A Clinician’s Guide to Statistics and Epidemiology in Mental Health: Measuring Truth and Uncertainty

There may be a professor of psychiatry out there who does a better job than Nassir Ghaemi in transmitting his wisdom directly to you - but in twenty years I have not found one. I have read the authors research papers for years. As an editor, I became familiar with his book "The Concepts of Psychiatry" as I considered the philosophical aspects of the field. His writing is always clear and his thinking consistently brilliant. In this brief volume on statistics and epidemiology his historical and original observations and descriptions of recent concepts is worth the price of purchase alone. A good example is his chapter on meta-analysis. He reminds the reader why this statistical method was invented in the first place and goes on to discuss significant limitations, significant historical critiques, and where the method might be useful. His opinions are well thought out and in a few brief pages he touches on issues that seem to be rarely discussed in the literature. This is an important chapter for a physician to read during a time when more and more meta-analyses are considered the gospel and end up as front page truths. He also provides a "defense and criticism" of evidence based medicine. He provides a philosophical context for the discussion and reminds us of "the cult of the Swan-Ganz catheter". Anyone who was an intern or resident in intensive care settings in the 1980s and early 1990s can recall the widespread use of this device despite the lack of evidence in randomized clinical trials (RCTs). It became the standard of care despite the lack of evidence. He pays homage to Feinstein his original observations that the evidence for evidence-based medicine goes beyond RCTs. The remaining chapters are concise discussions of statistics and epidemiology but they are anything but dry. An example would be his discussion of effect estimation and the number needed to treat or NNT method he describes the calculation and its advantages. He goes on to describe the meaning of particular numbers and also why the context is important. He uses a timely example of the issue of antidepressants and whether or not they lead to suicidality. This book succeeds as a volume that can rapidly bring the clinician and researcher up to speed on most current topics in statistics and epidemiology in medicine. It is not a book that reviews mathematical theory. It does not provide exhaustive calculations and examples. It is written for clinicians. It is a book that could provide a basis for discussion and seminars in this field for advanced residents using some of the author's references or recent literature searches to look at specific concepts. It could also be developed into a much more comprehensive text on the subject. Dr. Ghaemi brings a very unique viewpoint to the subject matter and he has produced a very readable book that I highly recommend. George Dawson, MD