Clinical trials produce results. From these results investigators will draw one of two conclusions: "experimental" and "control" treatments produce different outcomes (a "positive" conclusion), or experimental and control treatments produce similar outcomes (a "negative" conclusion). Because trials are always performed on a subsample of all patients, these conclusions may be wrong. We may draw the "false-positive" conclusion that the experimental treatment works, when, in truth, it does not; and we may draw the "false-negative" conclusion that the experimental treatment does not work, when, in truth, it does.
When the risk of drawing a false-positive conclusion (a type I error) is unacceptably high, it is tempting for investigators to conclude that the experimental treatment does not make a difference. However, statisticians have heightened our awareness of the illogic of this conclusion. A high risk of a false-positive conclusion does not guarantee a low risk of a
Detsky AS, Sackett DL. Establishing Therapeutic Equivalency: What Is a Clinically Significant Difference? Arch Intern Med. 1986;146(5):861–862. doi:10.1001/archinte.1986.00360170057006
Coronavirus Resource Center
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: