AI finds drug safety signals in clinical notes

Large language models were tested on cancer patient notes from two academic medical centers and seven drug trials.

Patients with atrial fibrillation face higher bleeding risks when treated with 2 common medications

While diltiazem remains an effective option for heart rate control in selected patients, bleeding risk posed by anticoagulants should be considered.

Study uses AI to cast light on harmful drug-drug interactions

Using AI, a team identified a gap in post-market drug safety: not all published reports of dangerous drug interactions are taken up by recognized purveyors of drug safety information.

Study of severe drug reactions finds skin cells activate their own assassins

Single-cell analysis reveals the inner workings of the life-threatening adverse drug reactions Stevens-Johnson syndrome and toxic epidermal necrolysis.

A clue to an adverse drug event in children

Considering metabolic (CYP2D6) enzyme activity score and patient age may aid in determining an individual’s risk for an adverse event with administration of the anti-arrhythmic drug propafenone.

Drug allergy labels in medical records power searches for gene-drug associations