The aim of this study was to compare the performance of two formats (prose and flow diagram) of the guidelines for management of paracetamol poisoning, and to assess the likely performance without access to the guidelines. A prospective questionnaire study of the management of seven hypothetical cases of paracetamol ingestion was carried out by accident and emergency senior house officers at a regional induction course. No differences were found between the two formats. The proportion of correct answers was 37% in the flow diagram and 31% in the prose group (95% confidence interval for the difference -8% to 20%). In the group with neither format of the guideline the proportion of correct answers was lower: 19% (95% confidence interval for the difference between this group and the group with flow charts 6.9% to 30.6%, for the difference between this group and the group with the prose format 0.4% to 24.8%). The time taken to answer the questions did not vary between the groups. These data do not support the exclusive use of either format. They suggest that management of paracetamol poisoning is less likely to be correct if staff do not have access to the guidelines.