Concurrent Use of Prescription Opioids with Other Medications and Risk of Fall and Fracture in Rheumatoid Arthritis Patients

Background: Nearly 70% of individuals with rheumatoid arthritis (RA) use five or more medications concurrently, including prescription opioids for pain management. However, the safety of drug combinations involving opioids has not been evaluated with respect to fall or fracture risk.

Objectives: To identify medications that may increase the risk of falls or fractures in RA patients taking prescription opioids continuously for at least 90 days.

Methods: We used a case-crossover design to screen two commercial claims and Medicare databases (2003 to 2020). This approach selects RA patients who had a fall or fracture while on opioids and assesses exposure to non-opioid medications during a hazard window (days 1-30 before the event) and referent window (days 61-90 before the event). RA patients aged 18 years or older (65 years or older in Medicare) were required to have at least 365 days of continuous enrollment and at least 90 days of continuous prescription opioid use before a fall or fracture. We quantified the association between each drug pair and outcome using the odds ratio (OR). We also incorporated a control group of patients who experienced fall or fracture without opioid use in the prior 90 days to account for potential confounding by indication in a case-case-time-control analysis. We further limited analyses to drug pairs with at least 500 individuals for sufficient statistical power. Results were evaluated in order of ascending p-value, and the false discovery rate (FDR) was used to adjust for multiple testing.

Results: We identified 37,389 RA patients who experienced a fall and 17,991 RA patients who experienced a fracture while on opioids. Among 37 medications evaluated for fall risk, fluticasone (adjusted OR 1.16, p-value = 0.05, FDR > 0.99), losartan (adjusted OR 1.12, p-value = 0.17, FDR > 0.99), and carvedilol (adjusted OR 1.12, p-value = 0.18, FDR > 0.99) were highly ranked as associated with fall. Among 23 medications evaluated for fracture risk, hydrochlorothiazide (adjusted OR 1.31, p-value = 0.03, FDR = 0.67), clopidogrel (adjusted OR 1.27, p-value = 0.07, FDR = 0.76), and duloxetine (adjusted OR 1.23, p-value = 0.07, FDR = 0.51) were highly ranked as associated with fracture.

Conclusions: In this screening study of RA patients, we identified several drugs that may increase the risk of fall or fracture when taken concurrently with prescription opioids. These medications may indicate drug-drug interactions, chance findings, or acute changes in underlying health status immediately before the event.