The Perils of Cookbook Medicine

I have a confession to make. I am guilty of practicing cookbook medicine.

I make this confession with much angst and embarrassment. In my daily practice, I actively rail against making diagnoses by using "cookbook" recipes. I believe that doctors need to pay attention to the art of medicine that prioritizes the individual's story. In fact, I am writing a book on this topic!

Yet, this is a true story of what happened last week. It was 5PM, and I just started my evening shift.  The E.R. was in black “crisis” mode. The waiting room had been converted into patient bays, and there were patients out in the hallways stretching to the main lobby of the hospital.

I rolled up my sleeves and went to work. First, I met Mr. A, a middle-aged man with atypical chest pain for a day. Never had heart problems before but his father had an MI at age 50. Normal vitals and EKG. Check! Here was an easy dispo for low-risk chest pain. I sent off for cardiac enzymes and put Mr. A. in our observation unit for a second set and stress in the morning.

Next patient. Mr. B., a man in his seventies being treated for lung cancer who came in with fever and a productive cough. He was tachycardic, looked weak, and his lungs sounded junky. Again, easy dispo. Labs, x-ray, antibiotics for pneumonia, and oncology admission.

And the next. For kicks, let’s call her Mrs. M. A. a well-appearing woman in her sixties transferred from an outside hospital to get an MRI for a one-hour episode of aphasia eight hours ago. When I saw her, she talked fluently and had no neuro deficits. Once again, I knew what to do. I ordered an MRI, called neurology, and put her into the observation unit to await the scan and neuro consult.

My shift started just 15 minutes ago, and I had dispos on three patients! But something was wrong with this picture. While a lot of attendings would have applauded my actions as “efficient” and “having solid plans,” my very astute attending that day frowned. “You’re seeing patients fast and that’s good,” she said, “But think about what they have. Don’t let your practice become algorithmic.”

She was right. In my misguided attempt at becoming faster to and help with ED flow, I was evaluating my patients based on how quickly I could send them somewhere else, instead of stopping to figure out what they actually came in with. My reaction to each patient was reflexive, almost as if I were doing a multiple-choice board exam. Low-risk chest pain? To the observation unit Mr. A. goes. But had I even heard him talk about how the pain got worse after eating, how it came and went, how it traveled to his right side and back? Fever and productive cough? Must be pneumonia. Or maybe not. After Mr. B. undressed, I went back and saw that he had raging erythema covering one of his legs. He’d been telling me about scratching his legs but I hadn’t listened—and I could have missed diagnosing and treating a bad cellulitis.

And Mrs. M. When I went back to ask her what exactly happened, she said that she and her husband were driving on the road when she forgot the names of two streets she had lived on when she was young. She thought they were on the tip of her tongue but she couldn’t quite remember them. She panicked and her husband drove her to the nearby hospital. An hour later, she finally recalled the names of the streets. Never during this time did she have slurred or unclear speech. This was the “aphasia” that the other hospital had sent her in for, that was not actually aphasia and definitely not a stroke! But the joke was on me—I hadn’t done my own evaluation, and had asked for a consult and ordered a scan that Mrs. M. didn’t need at all.

By its high-stress, high-intensity nature, emergency medicine is a challenging and demanding profession. We have to make quick decisions with limited information, while seeing many complex patients at once. Being efficient is important, and we often feel the temptation to take shortcuts on the history and physical and call a consultant to figure things out instead of working things out for ourselves. But this practice of reflexive, algorithmic medicine doesn’t lead to better outcomes, and, I would argue, isn’t any faster than thoughtful, common sense medicine. Having our patient be "ruled out" for a heart attack while he has gallstone pain doesn’t help anyone. Missing a raging infection in could lead to bad outcomes. And doing hours of unnecessary tests in a woman who doesn’t have a disease based on history alone is a waste of time and resources, and potentially harmful to the patient.

So how can we avoid the tempting low-hanging fruit of cookbook medicine? One tip is to always think about your patient’s diagnosis. “Chest pain” and “abdominal pain” are symptoms, not diagnoses. Ask yourself each time what the patient has. Most of the time, there is a diagnosis. Some of the time, you may not know the exact diagnosis after your evaluation, but at least thinking about the diagnosis—and not just what the patient DOESN’T have—forces you to break out of comfortable cookbook pathways before ordering tests and thinking about dispos.

Something else that’s helpful is to explain your thought process to the patient. Patients like it when you talk to them about what tests you are ordering and why, and what diagnoses you are considering and why. If you find yourself unable to justify these tests to your patients, you might want to consider why you want them in the first place—perhaps you’ve unwittingly entered a cookbook practice and are reflexively (rather than thoughtfully) putting in orders.

Another tip is to try to figure out what the patient has before calling a consultant. The E.R. is the modern home of diagnosis; it’s exciting to take the first crack at a diagnosis and to send a patient to the floor with a diagnosis in hand! To that end, make sure to follow-up on your patients. You will learn an awful lot and be satisfied—and humbled—by your diagnostic acumen.

As we residents advance another year, it’s a good time to remind ourselves of the perils of cookbook medicine. Not only does it dehumanize our patients, cookbook practice leads to unnecessary tests and missed diagnoses, not to mention dissatisfaction—by both the patient and the practitioner. Let’s try our best to uphold the excellent standard of care that we are so proud of in our field, and practice the type of medicine that our patients deserve.

Parts of this article have been published in AAEM's Common Sense magazine and on Medscape.  I welcome your feedback and comments.