Treatment decisions should be based on high-quality data, ideally obtained from randomized controlled trials. However, randomized trials are not immune from bias and other important limitations. Critical appraisal of such studies must consider both the methodological rigour of the study and the applicability of the results to routine clinical practice.Readers should work systematically through trial reports to establish the aims of the study and consider whether the methods used are able to provide an unbiased answer. Particular attention should be directed towards randomized patient allocation, ensuring that the treatment and control groups are equally balanced at the start. There should be adequate follow-up and sufficient blinding of the investigator and participants so that important outcomes are reliably recorded without any interference from preconceived notions. The trial’s Results section should be reviewed to confirm that the researchers have transparently reported all data (positive or negative) relevant to the study question. Critical appraisers should also consider how closely the conduct of the trial (e.g. selection of patients, follow-up arrangements) reflects real-world medicine, and the applicability of the trial results to everyday clinical practice.