(59) Risk of Falls and Fractures with Concomitant Use of Prescription Opioids and Other Medications in Osteoarthritis Patients

Publication Number: 733

Background: Prescription opioids are used commonly in osteoarthritis (OA) patients for pain management. Concurrent use of other medications is often required to manage several comorbid conditions in these individuals. Co-prescribing opioids with other medications may increase the risk of falls or fractures in this high-risk group.

Objectives: To identify non-opioid medications associated with falls or fractures in OA patients on opioid therapy for at least 90 days.

Methods: Using two US healthcare claims databases (2003 – 2020) and a case-crossover design, we conducted a semi-automated evaluation of non-opioid medications and their association with a fall or fracture (outcomes) in patients on opioids. The analytic cohort consisted of OA patients ≥ 40 years who experienced a fall or fracture with at least 365 days of continuous enrollment and at least 90 days of continuous opioid use prior to the outcome date. The hazard window was defined as days 1-30 and referent window as days 61-90 before the outcome date. We quantified the odds ratio (OR) between each non-opioid medication and each outcome. Analyses were limited to non-opioid medications with at least 500 individuals with discordant exposure across the windows to avoid statistically unstable estimates. Case-crossover estimates in patients with no exposure to opioids were used to adjust for potential time-varying confounding in a case-case-time-control analysis. Results were evaluated in ascending order of p-value, and the false discovery rate (FDR) was used to account for multiple testing.

Results: We identified 41,693 OA patients who experienced a fall and 24,891 OA patients who experienced a fracture while on long-term opioid therapy. A total of 25 medications were evaluated for their potential association with falls when co-administered with opioids. The top three drugs associated with falls based on ascending p-value included meloxicam (adjusted OR 1.22, p-value < 0.01, FDR = 0.10), venlafaxine (adjusted OR 1.14, p-value = 0.18, FDR > 0.99), and clonazepam (adjusted OR 1.13, p-value = 0.20, FDR > 0.99). Among 13 non-opioid medications evaluated for their potential association with fractures, the top three drugs were losartan (adjusted OR 1.21, p-value = 0.04, FDR = 0.56), alprazolam (adjusted OR 1.13, p-value = 0.20, FDR > 0.99), and duloxetine (adjusted OR 1.11, p-value = 0.22, FDR = 0.95).

Conclusions: We identified several medications that may have precipitated falls or fractures in OA patients on opioid therapy. Further studies are needed to identify the true cause of the associations, which may include drug-drug interactions, drug-disease interactions, and worsening health status or OA progression.