(104) Effectiveness of Gabapentinoids versus Muscle Relaxants on Opioid Discontinuation among Older Adults with Prolonged Opioid Use

Background: Older adults are at risk for adverse events when co-prescribed opioid analgesics with other central nervous system-acting drugs. Yet, it remains uncertain whether the co-prescription of gabapentinoids compared to muscle relaxants can facilitate the process of discontinuing opioids.

Objectives: To compare opioid deprescribing between adjuvant gabapentinoids vs. muscle relaxants in community-dwelling older adults with prolonged opioid use.

Methods: We conducted a retrospective cohort study using U.S. fee-for-service Medicare data (2013-2018) and Optum Clinformatics Data Mart (2004-2023) to establish an active comparative cohort of adults aged 65 and older with a new use (no preceding use for 365 days) of gabapentinoids vs. muscle relaxants, following the commencement of a prolonged opioid analgesic treatment of ≥90 days. Exclusions were applied for cancer, palliative care, indications other than pain (e.g., seizure disorder, diabetes), and contraindications for opioid use like substance use disorder. The cohort entry date was the dispensing date of gabapentinoids or muscle relaxants. The outcome was time to discontinue opioids, defined by a dispensing gap exceeding 15 days. We used 1:1 propensity score (PS) matching to balance baseline measures of sociodemographic factors, health status, and healthcare utilization recorded during the year preceding cohort entry. We estimated adjusted hazard ratios (aHRs) and incidence rate differences per 1,000 person-years (aRD) using Cox proportional hazards regression and generalized linear models, respectively.

Results: Our Medicare cohort included 11,602 muscle relaxant initiators (median age 70 years, female 67%) and 14,949 gabapentinoid initiators (median age 72 years, female 67%). The Optum cohort included 15,226 muscle relaxant initiators (median age 70 years, female 64%) and 12,259 gabapentinoid initiators (median age 72 years, female 64%).
Censoring for death occurred in 2% of the Medicare cohort and 1% of the Optum cohort. The opioid discontinuation rate, per 1,000 person years, among new initiators of gabapentinoids versus muscle relaxants was 671.86 vs. 612.48 in Medicare and 789.84 vs. 759.16 in Optum. Compared to muscle relaxant initiators, gabapentinoids were associated with an increased likelihood of opioid discontinuation among Medicare (aHR, [95% confidence interval]: 1.06, [1.03-1.10]; aRD: 27.14, [14.04, 40.24]) and Optum (aHR: 1.02, [1.00-1.05]; aRD: 11.57, [-1.91, 25.04]) beneficiaries. These findings did not vary between those with versus without dementia.

Conclusions: Based on two US national cohorts, we found initiating gabapentinoids was associated with a higher likelihood of opioid deprescribing, compared to initiating muscle relaxants.