The topic of how doctors think has been popular for several decades. Cognitive psychologists who study reasoning and decision making have found medicine to be a fruitful area of application, and many practicing physicians have enjoyed reflecting on their own thought processes. In How Doctors Think, Jerome Groopman combines these two lines of work, integrating clinical anecdotes with results from research studies. Groopman, a physician whose own work combines laboratory science and clinical practice, is well qualified to write this book, and readers will enjoy his wise advice to patients about how they can help their doctors to think better.

Much of the work discussed in the book is depressing, since one of the themes is the variability in the way that doctors respond to the same clinical situation. In one chapter, Groopman illustrates this by recounting his own frustrating experience as a patient dealing with severe wrist pain. In consulting five different hand surgeons, the range of advice that he receives is enormous, with some surgeons referring to medical phenomena that others denied being even physically possible. Moreover, the surgeon who made the best diagnosis did not utilize the most recent technology (he dismissed the results of the MRI scan of Groopman's hand), nor did he have the most experience (he had only performed the operation that he was proposing a single time). According to How Doctors Think, though, it is certain thinking styles, rather than technology and experience, which lead to good medicine.

Groopman emphasizes the role of several "cognitive errors"–problems in thinking–that lead to poor decisions, including diagnostic determinations and treatment plans. One of these is "search satisficing," in which, once we find an answer to a question, we stop searching for a better one. A physician who "explains" a stuffed nose and cough as being due to the common cold may fail to test for a fever (diagnostic of the flu). Similarly, a physician who finds a drug that is efficacious in treating a condition may fail to search for drugs that are more effective or that have fewer side effects. To guard against these events, Groopman suggests that patients be ready to ask their doctors such questions as "What else could these symptoms mean?" and "What are some alternative treatments for this condition?"

Another cognitive problem that Groopman discusses is one that he calls the "representativeness error," but that is akin to what psychologists have called "top-down thinking" or "confirmation bias." In any of these, initial information about a patient or their condition leads the physician to make a general judgment, and any evidence that is not consistent with that judgment is ignored, while evidence that seems to confirm the general judgment is eagerly sought. Groopman gives the example of an apparently vigorous and healthy man presenting to an emergency room with chest pain (a common presentation) that began while he was working outdoors. A physician impressed with the man's physical stamina never suspected that the pain signaled an impending heart attack (which occurred the next day), and he put too much faith in blood tests that were not sensitive enough to disclose the man's problems but whose negative results confirmed the initial judgment.

In addition to describing these and many other cognitive errors, Groopman attempts to find their personal and institutional sources. For instance, he reviews the ways in which medical schools train students to think about diagnosis in a linear fashion: physicians respond to a presenting symptom by asking a series of questions, with each answer narrowing the range of conditions that could be causing the symptom. This type of thinking works well for "typical cases"–individuals who have only one symptom, who do not have chronic medical problems that are difficult to separate from the acute complaints, and who are able to give accurate and detailed descriptions of their problems. But, of course, the average case may not be a "typical" one in many medical settings–many physicians deal daily with patients who have many symptoms, some of which stem from chronic conditions, and who, due to their discomfort, cannot provide the types of answers that help physicians trained to think in such a way.

A more destructive source of cognitive errors that Groopman discusses is money, whether in the form of insurance reimbursements or pharmaceutical company handouts. Interestingly, he does not accuse his colleagues of making diagnoses or endorsing treatments for any direct monetary gain. Instead, he offers a more complex chain of causation, in which drug companies give slanted information to physicians, who are then persuaded to behave in ways that benefit the companies. The lavish treatment of the physicians by the companies (e.g., expensive dinners and trips to exotic locales) does more to put the physicians in the appropriate mood to be persuaded.

Although much of this information is, as I said, somewhat depressing, the tone of the book is remarkably light and the writing is almost fluffy in terms of how many stories are used to illustrate simple, intuitive points. Each of the chapters opens with a case study of an actual patient, whose diagnosis (and sometimes treatment) is followed. The material that follows each case study is presented in the form of interviews with experts, more case anecdotes, or Groopman's personal reflections. All this makes for a book with a small amount of substantive content (few technical concepts and terms are actually explicated), but the low density of the material also makes for an enjoyable read for both patients and health care providers.

© 2007 Benjamin J. Lovett

Benjamin J. Lovett, Ph.D., is an assistant professor of psychology at Elmira College, where he teaches classes on a variety of topics in applied psychology and his research focuses on the conceptual and psychometric foundations of psychoeducational assessment and psychiatric diagnosis.

Comment on this review.

Categories: General, Psychology