Objectives: Transmucosal fentanyl is used to treat transient exacerbations of cancer pain. Several immediate release products are available, presented as intranasal sprays, sublingual and buccal tablets, or lozenges. These are not interchangeable, creating potential for medication errors. We compared the incidence of medication errors in a simulated scenario using handwritten drug charts and charts labelled with preprinted self-adhesive stickers with full pictorial fentanyl prescriptions., Methods: 54 nurses were shown 5 handwritten drug charts and 5 with self-adhesive pictorial labels. Nurses indicated which preparation and dose they would administer from boxes of Instanyl, Abstral, Effentora and Actiq (Nycomed, ProStrakan, Cephalon and Teva, respectively). We measured the frequency of drug administration errors and asked them to rate the prescriptions for clarity on four-point Likert items., Results: The use of pictorial self-adhesive prescriptions significantly reduced errors in choice of preparation, from 20 with traditional handwritten charts to 6 with self-adhesive labels (OR 3.52, 95% CI 1.39 to 8.90, p=0.006), but the incidence of dose error was not significantly different (OR 1.47, 95% CI 0.80 to 2.70, p=0.281). Analysis of Likert items showed using pictorial printed labels significantly improved nurses' understanding of choice of preparation, dose and maximum four hourly dose (p<0.0001, p=0.006 and p=0.028, respectively)., Conclusions: The use of pictorial prescribing appears to be a promising strategy that could reduce medication errors in choice of fentanyl preparations. There may be a wider use for pictorial prescribing where non-interchangeable preparations of the same drug exist., (Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://group.bmj.com/group/rights-licensing/permissions.)