Evidence-based medicine

In this first article in the series on Evidence-Based Medicine for the rheumatologist, we discuss the basics of hypothesis testing. Hypothesis testing is the process by which one seeks to reject a certain claim, such as that the two groups in a clinical trial have the same outcome. The basics of hypothesis testing include types and distributions of data, Type I and Type II error, power and its implications for interpreting and designing studies, the choice of one-sided versus two-sided tests, and p values versus confidence intervals. We also discuss the difference between a clinically significant result and a statistically significant result. We hope that this article provides a framework from which to introduce other important topics such as comparing two or more groups of subjects, comparing survival rates of subjects, and comparing agreement between observers.