The Impact of Abuse Deterrent Technologies on Prescription Opioid Trends

Published Online: October 11, 2013
Howard Oremland, RPh, MBA; Lauren G. Gallagher, MPH; John Schlie, PharmD; Dick Creager, MD, MPA; Anthony Palmieri, PharmD, MBA; Olga S. Matlin, PhD; and Steven M. Kymes, PhD
The past 2 decades have been marked by an increased awareness of the need for improved management of chronic pain and the aggressive marketing of new opioid formulations. This has been associated with an increased prevalence in prescribing of opioids, alongside an increase in the abuse of prescription drugs. In response, the Centers for Disease Control and Prevention (CDC) has described prescription drug abuse as the fastest-growing drug abuse problem in the United States.1

Oxycontin (Purdue Pharmaceuticals, LP) is a controlled-release oxycodone first sold in 1995 and is today among the most widely prescribed opioid medications. The controlled-release formulation is expected to deter abuse, as the active ingredient is released over a specified time period, reducing the patient’s ability to seek a “high” by taking more medication than prescribed. However, prescription drug abusers can achieve an enhanced high by chewing, crushing and snorting, or liquefying the pills. To inhibit this behavior Purdue developed an abuse deterrent technology version of Oxycontin that was approved for sale by the US Food and Drug Administration (FDA) in April 2010; and in August 2010, the company began limiting sales exclusively to the abuse deterrent technology formulation.2 A 2012 survey of 2566 opioiddependent patients conducted in part to assess the impact of the abuse deterrent technology on opioid abuse behavior reported that following release of the abuse deterrent formulation, there was indeed a decline in Oxycontin as a primary drug of abuse. However, this was accompanied by increased use of hydrocodone and other oxycodone agents, as well as increased use of high-potency agents such as fentanyl, hydromorphone and heroin.3

We analyzed de-identified prescription drug claims to examine the impact of abuse deterrent formulations on the utilization of opioid medications. We specifically considered the impact of the introduction of an abuse deterrent formulation of controlled-release Oxycontin on medication utilization in a cohort of prescription drug benefit plan members who were long-term users of Oxycontin at the time of the release of the abuse deterrent formulation in August 2010.


This analysis was conducted using de-identified CVS Caremark pharmacy benefit claims data. Prescription drug benefit members who had at least 1 pharmacy claim in calendar years between 2009 and 2012 for medications commonly used to treat pain were selected for analysis (Table 1). Members with claims for antineoplastic medications were excluded to limit the analysis to members with pain not related to cancer. Our analyses were limited to opioid medications categorized as follows: Oxycontin, other oxycodones (including generic versions of Oxycontin), morphine sulfate, hydrocodone-acetaminophen combinations, and other opioid-based pain medications(eg, fentanyl, oxymorphone, tramadol).

We examined medication utilization rates among chronic and acute opioid users, measured as the number of prescriptions filled per 1000 eligible prescription drug benefit eligible members per year from 2009 to 2012. In addition, we examined utilization trends among chronic opioid users to identify changes in medication selection following introduction of abuse deterrent technology formulations in August 2010. For this second analysis we compared 2 cohorts: 1) patients who received at least a 90-day supply of Oxycontin during the 6 months before the introduction in 2010 of Oxycontin with abuse deterrent  technology; and 2) a comparison group of patients who received at least a 90-day supply of Oxycontin during the 6 months prior to August 2009. Both cohorts were limited to members for whom Oxycontin was the only pain medication used during the 6-month look-back period. The outcomes of prescription drug use in the 12-month follow-up period (August to July) were: 1) patient remained on Oxycontin only; 2) patient discontinued all prescription opioid medications; 3) patient changed to a prescription opioid medication other than Oxycontin; and 4) patient added another prescription opioid to the Oxycontin regimen. We compared thpattern of utilization in the post period between these groups using the χ2 statistic. All analyses were conducted using SAS 9.1 (SAS Institute, Cary, North Carolina).


During each year from 2009 to 2012, approximately 4 million unique members from the analytic data set fi lled at least 1 prescription for an opioid pain medication and met our inclusion and exclusion criteria. The average age of members with a prescription for pain medication was 44.7 years (standard deviation ± 18.3) and 56% of opioid users were female.

Of the total prescriptions filled for opioid medication, over 99% were fi lled at retail pharmacies, 52% were for 10 days or less, and an additional 13% were for 11 to 20 days. Hydrocodone-acetaminophen combinations were the largest category of opioids, accounting for over half of all opioids prescribed. The overall rate of opioid use decreased by 8.3% between 2009 and 2012 (Table 2), with the even larger decreases observed in utilization of Oxycontin, morphine sulfate, and  hydrocodoneacetaminophen combinations (19%, 10.5%, and 12.9% decreases, respectively).

The results of our analyses comparing a cohort of members who were Oxycontin users at the time abuse deterrent technology was introduced in August 2010 to a similar cohort of users in August 2009 are presented in Table 3. In both cohorts,  approximately 10% discontinued use of all opioid medications during the year of follow-up. In the cohort that were chronic Oxycontin users in August 2009 (the year prior to the introduction of the abuse deterrent formulation), 2.2% of members switched to another opioid medication during the subsequent 12 months. Among members who were chronic Oxycontin users in August 2010 (when the new technology was introduced), 2.6% switched to a new opioid after the introduction of the abuse deterrent formulation. The differences between these cohorts were not statistically significant (P = 0.68).

Discussion of Findings and Pipeline Therapies

PDF is available on the last page.

Issue: September/October 2013
More on AJMC.COM