In 2007, Purdue pleaded guilty to misleading consumers about the risk of abuse associated with the drug. The company paid $600 million in fines, and in 2010 it released a new formulation of the drug that made it extremely difficult to crush or dissolve the pills in an attempt to make it harder to abuse. It was the first drug to receive an "abuse-deterrent" designation from the FDA.
That reformulation was one of a number of steps taken by authorities across the country to limit the number of prescription painkiller overdose deaths, which by then had been trending sharply upward for more than a decade. Other interventions included the use of Prescription Drug Monitoring Programs (PDMPs) to track the sale of prescription painkillers, and a "pill mill" law in Florida passed in 2010 putting tighter regulations on pain clinics that in some cases dispensed painkillers recklessly.
Taken together, these interventions have been widely credited with staving off the rise of prescription painkiller deaths. But they came with an unintended side effect: a subsequent heroin epidemic that now kills more people each year than most prescription painkillers combined.
The working paper, published this week by the National Bureau of Economic Research, shows that the Oxycontin reformulation was by far the single largest driver of the shift to heroin, accounting for "as much as 80% of the three-fold increase in heroin mortality since 2010," the authors found.
They arrived at this estimate by looking at the differences in heroin deaths, post-2010, in states that had differing rates of OxyContin abuse prior to the 2010 reformulation.
If heroin deaths increased more in states that had large rates of OxyContin abuse before 2010, the authors reasoned, that would be a strong indicator that the reformulation drove more Oxycontin users to abuse heroin. And that's exactly what they found.
"States with the highest initial rates of OxyContin misuse experienced the largest increases in heroin deaths," they write. "Results show that this differential increase in heroin deaths began precisely in the year following reformulation."
Moreover, prior to 2010, there was no correlation between OxyContin abuse and heroin mortality. "Both the levels and trends in heroin deaths were nearly identical across states with high or low initial rates of OxyContin misuse before 2010," the study found.
The study determined that the OxyContin reformulation was indeed successful at reducing OxyContin abuse. But, "each percentage point reduction in the rate of OxyContin misuse due to reformulation leads to 3.1 more heroin-related deaths per 100,000."
The authors are confident the effects they observe are from the OxyContin reformulation and not from other efforts to crack down on prescription painkillers during that time. They controlled for the presence of Prescription Drug Monitoring Programs in different states and found it had little effect on their numbers.
They also re-ran the numbers to exclude Florida, as well as states like Kentucky and West Virginia, where people were known to buy drugs from Florida's "pill mills" which were largely shut down in 2010. But even then, their observed effect remained.
Overall, they estimate that the OxyContin change was responsible for 80 percent of the increase in heroin mortality after 2010. They stress that the change didn't appear to have any effect on the total number of overdose deaths observed after 2010 -- if fewer people died from OxyContin, more were killed by heroin. The overall trajectory essentially remained unchanged.
The authors position the effects of the OxyContin reformulation as a lesson in the limits of "supply-side" drug policies -- those that attempt to reduce drug abuse by cutting the supply of abused drugs, rather than the demand. "The findings from this study provide yet another example of how supply-side strategies alone are inadequate for dealing with the drug problem, particularly when substitute drugs exist," they write.
"That doesn't mean supply reduction is bad, you just need broader supply reduction policies," said co-author David Powell of the RAND Corp. in an interview. With the OxyContin reformulation, "they were basically just taking out one component of this abuse chain."
The FDA is actively encouraging the development of more abuse-deterrent opioids similar to the reformulated OxyContin. That could create a "tipping point," Powell said -- if an overwhelming majority of prescription opiates become harder to abuse, in the long term that may reduce the likelihood that people get hooked on any opioids to begin with. That could eventually lead to opioid overdose death rates declining.
On the other hand, "you're not gonna reformulate heroin," he added. As long as heroin remains an available substitute for opioid painkillers, cracking down on the latter will likely only lead to an increase in use of the former.
He and his co-authors would rather see a more holistic approach to drug enforcement that focuses more attention to reducing demand for illicit drugs to begin with -- treatment and prevention programs that either help people stop using drugs completely, or prevent them from starting in the first place.
"Treating underlying demand may prove to be the more effective strategy for dealing with the current opioid epidemic, particularly because substitutes are readily available," they conclude.