[Main biases in clinical research].

Abstract

In developing a research protocol, authors must consider the possible errors that may occur throughout the study. In clinical research, two types of biases are recognized: random errors and systematic errors; the latter are called biases. To date, dozens of biases have been described, which is why the purpose of this article is to describe the main biases that can occur in clinical research studies, as well as strategies to avoid them or to minimize their effects. Since there are several classifications, in order to provide a more practical overview in this review, the biases are grouped into three types: selection biases, information (or performance) biases, and confounding biases. In addition, to make it even more specific, we describe the biases considering the purpose of the research: prognosis, therapeutics, causality, and diagnostic test studies.