Information contained in clinical trial reports

In this paper, we assess the information contained in two hundred and fifty nine reports of clinical trials published in six major medical journals in 1960, 1965, 1970 and 1975. A checklist was used, against which the presence or absence of various items of information was recorded. This method may have wider applicability for assessing the information content of publications which have easily defined aims. Information about the type of drug under trial, dosage sche-dules, duration of therapy and whether or not the results were statistically significant was seldom provided in the summaries of the reports. Less than half of the reports included details of (i) the duration of the patients ’ diseases under treatment; (ii) whether or not the patients were con-currently suffering from any other diseases; (iii) whether or not other drugs were being taken in addition to the trial drug; (iv) whether checks were made on patient compliance. In the light of our results, we suggest that clinical trial reports should be required to provide a specified minimum of clinically important information. This should enable doctors to assess more readily the value of newly introduced drugs