MEDICAL EDUCATION Reviewing scientific papers

A recent informal review by the editors of this journal showed that on a scale of 0 to 5 the quality of 25 % of referees ' reports scored 2 or less. Little training is given in this country in how to review a scientific paper and, apart from how to judge trial design and data analysis, ' there is little help in the literature. The late Julius Comroe held seminars in the subject (personal communication). Fellows who attended were expected to have read all the references in the paper he had given them to review. Otherwise they were asked to leave! These doctors received formal training in all the aspects of critical appraisal of a scientific paper, something which perhaps should be encouraged for higher trainees in this country. about competence to review a paper within the time limit is appreciated by everybody. BIAS If the reviewer is aware that s/he could be biased- for whatever reason- then the manu-script should be returned. Bias could explain the disagreement between experts already mentioned. Institutions working on the same topic are hardly likely to provide unbiased reviewers of each others ' work. There is some suggestion that a competent reviewer in a non-competing field is more likely to provide a good quality review.