I was reading a well-written piece that revisited the issue of discharge notes as part of the transition of a patient from one type of care to another. When you watch a medical TV show, just as when you watch a TV police show, it seems that an entire complicated case can be summed up, managed and fully reported within 43 minutes. In the medical situation, it is equivalent to a patient with a three-year history of undiagnosed headaches who ends up having every test in the book and is then seen and diagnosed by the world expert in headaches, all within a few hours. And of course, the last few minutes of the show is when someone says to someone else that he loves her, or “I am your father” or the like.

In practice, especially with more complicated cases, care, diagnosis and treatment may extend over many years. In such a situation, how does one physician summarize the case such that the next physician  can effectively pick up where the previous doctor left off. This transfer of care may occur at discharge from the hospital, after seeing a specialist and even during shift change, when a new young (i.e. inexperienced) physician comes on, replacing the one who just completed his 36 hour shift. There is a great deal of literature about the potential for error at times of transfer of care. Sometimes, the most basic information such as an allergy to a medication or a critical treatment can miss being passed over to the next team. And sometimes patients die because of this.

Electronic medical records were supposed to fix this. With one click access to a patient’s entire history of medical care, a physician should be able to review an entire case within seconds. So the patient with three years of headache should be able to move seamlessly from doctor to doctor without ever having to repeat a test and without ever suffering from the lack of transmission of a key piece of medical information. Many physicians complain that EMRs not only have failed to help summarize and move care along, but in fact, EMRs have worsened the situation by becoming mindless collections of data that are almost impenetrable due to their endlessness.

I am a very vocal supporter of electronic medical records and it is legitimate that I should be able to answer this apparent conundrum. Quite simply, computer databases are nothing more than the virtual equivalent of a desk drawer. If all you do is stuff in more and more junk, you will most likely only be able to retrieve lots and lots of junk. Admittedly, once computerized, it is possible to search thousands of pages of medical documentation, easily and quickly, for a given name of a specialist. So, I could review the three or four summary letters that a particular neurologist wrote about a patient, despite these summaries being lost within multiple charts scattered across multiple visits and, very likely, multiple healthcare facilities. On the other hand, having a record of the hundreds to thousands of lab tests that a particular patient has undergone, could literally make it impossible for me to discern if this patient is presently stable versus losing blood due to an, as yet, undiagnosed cancer.

How can things be made better? I would argue that one needs to start from the beginning, rewrite everything, rethink everything and then retrain everyone. This actually isn’t as Sisyphean as it sounds. What one needs to do is something which I have described multiple times in the past – redefine the role of the treating physician.

During my medical school training, I had multiple classes about the basic biochemistry of how a cell processes energy. This is unquestionably an important thing to understand as a foundation for understanding various types of diseases  and the actions of certain medications. On the other hand, as important as this is, it is no more important for sure than learning how to review a mass of information and summarize it in, perhaps, just a half a page.

I can imagine a course in medical school where you are given a thousand page chart each week and it is your responsibility to review that chart and generate a summary. In the first week, I would not be surprised if a number of the medical students failed to complete the task. But over the course of the two or three months of training, the medical students would become adept at this practice. A couple of years later, when at two in the morning a 75-year-old patient presents with chest pain and his thousand page chart, that medical students/intern/young resident should be able to extract the key points and summarize them within a reasonable amount of time. With that half page summary, the upcoming management of that patient will unquestionably be streamlined. More so, the next intern/young resident who receives this course will already have an excellent half page summary waiting from the previous admission. Over the course of a few years, many patients will begin to benefit from this rewriting of their charts.

One can definitely ask whether this is time well spent by a physician. Rather than playing the role of an editor of 1000 page cases, one could easily argue that a physician should be focusing on the healthcare of the patient rather than the typing of a numbered list of every procedure the patient has had. If the intent is to move this responsibility away from the physician, someone or something else has to take it on.

One could easily imagine a new subspecialty in healthcare that focuses entirely on the rewriting and effective summarization of medical chart information. When a patient comes to the hospital, an individual who is trained in medical chart summarization would spend 15 minutes reviewing the given case and would generate the half page summary I noted above.  The receiving physician would look at that page and move forward.

Alternatively, as computers become smarter and smarter, they should be able to create such a summary at any time (and much more quickly than within 15 minutes). With advanced language processing, computers will eventually be able to edit such masses of medical information. Especially, if the majority of the information is already computerized [meaning, there is no need to scan and OCR hundreds of pages of handwritten documentation], you can easily imagine a computer being able to display, in one graph, megabytes of various types of information. The advantage of computers, of course, is that they can work 24/7 and for no additional fee, once the software has been created.

Is a doctor still a doctor if he or she is not the one reviewing the original chart? I purposely am avoiding any issue  of medicolegal responsibility because that requires a whole different rationale. Purely in relation to quality of care, does a patient benefit from the fact that the doctor has just spent an hour reviewing an angiography report from three years ago, when the same patient had a repeat angiography one year ago? In this day and age, when more and more patients have more and more tests and are living longer and longer, a solution for managing endless documentation is a must. EMRs are the solution, but only if they are designed to perform summarizations, and if the physicians are effectively freed of this task.

I have also mentioned in the past that the physician role should change to one of overseer of the work of other healthcare professionals. This new role is not related to the IQ of the physician relative to others. It would be a task that the physician is specifically trained for.

In training a physician to be such an overseer, the physician would lose proficiency in certain other tasks. If I have three uniquely trained healthcare providers in three different clinics able to stitch a patient appropriately, isn’t it more appropriate for me to oversee the care and guide the appropriate type of closure rather than be the one who physically does it? If I have a specialized technician who knows how to apply a cast, isn’t it much more reasonable that I be the one to decide what cast to apply, rather than have to be physically present in each of the three clinics to apply the cast? At some point in the future, robots will be placing casts on people. At that point, it really will become tremendously inefficient for a physician to perform this task.

Everyone has to swallow hard and retrain to some extent. The professors, who presently profess in medical school, will have to abandon their perspectives on the practice of medicine and accept an entirely new approach. In time, I believe, that young physicians who grew up on Twitter, will demand systems that can summarize complicated situations in 140 characters. It might very well be that we need to wait for this entire new generation of doctors, in order to get the necessary support from the medical community, in order to transition the entire field of medicine.

Thanks for listening