Patient on a GLP-1 agonist needs an opioid? Oxycodone carried the highest bowel risk

Design

Population-based new-user cohort, propensity-score weighted

Population

411,188 US adults with type 2 diabetes on a GLP-1 agonist starting an opioid

Primary outcome

30-day severe constipation, bowel obstruction or gastroparesis

Effect

0.51% oxycodone vs 0.35% hydrocodone vs 0.33% tramadol; RR 1.55 (1.33 to 1.79) vs tramadol

A new-user cohort study in Diabetes Care (September 2026) used US insurance claims from 2016 to 2025 to follow 411,188 adults with type 2 diabetes on a GLP-1 receptor agonist who started oxycodone, hydrocodone or tramadol. The outcome was a 30-day composite of severe constipation, bowel obstruction and gastroparesis, with the groups balanced by propensity-score weighting.

The 30-day absolute risk was 0.51% with oxycodone, 0.35% with hydrocodone and 0.33% with tramadol. Oxycodone was associated with higher risk than hydrocodone (RR 1.48, 95% CI 1.30 to 1.69) and tramadol (RR 1.55, 1.33 to 1.79) — about 1.7 to 1.8 extra events per 1,000 patients. The excess was in constipation and obstruction; gastroparesis did not differ. Hydrocodone and tramadol were similar.

It is observational, so residual confounding by pain severity is possible — oxycodone may be chosen for worse pain. The absolute difference is small. But when the opioid choice is otherwise equal, it is an easy decision to make well.

Where the practical choice is between oxycodone and tramadol, this favours tramadol for the bowel — but tramadol brings its own cautions: serotonergic interactions, a lowered seizure threshold and hypoglycaemia.

• When a patient on a GLP-1 agonist needs a short opioid course and the options are otherwise equal, prefer tramadol or hydrocodone to oxycodone
• Co-prescribe a laxative with any opioid in a patient on a GLP-1 agonist
• Keep the opioid course short and review within a week
• Tell patients to report no bowel movement for three days, bloating or vomiting
• With tramadol, check for serotonergic drugs and warn about hypoglycaemia

Why it matters

Opioid choice is rarely thought of as a diabetes decision, yet with incretin therapy now common it shifts the risk of a serious bowel complication.

Don't overread it

This is an association in claims data; confounding by pain severity cannot be excluded, and the absolute difference is under 0.2 percentage points.

The statistics, in plain English

A risk ratio of 1.48 means about 48% more events with oxycodone than hydrocodone, but because the events are rare the risk difference is 0.17 percentage points — about one or two extra events per 1,000 patients over 30 days. Hydrocodone versus tramadol (RR 1.05, 0.91 to 1.21) crosses 1, so those two cannot be told apart.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.