You are here

Statistical Evidence in Medical Trials: What Do The Data Really Tell Us?

Stephen D. Simon
Oxford University Press
Publication Date: 
Number of Pages: 
Oxford Statistics
[Reviewed by
Sarah Boslaugh
, on

Stephen D. Simon is a research biostatistician at the Children's Mercy Hospital and a Associate Professor in the School of Medicine at the University of Missouri-Kansas City. He is well-known to web-surfing students and practitioners of statistics as "Professor Mean", where he fields questions such as "Can the p-value actually equal 1.0?" and "Could you explain what a Bonferroni correction is and why we want to keep scientists from fishing?" (See his web site at His answers are models of humor, clarity, and detail, qualities shared by his recent book, Statistical Evidence in Medicine: What Do the Data Really Tell Us?.

Simon wrote Statistical Evidence in Medicine to aid physicians and others in evaluating published medical research. It is therefore not an instruction manual in how to perform statistical procedures, but an guide to understanding and interpreting the medical literature. In fact, it is an outgrowth of a lecture he prepared for a class at the Children's Mercy Hospital, in response to requests for guidance in understanding the statistics used in medical journal articles. The book is about much more than statistics, however: its larger topic is really the clear application of logical thought and standard research methods to investigate medical issues and draw appropriate conclusions.

There are seven chapters, each covering a major topic in statistics and illustrated with numerous examples from published research. Simon's emphasis is on understanding the reasons behind different techniques and "rules" of research practice (such as randomization) and on evaluating whether an appropriate technique was used in a given circumstance. He provides lists of questions to ask of specific types of studies, and each chapter concludes with exercises which require the reader to apply the critical techniques presented in the chapter. He also provides a "counterpoint" section in each chapter, discussing the arguments against techniques which are the topics of the chapters.

The topics covered in Statistical Evidence in Medicine include the use of control groups and randomization, external validity, effect size and clinical significance, evaluating causality using Hill's criteria, meta-analysis, explanations of common terms such as p-values and confidence intervals, and how to search for research articles. Although these are standard topics covered in many books, Simon's approach is fresh and often his suggestions are not what is commonly practiced (e.g., he suggests that a literature search not begin with PubMed). This book is appropriate for anyone who needs to understand medical research as part of their job. It can be understood by people with no statistics background and yet (despite having a PhD and MPH) I learned quite a bit from it. It is particularly appropriate for medical students and others who are just beginning to read and evaluate the medical literature.

Sarah Boslaugh, ( is a Senior Statistical Data Analyst in the Department of Pediatrics at the Washington University School of Medicine in St. Louis, MO. She wrote An Intermediate Guide to SPSS Programming: Using Syntax for Data Management for Sage Publications in 2005 and is currently writing Secondary Data Sources for Public Health: A Practical Guide for Cambridge University Press. She is also Editor-in-Chief of The Encyclopedia of Epidemiology which will be published by Sage in 2007.

1. Apples or Oranges? Selection of the Control Group
2. Who was Left Out? Exclusions, Refusals, and Drop-outs
3. Mountain or Molehill? The Clinical Importance of the Results
4. What do the Other Witnesses Say? Corroborating Evidence
5. Do the Pieces Fit Together? Systematic Overviews and Meta-analysis
6. What do All These Numbers Mean?
7. Where is the Evidence? Searching for Information