Evidence-Based Diagnosis

Evidence-Based Diagnosis is a textbook about diagnostic, screening, and prognostic tests in clinical medicine. The authors’ approach is based on many years of experience teaching physicians in a clinical research training program. Although requiring only a minimum of mathematics knowledge, the quantitative discussions in this book are deeper and more rigorous than in most introductory texts. The book includes numerous worked examples and 60 problems (with answers) based on real clinical situations and journal articles. The book will be helpful and accessible to anyone seeking to select, develop, or market medical tests. Topics covered include: the diagnostic process, test reliability and accuracy, likelihood ratios, and ROC curves, testing and treatment thresholds, critical appraisal of studies of diagnostic, screening and prognostic tests, test independence and methods of combining tests, quantifying treatment benefits using randomized trials and observational studies, Bayesian interpretation of P values and confidence intervals and challenges for evidence-based diagnosis and evidence-based medicine.

Evidence-Based Diagnosis is a textbook about diagnostic, screening, and prognostic tests in clinical medicine. The authors' approach is based on many years of experience teaching physicians in a clinical research training program. Although requiring only a minimum of mathematics knowledge, the quantitative discussions in this book are deeper and more rigorous than those in most introductory texts. The book includes numerous worked examples and 60 problems (with answers) based on real clinical situations and journal articles. The book will be helpful and accessible to anyone looking to select, develop, or market medical tests. Topics covered include:

Doody Review Services

Reviewer:Martha L Carvour, BS(University of Iowa College of Public Health)
Description:This book provides a working statistical framework for the assessment and integration of diagnostic tests in the clinical setting. Topics include the interpretation of diagnostic and prognostic data, including information derived from routine clinical tests as well as the medical literature. All topics are presented with the clinician's perspective in mind, using clinically relevant in-text examples and multiple sample problems with each chapter.
Purpose:The authors set out to demystify the statistical principles underlying evidence-based medicine, particularly those related to diagnostic testing, in a format accessible to clinicians. Effective teaching tools and resource materials in this field are widely sought after. This book may offer a useful supplement to an evidence-based medicine curriculum but will be less useful as a reference guide for busy clinicians.
Audience:The authors apply years of experience in both the clinic and classroom to address a broad clinical audience. The book is geared toward residents, fellows, and junior faculty, although students could make use of it early in their evidence-based medicine training. The sample problems may also be useful during preparation for board exams. To many in the clinical arena, this may come across as a clinically relevant but statistically daunting book.
Features:Topics covered range from common terms in evidence-based medicine -- sensitivity, specificity, and positive and negative predictive values -- to those less thoroughly addressed by some resources in the field, such as measures of reliability, likelihood ratios, and propensity scores. The authors provide many illustrative, clinically relevant examples and sample problems, along with a detailed answer key. On the whole, however, the figures are a bit text-heavy (or data-heavy) for the intended audience.
Assessment:While this book is not intended to review all concepts in a complete evidence-based medicine curriculum, it may serve as a useful supplement for existing curricula or for board exam preparation. The format will appeal most strongly to readers interested in the statistical underpinnings of evidence-based diagnosis. However, sections pertaining to critical appraisal of the clinical literature and interpretation of p-values and confidence intervals may be worthwhile to readers at any stage in their clinical training.

Thomas B. Newman, currently Chief of the Division of Clinical Epidemiology and Professor of Epidemiology and Biostatistics and Pediatrics at the University of California, San Francisco, previously served as Associate Director of the UCSF/Stanford Robert Wood Johnson Clinical Scholars Program and Associate Professor in the Department of Laboratory Medicine at UCSF. He is a co-author of Designing Clinical Research and a currently practicing pediatrician.

Michael A. Kohn is Associate Professor of Epidemiology and Biostatistics at the University of California, San Francisco, where he teaches clinical epidemiology and evidence-based medicine. He is also an emergency physician with more than 20 years of clinical experience, currently practising at Mills-Peninsula Medical Center in Burlingame, California.