A brief exchange I had with a neurosurgeon in the comment section on KevinMD the other day left me pondering the diversity of skills needed in different types of medical specialties, and also how differently technology has impacted various specialties during my years in medicine.
Neurosurgeon F. X. Wall disagreed with the post author, Dr. James Aw, about the value of old-fashioned physical exam skills, because in neurosurgery the anatomical accuracy of interventions has approached 100% as a result of new technology.
I can see that in neurosurgery and many other surgical specialties the advances in imaging have made clinical exam skills too inaccurate to guide treatment in this day and age, just like few cardiologists would forego an echocardiogram in evaluating a heart murmur.
My reply to the neurosurgeon was:
“Good points, but possibly more relevant in specialty care. When a patient in primary care has nonspecific symptoms, like shortness of breath, we need doctors with enough clinical exam skills to notice pallor, prolonged expiratory phase, JVD, irregularly irregular pulse, tachycardia and all the other clues that help us decide what tests to order first.”
The more I thought about it, the more fundamental this seems to me: In primary care, we don’t have many technologies that make clinical exam skills entirely obsolete. When I see a patient in my office 20 miles from the nearest X-ray machine, when a simple lab test won’t be resulted for 6-24 hours, when there is almost no way I could get a same-day echocardiogram or MRI, clinical exam skills are essential.
Time and distance aside, primary care doctors also need enough clinical skills to either make the diagnosis without technology or at least to know which diagnostic possibilities to pursue before others; if we did every possible test in every case, we would obviously waste a lot of resources. Just like in my example of shortness of breath above, almost every presenting complaint in primary care has many diagnostic possibilities, ranging from trivial or self-limited to serious or even life-threatening.
The broad range of differential diagnoses to consider when we evaluate both common and unusual symptoms people see primary care providers for is something to consider when we look at what type of clinician we assign to front-line duty. In many practices, this task falls on the least experienced providers. This is also the case in some freestanding urgent care centers. Having more seasoned doctors available as back-up isn’t necessarily a good system if the clinician on the front line hasn’t seen enough to know what he or she doesn’t know.
There have been many attempts to use technology as a substitute for clinical experience in front-line medicine. In my opinion none have really emerged that can compare with the technological revolution we have seen in imaging, microsurgery or laboratory diagnosis.
Systems that require the clinician or the patient to enter data in order to produce differential diagnoses, for example, are clumsy and either simplistic or bogged down with detail, and assume that everybody shares language and values they in fact don’t. In real practice, the patient who says “it only hurts a little”, but whose pained or panicked facial expression makes the hairs stand up on the back of a seasoned doctor’s neck is not likely to be better diagnosed by today’s available technology.
Even in more technology dependent specialties, there are good reasons to cultivate low tech proficiency. What does a doctor do during a hurricane or an ice storm, during a war or on a foreign assignment when there is no technology available? Why would we not listen to hearts, lungs and peripheral blood vessels and then compare our impressions with the results of the imaging?
And, without excellent clinical exam skills, how do we evaluate unexpected or conflicting technology-derived results?
Ultimately, we need both hands-on and technical assessments in health care. But on the front lines, we are perhaps more dependent on our clinical assessment skills. I never get praised for ordering lots of tests, only for ordering the right one.
A Country Doctor Writes: 
Appalling. I rotated with a neurosurgeon but used it as an opportunity to learn when to manage back pain conservatively and when to refer. One day a patient came in with back pain. Take her story and do the 2-3 minute abbreviated neuro exam. Brisk reflexes. She about came off the table with her patellar reflexes. Her simple lower back pain just went to cervical myelopathy. And because of that single finding, we surveyed her entire spine, rather than just the lumbar. And yes, there was a lesion there in the cervical spine.
So, yes, it’s true that technology has advanced our diagnostic accuracy tremendously. But you have to know where to look at how to apply it. The patient’s story won’t always tell you that. A simple physical finding will tell you how to apply said technology. In other words, when you have an MRI, EVERYTHING is a stroke to change the hammer-nail analogy. Unless you know how to apply it.