Opioid safety among high-risk cancer survivors.

Abstract

e19169 Background: To mitigate risks of opioid-related harms, ASCO’s pain management guidelines for cancer survivors recommend that opioids be used in conjunction with other pharmacologic and non-pharmacologic approaches. The guidelines also recommend caution when prescribing opioids and benzodiazepines concurrently. We evaluated these 2 metrics of safe prescribing as applied to chronic opioid therapy (COT) among older survivors of head and neck cancer (HNC) and lung cancer (LC), two growing populations with high pain burden and prevalent risk factors for opioid-related harms (e.g., opioid use during treatment, history of substance use, distress). Methods: Using SEER-Medicare, we identified opioid-naïve adults diagnosed 2008-2015 with HNC or LC. We restricted analyses to survivors with ≥1 COT episode (≥90 days) occurring ≥1 year after cancer diagnosis and ≤120 days prior to hospice entry or cancer-related death (survivorship period). We report 2 opioid safety metrics during the survivorship period: 1) the proportion of survivors with non-opioid pain management (≥1 dispensing for a non-opioid, non-benzodiazepine pain medication or ≥1 claim for pain management procedure) concurrent with the first 90 days of the first COT episode and 2) the proportion of survivors with 0 dispensings for benzodiazepines within the first 90 days of the first COT episode. Results: Among opioid-naïve HNC (N = 5,500) and LC (N = 21,090) patients, 306 HNC (5.6%) and 927 LC survivors (4.4%) received COT during follow-up. Median duration of first survivorship COT episode was 5.2 and 4.9 months for HNC and LC, respectively. 64% of HNC survivors received non-opioid pain management concurrent with their first COT episode; 55% received an analgesic and 24% underwent a procedure. 75% of LC survivors received non-opioid pain management concurrent with their first COT episode; 67% received an analgesic and 35% underwent a procedure. 79% of HNC and 81% of LC survivors did not receive benzodiazepines during the first COT episode. Conclusions: Among older survivors of LC and HNC, less than 6% receive COT. However, of those, one-half of HNC survivors and more than a third of LC survivors receive guideline-discordant care by using COT without other pain management strategies or while using benzodiazepines. To minimize opioid-related harms, efforts should focus on improving safe COT prescribing practices for survivors. [Table: see text]