Evidence-based advertising? A survey of four major journals.

BACKGROUND Pharmaceutical advertisements are an important means of bringing drug information to physicians. Most advertisements are intended only to raise awareness, though there are those that do seek to persuade through presentation of research findings. Researchers have questioned the quality of the research reported in advertisements and wonder whether these advertisements would lead to improper prescribing. METHODS A consecutive 6-month sample of advertisements in 4 general medical journals, 3 from the United States and 1 from Canada, were reviewed to determine how research results are presented in pharmaceutical advertisements. RESULTS During this time there were 187 distinctive advertisements, with 43 data presentations in the 33 advertisements that contained quantitative research results. These results were examined using a critical appraisal worksheet. References to randomization and blinding were found in less than one half of the 43 data presentations. P values were frequently provided, but confidence intervals and references to power and number needed to treat were not provided in any of the advertisements. CONCLUSIONS Descriptions of research in pharmaceutical advertisements were brief and incomplete, and they inconsistently provided the basic design and statistical information needed to judge the results reported. More detail could make these advertisements more meaningful to critical readers.