Design: Population-based cohort with propensity overlap weighting
Population: 258,624 adults after elective hip or knee arthroplasty, Ontario, 2013–2022
Primary outcome: Persistent postoperative opioid use
Effect: 6.0% vs 5.2%; aOR 1.08 (95% CI 1.01–1.15)

Using linked Ontario data, researchers studied 258,624 adults having elective hip or knee replacement between 2013 and 2022. One in eight filled a new extended-release opioid prescription within a week of discharge.

After overlap weighting, filling one was associated with slightly higher odds of persistent postoperative opioid use (6.0% vs 5.2%; aOR 1.08, 95% CI 1.01–1.15), 90-day readmission (aOR 1.09) and emergency visits (aOR 1.16). One-year health costs were lower, and days alive at home were marginally higher.

Guidelines already discourage extended-release opioids for acute pain. The harms here are small but consistent, and the cost finding does not outweigh them. This is observational; patients given these drugs may differ in ways weighting could not fix.

Avoid starting extended-release opioids for acute pain after arthroplasty.
Prescribe short-acting opioids at the lowest effective dose and duration, if needed at all.
Build discharge analgesia around paracetamol, NSAIDs where safe, and regional techniques.
Plan a clear stop date for any opioid prescribed at discharge.

Why it matters

It confirms with population data a prescribing habit guidelines already discourage.

The statistics, in plain English

An odds ratio of 1.08 is small: about one extra patient with persistent opioid use for every 125 given an extended-release prescription. Large numbers make such small differences statistically clear.