This is unrepresentative publication of research reports that is not due to the quality of the research but to other characteristics, e.g., tendencies of investigators and sponsors to submit, and publishers to accept, “positive” research reports, e.g., ones that detect beneficial treatment effects of a new intervention. P