Hospitals, Office Visits of Great Use In the Future, Doctor

January 27, 2014

As big a proponent of technology and data as I am, the truth is data has never cured anything and tech can't replace in-person interactions with physicians.

The headline of a recent Medscape article reads "Topol: Hospitals, Office Visits of Little Use in the Future." Eric J. Topol, MD, has written a book that I have not read called "The Creative Destruction of Medicine" and blogs under the same moniker. He clearly sees himself as a medical futurist, frequently making interesting and challenging assertions about where medicine is going and where it should go. Technology looms large in his view of the future.

As you may have gathered, I am no stranger to technology but I am probably less impressed with it than many. This is primarily because, when it comes to technology, we collectively seem to be spending more time rearranging the deck chairs on the Titanic than on understanding and using technology to its full potential. Had the captain of the Titanic been less confident about their "sink-proof" technology, he might have slowed down. Had he known what we know now about the atmospheric science that explains why icebergs become invisible when there is an Arctic inversion layer, he might have slowed down. Had he realized that the rudders have no effect if you throw the engines in reverse and cause cavitation and turbulence in their vicinity making it impossible to steer and that inertia makes it impossible to turn a large vessel on a dime (see Exxon Valdez), he might have slowed down. It was incomplete understanding of science and a misplaced trust in technology that sank the Titanic.

Now Topol asks us "Why do we need hospitals in the future? Why do we need to have office visits with physicians in the future? "He says we have the technology we need to do most everything the patient needs by remote monitoring. He also thinks that video chat is equivalent to a face-to-face meeting and I guess he also believes that the high-tech equipment and high bandwidth communication channels needed to support it are available where they are needed most, like poor neighborhoods and rural areas where the population is so sparse that it is not economical to run fiber or high-speed cable for miles just to reach one home.

This vision of the future is based on one underlying assumption: The data tell the story, that is, make the diagnosis. I don't agree. I was taught and it has been my experience that the history makes the diagnosis most of the time but that getting the history is difficult and not everyone is equally good at it. I have a whole file of interesting cases that I have resolved as the "expert consultant" simply because I sat down with the patient, gained their trust, and asked one or two simple, obvious questions that no one had asked before. Like when I asked a patient who had supposedly made a suicide attempt, "Why did you take those 14 pills all at once?" The answer: "The doctor told me to take all the medicine. I had a visit scheduled the next day and I had these 14 pills left, so I did what the doctor said and took them." Neither data nor remote monitoring would have helped. I could not have gotten the critical history using video chat. The doctor that failed to ensure that she understood how to use the meds correctly in-person, is unlikely to have done a better job using video chat. It's true, the patient was forced to travel an hour to reach our clinic and waited 45 minutes to see me. Topol would probably label that a waste of time, but as a result of that 105 minutes, she was spared weeks of unnecessary psychiatric care including psychotropic meds and of being permanently and incorrectly labeled as a suicide attempt.

When neonatology was new and the hospital got its first monitors suitable for use on small babies, our lead physician was totally captivated by the new equipment and the data it would produce. There were times when he would sit there glued to the monitor, watching, as a baby's vital signs became unstable. The baby needed attention urgently but the doctor was fixated on the data. 

Data never cured anything. In some respects, I wish that Topol was a pediatrician. Pediatricians know that the vital signs alone (temperature, respirations, heart rate, etc), the things you could get by remote monitoring, are poor indicators of whether an illness is serious or not. The same can be said for lab tests. Some can be disturbing, but none "prove" that the child is sick enough to require treatment. You need to observe the child closely, from multiple angles, for a prolonged period of time in order to get a clue as whether what you see is a significant deviation from their normal state. Some of your appreciation of their normal state comes from having spent time with them when they were normal, at those unnecessary office visits.

Cut out the face-to-face contact and the well-visits and you will drive up the number of patients who are incorrectly assumed to be seriously ill by doctors who are seeing them, in the ER, for the first time. The assumption that data, especially the kind of data that has been bent, folded, spindled and mutilated to get it entered into an EHR, is going to eliminate the need for patients to see their doctor represents the same sort of blind trust in technology that was the undoing of the Titanic.