[Skip to Content]
[Skip to Content Landing]
Article
March 26, 1982

Studying a Study and Testing a Test: How to Read the Medical Literature

JAMA. 1982;247(12):1761-1765. doi:10.1001/jama.1982.03320370065043

This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.

Abstract

This book is the converse of the more numerous basic biostatistics texts extant: the latter begin by teaching statistics and extrapolate to their use in clinical studies. This book begins, as the title suggests, with clinical studies and then describes the statistical methods used in their interpretation. Statistics texts will give one a firmer grounding in methodology, but Riegelman's approach to understanding the strengths and limitations of published studies is eminently pragmatic.

Excellent is the book's initial section, "Studying a Study," in which the author deals in turn with sample selection, assessment (defining and collecting results), analysis of results, interpretation, and extrapolation from results. Included are explanations of cohort, case-control, retrospective, and prospective studies and of clinical trials, selection bias, type I and type II error, and the proper inference of causality. A better discussion of these concepts at the level most useful to the reader of medical literature would

×