Among GLP-1 users, oxycodone carried the highest short-term gut risk

GLP-1 receptor agonists slow the stomach; opioids slow the whole gut. A cohort study in Diabetes Care, published 2 September, asks a practical question that follows from putting the two together: among people already taking a GLP-1 drug, does it matter which opioid they are started on? The answer, at least for short-term gut complications, is yes [s1].

What the study did

The researchers built a new-user cohort from US insurance claims spanning 2016 to 2025, restricted to adults with type 2 diabetes who were taking a GLP-1 receptor agonist and then initiated one of three commonly prescribed opioids: oxycodone, hydrocodone or tramadol [s1]. The analysis covered 411,188 patients, mean age 62.8 years and 53.8% female [s1]. Of these, 24.4% started oxycodone, 48.5% hydrocodone and 27.1% tramadol [s1].

The outcome was a composite of motility-related gastrointestinal events — severe constipation, bowel obstruction and gastroparesis — assessed as the weighted absolute risk over 30 days, using propensity-score-matching weights to make the three groups comparable on measured characteristics [s1].

The numbers

The 30-day absolute risk of a motility-related GI event was 0.51% for oxycodone, 0.35% for hydrocodone and 0.33% for tramadol [s1]. Oxycodone carried a higher risk than hydrocodone (risk ratio 1.48, 95% CI 1.30–1.69; risk difference 0.17, 95% CI 0.11–0.22) and than tramadol (risk ratio 1.55, 95% CI 1.33–1.79; risk difference 0.18, 95% CI 0.12–0.24) [s1]. Hydrocodone and tramadol were statistically indistinguishable from each other (risk ratio 1.05, 95% CI 0.91–1.21; risk difference 0.02, 95% CI −0.03 to 0.06) [s1].

Two things are worth holding in view at once. The relative differences are real and reasonably precise. The absolute differences are small: even for oxycodone, roughly five events per 1,000 patients over a month, and the gap between opioids is a fraction of one percentage point [s1]. A signal can be genuine and still be modest in everyday terms.

Notably, the excess appeared to be driven by severe constipation and bowel obstruction; the authors did not observe a difference between opioids in the risk of gastroparesis specifically [s1].

Context: a known interaction, now made comparative

That GLP-1 drugs themselves raise GI risk is not new. A 2023 analysis in JAMA linked the class, when used for weight loss, to higher rates of gastroparesis, bowel obstruction and pancreatitis than an active comparator [s2]. What the new study adds is a within-class comparison on the other side of the interaction — holding GLP-1 use constant and varying the opioid — which is closer to the choice a prescriber actually faces when a patient on a GLP-1 drug needs analgesia [s1].

The biological rationale for the ranking is straightforward. Both drug classes slow the gut, but they do it through different routes: GLP-1 agonists delay gastric emptying, while opioids blunt motility throughout the intestine by acting on gut opioid receptors, an effect strong enough that opioid-induced constipation is a recognised clinical entity in its own right. Layering the two raises the odds that transit slows enough to matter. Among the three opioids studied, oxycodone is generally the more potent full agonist, whereas tramadol acts partly through non-opioid mechanisms and hydrocodone sits between — an ordering that broadly matches the risk gradient the data show, though the study measured associations rather than testing that mechanism directly [s1]. That the gap showed up in severe constipation and bowel obstruction, but not gastroparesis, also fits: the former are downstream, whole-bowel problems where an added motility brake would tell most [s1].

Limits

This is observational. Propensity weighting balances measured confounders, not unmeasured ones, and the reasons a clinician reaches for oxycodone rather than tramadol — pain severity, surgical context, prior tolerance — could themselves track GI risk [s1]. The window is short at 30 days, so the findings speak to acute events, not to what happens over months of concurrent use [s1]. The cohort is confined to insured US adults with type 2 diabetes, and the authors themselves flag that variation across secondary analyses warrants cautious interpretation of the hydrocodone–tramadol comparison [s1]. No trial randomised anyone to an opioid, and none should be read into this.

What to watch

The useful next step would be whether this ordering holds outside diabetes, over longer follow-up, and for GLP-1 drugs used at the higher doses now common in obesity care. For now the study supports a narrow, testable observation rather than a rule: in this population, the specific opioid chosen was associated with a measurable difference in short-term gut complications.

This article is informational and does not constitute medical advice. Decisions about analgesia belong with a prescribing clinician.

Sources

• [s1] Comparative Short-term Risk of Severe Gastrointestinal Events Associated With Opioid Type Among Patients Receiving Glucagon-Like Peptide-1 Receptor Agonists. Diabetes Care, 2 September
  1. https://doi.org/10.2337/dc26-0757
• [s2] Risk of Gastrointestinal Adverse Events Associated With Glucagon-Like Peptide-1 Receptor Agonists for Weight Loss. JAMA, 5 October 2023. https://doi.org/10.1001/jama.2023.19574