absence of a written diagnosis of addiction in doctors’ or nurses’ notes in patient files. This is likely to have resulted in an under-recording of the “addiction” diagnosis.
With these glaring methodological flaws, Jick’s study should never have been used to support the contention that addiction in pain patients (including OxyContin patients) is “rare.” At the same time, Purdue failed to inform physicians of several studies that do indicate high rates of opioid addiction for chronic pain patients.
“Data are Not Available to Establish the True Incidence of Addiction”
In 1992, Fishbain et al. found rates of addiction of 3.2-18.9 percent in their meta-analysis of the scientific literature on chronic pain patients. These percentages were reiterated by Barry Dickinson in a 2000 review article for the American Medical Association Council on Scientific Affairs. A review of the literature on chronic pain and addiction reveals that numerous papers document rates of addiction between 3 and 34 percent in chronic pain patients.
In 2001, the FDA concluded that the three references did not support Purdue’s assertion of low risk, and required the company to change OxyContin ’s patient package insert wording from “iatrogenic ‘addiction’ to opioids legitimately used in the management of pain is very rare” to “data are not available to establish the true incidence of addiction”.
Unfortunately, this FDA-ordered “correction” occurred years after it had approved the original wording, long after the original misleading label was in the hands of millions of patients and prescribers.
The 2019 formal complaint lodged against Purdue by the state of Massachusetts’s Attorney General revealed that the company “tracked Massachusetts doctors’ prescriptions, visited their offices, bought them meals, and asked them to put specific patients on Purdue drugs.” It selectively targeted doctors that they deemed most likely to change their prescribing habits to benefit OxyContin prescribing amounts. Purdue “staff told the [owners of Purdue] that they would increase the number of sales visits and had hired McKinsey to study how to get doctors to prescribe more OxyContin.”
Company marketing pieces disguised as scientific knowledge were sent to prescribers. For example, the piece Focused and Customized Education Topic Selections in Pain Management “falsely told doctors and patients that signs of addiction are actually ‘pseudoaddiction,’ and that doctors should respond by prescribing more opioids.” Purdue paid doctors nearly $100,000 to promote its opioids, while assuring them that OxyContin was safe.
Statement of Conflict of Interest:
David Egilman has consulted on opioid litigation at the request of addicted patients and counties. Gregory Collins has consulted on opioid litigation at the request of addicted patients. David Egilman is the owner of Never Again Consulting; Naomi Shembo, Ciara Keegan, and Sunil Tohan were compensated by Never Again Consulting for their contributions.
The ProMarket blog is dedicated to discussing how competition tends to be subverted by special interests. The posts represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty. For more information, please visit ProMarket Blog Policy.