Methods and Applications of Statistics in Clinical Trials, Volume 1: Concepts, Principles, Trials, and Designs

Chapter 50

Nonrandomized Trials

Zhengqing Li

50.1 Introduction

A clinical trial, as defined by Friedman, Furberg, and DeMets, is “a prospective study comparing the effect and value of intervention(s) against a control in human subjects” [1]. Following this definition, a clinical trial must be prospective. Retrospective studies such as case-control studies in which subjects are selected on the basis of presence or absence of an event of interest do not meet this definition and will not be discussed here. Following this definition, a clinical trial must employ one or more intervention techniques and must contain a control group. Without an active intervention, a study is observational because no experiment is being performed. Without a control, there is no comparison group one can use to assess the effect of the intervention. We will be focusing on studies on human beings. Animal (or plant) studies will not be covered in the discussion although they may be studied using similar techniques.

This article is a tutorial description of nonrandomized trials under the framework of clinical trial definition. Basic concepts of studies, design features, statistical methods, and applicability of nonrandomized trials will be described and their limitations discussed. In addition, references are provided for readers who are interested in exploring relevant topics further. Readers are also encouraged to read articles under the topics of randomization, stratification, historical control, observational ...