Unintended Consequences of Opioid Policy on Disparities in Pain Management for Older Adults with Serious Illness

Outcomes 1. Characterize opioid and anticonvulsant (gabapentin, pregabalin) prescribing trends from 2010-2020 among older adults with dementia (PWD) and older adults with cancer (PWC).2. Identify the association of the 2016 CDC opioid guideline with racial/ethnic disparities in opioid prescribing among PWD and PWC. Key Message Opioid prescribing for older adults with dementia or cancer rose from 2010 until a peak in 2013-2018 and declined thereafter to a nadir in 2020. Anticonvulsant prescribing continued to rise for these populations during that time, suggesting that pain management may be shifting from opioid to anticonvulsant therapy. Introduction In response to the opioid epidemic, the CDC published opioid prescribing guidelines in 2016 cautioning against opioid use, particularly for adults with chronic, non-cancer pain. However, little is known about the unintended effects of such opioid policy on pain management for older adults with serious illness, who commonly have moderate-to-severe opioid-responsive pain (e.g., cancer, dementia)[1-3] and are at increased risk of pain undertreatment (e.g., dementia, racial/ethnic minorities).[4-7] Objectives 1) Characterize opioid and anticonvulsant (gabapentin, pregabalin) prescribing trends from 2010-2020 among older adults with dementia (PWD) and cancer (PWC). 2) Identify the association of the 2016 CDC guideline with racial/ethnic disparities in opioid prescribing among PWD and PWC. Methods This retrospective observational study of older adults (age ≥65) with dementia or cancer utilizes 2010-2020 data from Medicare Current Beneficiary Survey, a nationally representative, longitudinal survey of Medicare beneficiaries linked to Medicare claims. We used a difference-in-differences design to determine whether the 2016 guidelines were associated with reduced opioid prescribing for PWD and PWC and worsened racial/ethnic disparities in opioid receipt. Results The annual proportion of PWC (n=18,292) prescribed an opioid rose from 21.9% (n=720/3,237) in 2010 to a peak of 26.4% in 2013 and 2015 (n=923/3,395 and 800/3,170, respectively) and declined to a nadir of 15.4% in 2020 (n=538/3,120). Opioid prescribing for PWD (n=6,056) rose from 14.2% (n=121/960) in 2010 to 20.0% in 2018 (n=199/947), and then decreased to 15.4% in 2020 (n=123/806). Anticonvulsant prescribing rose consistently from 2010-2020, doubling for PWC (6.9% [n=237/3,237] to 13.4% [n=428/3,120]) and nearly tripling for PWD (5.6% [n=48/960] to 14.0% [n=105/806]). Remaining results expected by March 2024. Conclusion Opioid prescribing for PWD and PWC has declined since its peak in 2013-2018, while anticonvulsant prescribing continued to rise since 2010. These findings may indicate a shift in pain management from opioids to less-effective anticonvulsant therapy. Keywords Scientific Research / Diversity, Equity, Inclusion, Belonging, Justice