The Weight of a Term: "Substantial Evidence" and Buried Data

When Congress amended the Food, Drug, and Cosmetic Act in 1962 to ensure the efficacy of drugs before they reach the market, it imposed a standard of evidence distinctly weaker than the reasonable one of preponderance. The difference is material. Some drugs now on the market may or may not have a preponderance of trial evidence of efficacy. An obstacle to a finding of efficacy is the well-known nemesis of drug makers with a definite interest in favorable trials: the placebo effect. Trials where the placebo effect runs high are vulnerable to negative findings, and trials with such findings all too often find their way to burial sites like regulatory archives. Publication bias-the preferential reporting of positive findings-has been abetted by a regulatory standard that does not require a preponderance of evidence and discounts negative trials, provided only that two trials show positive results.