A few weeks ago, I was chatting with another physician in the emergency department after finishing my shift. He had thrown a birthday party for his nine-year-old son Jake, who had recently recovered from a sports-related ankle fracture. “During the party, a parent pulled me aside and said he had recently injured his shoulder doing yard work,” my colleague said. “I assumed he wanted me to examine his shoulder or give him advice – but instead he explained that he didn’t have time to see his physician, and he asked whether we had any Vicodin left over from Jake’s surgery. He was pretty casual about it.”

Situations like this are surprisingly common. National data suggest that two-thirds of patients who abuse prescription opioids obtain these medications from people they know, such as friends and family members, rather than seeking medications directly from a doctor. The FDA strongly supports patient disposal of unused opioid pills to prevent “diversion” of pills. The favored approach for drug disposal is to mix the pills in coffee grounds and place them in the trash. Unlike with other medications, the FDA suggests that patients may even dispose of unused opioids by flushing them down the toilet, a recommendation that reflects the substantially higher risk of keeping opioids in the home. The Drug Enforcement Agency hosts twice-annual “drug take-back days” at which patients can drop off medication for disposal.  And since 2010, the Secure and Responsible Drug Disposal Act has allowed retail pharmacies (and several other health care entities) to provide drug drop-boxes and mail-back packaging for patients to dispose of prescription opioids.

Despite these options, patients often indicate to researchers that they intend to keep unused opioid medications. On the surface, this is a rational decision: patients may believe that opioids may have substantial future value in the case of a painful injury, and disposing of these medications may be perceived as wasteful. (“I paid for that prescription! Would you flush money down the toilet?”) Yet this calculus ignores the harms that may occur from keeping opioid pills in the medicine cabinet.

How big is this problem? In early 2015, we received pilot funding from LDI to estimate the number of opioid pills left unused following a common outpatient surgery: tooth extraction. We enrolled 79 dental patients over six months. Our results showed that patients used an average of 46% (13/28) of the opioid-containing analgesic pills prescribed for them after surgery, and 1,010 pills remained unused among this group. If generalized to all patients, these results suggest that more than 100 million opioid analgesic pills remain unused following tooth extractions in the United States each year.

We also explored ways to promote drug disposal. Although the Drug Disposal Act provided patients with more options, we were skeptical about uptake — if patients were already unwilling to put medications in the trash or flush them down the toilet, it seemed they would be equally unlikely to walk or drive to a pharmacy just to get rid of leftover pills. However, since retail pharmacies are commercial enterprises, we developed an incentive program in which patients receive a non-cash incentive for disposing of their medications, such as a store credit, along the lines of bottle recycling programs.

Pharmacies would benefit from this arrangement as well: patients may purchase other goods and services while disposing of medications, and the pharmacies could receive positive publicity for helping to reduce opioid abuse. (In February 2016, Walgreens announced it would install drug disposal kiosks in about 500 of its pharmacies across the country.) Since incentives may not always have the intended effect, we performed a randomized controlled trial to examine the potential impact of such a program.

We launched our trial among patients undergoing tooth extraction in early 2015. Study patients were randomized to receive either standard discharge instructions or those instructions plus an additional one-page sheet describing a new program in which patients could receive store credit for returning their unused opioids. (Since no drug disposal programs existed at the time of the study, we included a “hotline” for patients to call if they were interested in the program. Patients who called this number were connected to members of the study team, who provided them with information on safe drug disposal techniques). We used an automated text messaging service to deliver surveys to patients on a daily basis for the first week after their surgery, followed by additional surveys at days 14 and 21 after surgery. These surveys asked patients to rate their pain level that day and to report the number of opioid pills they had used. After 3 weeks, we followed up by phone, and asked patients whether they had disposed of, or intended to dispose of, their unused pills. To encourage study participation, during study enrollment patients received a debit card with a $10 credit on it. They received an additional $3 credit for completing each text message survey and $10 credit for completing the follow-up interview.

We found that patients who received information on the (hypothetical) drug disposal program were 22% more likely to report disposing (or intending to dispose) of their opioids (52% vs. 30%). This difference did not reach statistical significance, but it may inform the designs of larger studies on drug-disposal incentives in the future.

This study would have been impossible without the pilot grant funding from LDI. I also received invaluable support from LDI Adjunct Senior Fellow Karin Rhodes, MD, as well as from Elliot Hersh, DMD, MS, PhD from the Departments of Pharmacology and Oral & Maxillofacial Surgery at the Penn School of Dental Medicine.

Brandon Maughan, MD, MSHP is an emergency physician and health services researcher at The Lewin Group, a health policy consulting firm. He conducted the study while serving as a Robert Wood Johnson Foundation Clinical Scholar at the University of Pennsylvania.