Explaining the Opioid Crisis: Strategic Ignorance at the FDA
A Knology researcher reviews evidence from her upcoming book about the institutional logics that jump started the opioid crisis.
As far as the FDA is concerned, the opioid crisis began in December 1995. That month, Purdue Pharma received a letter approving their application for OxyContin, a new and promising opioid painkiller. Immediately thereafter, Purdue started promoting its new drug, which was prescribed for moderate to severe chronic pain—a condition that plagues 1 in 5 Americans. Soon, the money started pouring in. From 1997 to 2002 alone, OxyContin prescriptions for non-cancer related chronic pain patients skyrocketed, increasing nearly tenfold—from 670,000 to a whopping 6.2 million.
But as OxyContin became America's go-to pain medication, evidence of the drug's dangers began to emerge from all across the country. In the press, ominous stories about OxyContin overdoses began to circulate. Federal agencies became aware of a surge in false prescriptions (that is, black market purchases) of OxyContin. In 1999, the DEA reported that nearly all new arrests for false prescriptions in Maryland had been written for Purdue's newest painkiller. One year later, OxyContin overdoses were reported across the country. Baltimore, Boston, Billings, Denver, Detroit, Honolulu, New Orleans, Philadelphia, St. Louis, and Washington D.C. all reported to the Office of National Drug Control Policy that the "Oxy crisis" had spiraled out of control. When the DEA reviewed autopsy data to assess the damage, it concluded that OxyContin had played a role in hundreds of deaths. In all but a fraction, the people who died had taken the drug orally—exactly how it was prescribed.
The surge of deaths linked to OxyContin seemed to show that the painkiller posed risks. So why did the FDA approve it? After all, it was never a secret that opioids are powerful, habit-forming and—in some cases—dangerous. Was this bit of knowledge unknown to the FDA? Were they aware? Or were they ignorant?
The Pre-History of OxyContin
To answer these questions, we must go back to 1989, when the first clinical trials for OxyContin began. That year, 90 women in two different hospitals in Puerto Rico signed up to try the new opioid. All of these patients were suffering from pain, though not the kind for which OxyContin ended up being indicated for. They were not chronic pain patients, but surgical ones. This is significant, since it is well known in pain medicine that acute surgical pain is easier to treat than chronic pain. This being the case, OxyContin should have overperformed, especially considering that the acute pain sufferers in the trial were also opioid intolerant. Astoundingly, however, OxyContin did not perform well: half the patients who tried it did not experience the 12 hours of relief it promised. Many needed an additional dose within eight hours. But none of this mattered to the FDA, whose twin pillars of safety and efficacy indicate that a new drug can be approved as long as it is found to be no less safe and no less effective than other approved drugs in its class. Having set the bar low, the FDA set OxyContin up for immediate success. Although 90 Puerto Rican women would not have agreed, the drug was marked as having passed its test.
Aside from the obvious, what's wrong with this picture? How could a study carried out with a homogenous group of opioid-naive acute pain patients be used to assess how the drug would function among a diverse group of opioid-tolerant chronic pain sufferers? How reliable could this information be for the agency charged with anticipating the new drug's benefits, not to mention its risks? The short answer is, it could not and it was not. The slightly longer answer, however, is a bit more complex. It has to do with something called strategic ignorance.
As Linsey McGoey (2019) has written, ignorance is not always the opposite of knowledge. Sometimes burying one's head in the sand is not passive, but productive. Not a sign of vulnerability, but a purposeful strategy. When employed strategically, ignorance also produces knowledge, and that knowledge has consequences. These can be mild or—in the case of OxyContin—deadly serious. While OxyContin's clinical trials did produce knowledge about the drug, this knowledge was strategically limited. And when the painkiller was approved, much was still unknown about it and the patients who would come to use it. In my view, the strategic ignorance that defined OxyContin's clinical trials also defines the opioid crisis as a whole.
That is, it helps explain how we got here in the first place.
As I argue in my upcoming book, strategic ignorance has advanced both the interests of pharmaceutical companies and the ongoing over-production of prescription painkillers over the past two plus decades. The Oxycontin clinical trials are evidence of this. But let me point to another example that illustrates how this occurred at the FDA.
One of the FDA's regulatory practices is something known as "pharmacovigilance." It requires pharmaceutical companies to monitor the use and abuse of their products after they are released onto the market. In the case of OxyContin, pharmacovigilance helped produce strategic ignorance and, in doing so, set the stage for Purdue Pharma to assert control over the drug's consumption data in ways that would end up pulling the wool over the FDA's eyes. This is because the pharmacovigilance system Purdue put in place, RADARS, had a number of limitations. First, the opioid use data it compiled was done at infrequent intervals, making it difficult to measure the scale and scope of opioid misuse in real time. Second, the data itself was tethered to emergency room admissions and hospital records in metropolitan areas, which meant that data from rural or more remote areas was missing. This shortcoming is important for the opioid crisis, which first took shape among rural communities on the East coast and in the Southeast. Finally, databases like RADARS collect data on overdoses but are not capable of distinguishing between the different drugs that might have caused them. In cases where an overdose victim has more than one substance in their system, RADARS is unable to identify the specific drug that caused the overdose or death.
All these limitations benefited Purdue. The data they collected not only met the FDA's safety requirements, but also provided priceless information about OxyContin's use across the country, which Purdue then used to market the brand. That is, RADARS endowed Purdue with both a risk management plan and the ability to collect huge amounts of information about the precise areas (down to three digits of the zip code) where people were prescribing and consuming OxyContin. They then used this information to identify high-prescribing doctors and adjust their marketing plan to target them.
But that's not all. The same surveillance technology Purdue used to enhance its marketing tactics was also used to manipulate OxyContin misuse patterns. Purdue's promotion of RADARS, which boasted a new ability to track OxyContin trends to specific zip codes, also produced data where OxyContin misuse appeared more local than it really was. So while OxyContin use was devastating communities from Baltimore to Honolulu to Washington, D.C., Purdue was using the localized data produced in RADARS to testify in congressional hearings that OxyContin abuse was not an "epidemic," but rather an issue tied to specific localities.
Again, in RADARS's pharmacovigilance, strategic ignorance produced partial knowledge about OxyContin that limited the FDA's (and the public's) view of the true scope of the problem. What's more, the production of this limited knowledge worked to Purdue's benefit, allowing it to mask a surging opioid crisis while continuing to promote its newest product to chronic and acute pain sufferers across the country.
Ending the Opioid Crisis
If we are to understand how the opioid crisis evolved and why it looks the way it does today, we have to consider the FDA, because any pharmaceutical drug that enters the market must first pass through their regulatory apparatus. Yet even knowing this, it is still difficult to comprehend how this level of FDA-approved disaster could be achieved so easily, and why it continues decades later. One tool that can help us account for these questions is strategic ignorance. The systemwide production of strategically limited knowledge has, at nearly every step of the regulatory process, blinded regulators to the harsher realities of painkiller use. Perhaps the first step toward stopping the crisis is recognizing that strategic ignorance is not just noxious, but also forms part of a highly symbiotic relationship between our public health institutions and the interests of the pharmaceutical industry. If we do not find new ways of dismantling this relationship, it will continue wreaking havoc on the regulatory landscape, producing cycles of destruction that promise to thrust us deeper and deeper into the crisis.
Melina Sherman is the author of the forthcoming book How We Hurt: The Politics of Pain in the Opioid Epidemic. Click here to read more about her research on the opioid crisis.
Photo by Olga DeLawrence at Unsplash