The legal and medical systems both strive for truth while acknowledging that there are no absolutes. Both systems require evidence, which they categorize in a hierarchy of levels, on which to base decisions that can have major effects on the quality and even quantity of people’s lives. In law, the strictest standard of proof applies in criminal matters, in which the presumption of innocence requires that guilt be established “beyond a reasonable doubt” to attempt to rule out the possibility of convicting an innocent person — though of course the application of this level of proof carries the risk of occasional acquittal of a defendant who is actually guilty. A lower, but still relatively stringent, standard of proof, that of “clear and convincing” evidence, applies to certain discrete civil matters and criminal matters such as the setting of bail. A still lower standard requiring conclusions based on a “fair preponderance of the evidence” applies in the great majority of civil matters, and an even lower standard in which only “probable cause” must be established permits certain criminal proceedings to be initiated. These levels of legal proof have analogies in medicine. Clinical trials use alpha (significance) levels, confidence intervals, and statistical power to gauge levels of certainty. To reject the null hypothesis (that a result occurred merely by chance) and deem an intervention effective in a clinical trial, the level of proof analogous to law’s “beyond a reasonable doubt” standard would require an extremely stringent alpha level to permit researchers to claim a statistically significant effect, with the offsetting risk that a truly effective intervention would sometimes be deemed ineffective. Instead, most randomized clinical trials are designed to achieve a somewhat lower level of evidence that in legal jargon might be called “clear and convincing,” making conclusions drawn from it highly probable or reasonably certain.

Although errors can be made by both the judicial system and the medical research system, the former provides the opportunity to appeal a court’s decision, and in the latter, reproducibility or independent confirmation of a result greatly enhances the reliability of findings. Unlike the categorical decisions of the courts, which immediately carry the weight of the law regardless of their popularity, the results of a clinical trial can have greater or lesser impact depending on their eventual degree of acceptance by the medical community. The data from clinical trials are generally initially disseminated in peerreviewed medical journals, at scientific meetings, or both. The ultimate influence of a study then depends on the interpretation of the importance of its results by national guideline-setting committees, as well as by more local physician groups at journal clubs, morning reports, and rounds. In such scholarly dissections of the trial data, statistical tests represent only one aspect of the intense scrutiny applied in assessing the quality and robustness of the findings.