The main argument is that most medical mistakes are not obvious stuff ups: "...technical errors account for only a small fraction of our incorrect diagnoses and treatments," he writes. "Most errors are mistakes in thinking. And part of what causes these cognitive errors is our inner feelings, feelings we do not readily admit to and often don't even recogonise" (p40).
For example, a lot of doctors are quite averse to patients who are very ill - possibly because they have feelings of failure when dealing with diseases that resist their efforts to cure the patient. So in some cases - like it or not - they subconsciously avoid these patients, or stop trying. Groopman also points out that doctors (and vets, if you ask me) can become wedded or committed to a particular diagnosis, biasing their interpretation of clinical signs and response to treatment so that they pay attention to those things that confirm their diagnosis and ignore the others. Its too easy to judge a patient eg a dirty old man who is poorly clothed, rough around the edges and admits to sinking a glass of rum a day turns up with a liver problem and people assume he's an alcoholic with cirrhosis because they just want to get the diagnosis made and turf him out the door.
Another common source of error is the tendency to judge the likelihood of a diagnosis "by the ease with which relevent examples come to mind" (p64). For example, you think a patient has subclinical viral pneumonia because you've been treating a lot of those lately...when the symptoms could be explained by something altogether differently. In particular, the last BAD experience you have had with a particular illness always springs to mind. Little wonder that! They're the ones you can't forget because they keep you up at night.
Groopman argues that we should always consider a list of differential diagnoses, even when we think the diagnosis is obvious. I could not agree more. Another helpful strategy is to ask yourself, when you see a patient, "what is the worst thing this could be?"and "what ELSE could this be?" It helps you think broadly, more outside the square, relate the clinical signs to pathophysiology and ultimately not delay diagnosis in grave cases. He also argues that it is helpful to ask what parts of the patient's body occur near the symptom (ie a source of pain).
A pediatrician tells him that they always ask whether the child could have a serious problem - so they don't miss anything. I guess pediatricians are closest to vets in a way - neither can ask their patient where it hurts, the history is gleaned from a carer who may exaggerate or ignore clinical signs and we have to be advocates for them.
There's a great discussion of types of medical uncertaintly: "The first results from incomplete or imperfect mastery of available knowledge. No one can have at his command all skills and all knowledge of the lore of medicine. The second depends on limitations in current medical knowledge. There are innumerable questions to which no physician, however well trained, can provide answers. A third source of uncertainty derives from the first two: this is the difficulty in distinguishing between personal ignorance or ineptitude and the limitations of present medical knowledge"(p152).
Groopman argues that a doctor who is honest about their uncertainly has to be a better doctor - the patient will be more likely to trust them (I would argue that there are, however, some clients who are terrified by the words "I don't know"); it demonstrates a willingness to engage with the patient and change the course of treatment if it isn't working. All in all its a thought-provoking book. There are some very powerful anecdotes about the divergence of medical opinions. If I am ever diagnosed with cancer I'll certainly be seeking a second and third opinion before I commit to any treatment! But as a vet I think its worth a read. Vets and doctors are only human, after all. But some awareness of our fallibility can't hurt.