Tapentadol Linked to Lower Short-Term Fall Risk Than Oxycodone
Key Takeaways
- Among adults in New South Wales, Australia, who started publicly subsidized tapentadol SR or oxycodone CR products (including oxycodone CR/naloxone combinations), tapentadol initiation was associated with lower short-term risk of serious falls.
- Within 28 days, serious falls occurred in 0.4% of tapentadol initiators and 0.6% of oxycodone initiators, and the lower-risk association was also present at 7 and 14 days.
- The overall absolute difference was small at about 1 to 2 fewer serious falls per 1,000 patients starting tapentadol SR rather than oxycodone CR.
- Absolute differences were larger in adults aged 80 years and older, especially without recent opioid exposure, while the association was attenuated in older adults with recent opioid exposure.
The primary outcome was a composite of fall-related emergency department (ED) presentation, hospitalization, or death within 7, 14, and 28 days of initiation. After one-to-one propensity score matching, the cohort included 103,758 initiators in each group, and about 40% had non-study opioid exposure in the prior 90 days; subgroup analyses focused on older age and recent opioid exposure.
In the matched cohort, the 28-day serious fall comparison of tapentadol SR and oxycodone CR identified 457 falls after tapentadol initiation and 652 after oxycodone initiation, for a relative risk of 0.70 (95% confidence interval 0.62 to 0.79). This direction of association favored tapentadol for short-term serious falls.
The lower-risk association was also present at 7 days, with a relative risk of 0.57 (95% confidence interval 0.48 to 0.69), and at 14 days, with a relative risk of 0.62 (95% confidence interval 0.54 to 0.71). In absolute terms, that was about 1 to 2 fewer serious falls per 1,000 patients treated overall, with larger absolute differences in older adults; among adults aged 80 years and older, the 28-day risk difference (RD) was -0.0079 (95% confidence interval -0.0115 to -0.0039) and the number needed to treat to harm (NNTH) was 127 (95% confidence interval 87 to 257).
Immediate-release tapentadol use was not captured, dispensing records could not confirm ingestion, dose data were unavailable, and the endpoint captured falls that led to ED care, hospitalization, or death rather than all falls. Prior fall history also may have been incompletely recorded, and because the design was observational, the findings cannot establish causality; still, sensitivity analyses were consistent, adjustment for same-day non-study opioid initiation minimally changed estimates, and the cataract-surgery negative control showed no difference between groups.
In routine-care data from New South Wales, Australia, tapentadol SR initiation was associated with lower short-term serious-fall risk than a pooled comparator group of controlled-release oxycodone products, including oxycodone/naloxone combinations. The overall absolute difference was small, while larger absolute differences appeared in the oldest groups, particularly in this Australian publicly subsidized prescribing setting.
Clinician Questions
What counted as a serious fall in the tapentadol SR versus oxycodone CR comparison?
In the tapentadol SR versus oxycodone CR comparison, a serious fall was defined as a fall-related emergency department presentation, hospitalization, or death identified through linked administrative records. The endpoint therefore captured falls that reached health-system contact rather than all falls after treatment initiation.
How large was the overall absolute difference in serious falls after starting tapentadol SR instead of oxycodone CR?
Across the matched New South Wales cohort, starting tapentadol SR instead of oxycodone CR was associated with risk differences of -0.0014 at 7 days, -0.0018 at 14 days, and -0.0019 at 28 days, with corresponding NNTH estimates ranging from 739 to 529. In plain language, that was roughly 1 to 2 fewer serious falls per 1,000 patients starting tapentadol SR rather than oxycodone CR.
How did recent non-study opioid exposure change the fall-risk pattern in adults aged 80 years and older?
Lower serious-fall risk with tapentadol SR versus oxycodone CR was reported regardless of prior 90-day exposure to non-study opioids, but the largest absolute differences were seen in opioid-naive adults aged 80 years and older. In that subgroup, the estimated NNTH was 173 at 7 days and 97 at 28 days, whereas the association was attenuated in adults aged 80 years and older with recent non-study opioid exposure.
What analyses supported the stability of the tapentadol SR fall-risk signal?
Sensitivity analyses were consistent with the primary analysis, and adjustment for same-day initiation of non-study opioids minimally changed the estimates. Cataract surgery as a negative control showed no difference between groups, follow-up time was similar between groups, and very few serious falls resulted in death.