Greetings, gentle reader. Dr. Cranky has returned once again, filled to the brim with black bile, melancholia and lugubriousness as he sits in his padded cell experiencing a moment of transient lucidity. And what is the source of your correspondent’s ill humor? Quite simply, it is nothing less than his recent recollection of the notorious Dr. G. Memories of this truculent tyrant are second only to those of Gilda, the Wicked B*tch of the East in their ability to induce such an imbalance in your host’s mental stability. But all is not lost. In fact, such a recounting of the ghosts of internship past might be beneficial to you as an instructional aid.
“How could this be,” you might wonder. “What good can possibly come from torturing yourself in this way? Please inform us. We’re simply dying to know.” Well, my dear disciple, the one thing Dr. G loved to do most, other than chastise Dr. Cranky and his compatriots verbally, was to tell us how worthless we were on paper. You see, at the end of every clinical rotation he had the perverse pleasure of filling out our evaluations. Thoughts of reading these assessments resulted in the greatest trepidation imaginable to our young, impressionable minds. We even had a name for it, and one was said to have been “G-bombed” by the experience.
So what did this malicious martinet have to say about your earnest host? What was the worst thing he could think of? From the darkest corner of his blackened heart, the best invective Dr. G could come up with, after questioning your scribe’s basic hygiene and parentage, was to say that “Dr. Cranky is an algorithmic doctor.”
“This is most curious,” you might say. “Dr. Cranky, once again you intrigue us with these strange concepts. What is this mysterious thing you call an algorithm, and just what did Dr. G mean when he said you were algorithmic?” Dr. Cranky is glad you asked. Allow him to explain.
Quite simply, an algorithm is nothing more than a flowchart which contains of a series of key questions. At each decision point the person who uses it asks whether the interrogative can be answered in the affirmative or negative. If the answer is “yes,” the algorithm continues down one path. If the answer is “no,” things progress via a different route. The idea is to arrive at the correct answer to a problem using a set of precisely defined rules. Computer scientists use such instruction sets as part of their attempt to mimic the human decision-making process.
For example, let us say you have an itch. What do you do next? A simple algorithm to solve this problem might be as follows:
As anyone who has ever had an itch can tell you, once you scratch it and make it go away heaven and earth are yours for the taking. However, if you can’t reach the pruritic nerve endings in question you are in for an evening of torment unparalleled in the annals of human suffering.
It would seem that your fate lies between rapture and damnation, depending on whether or not your arms are long or limber enough to reach those affected nether regions. Oh, cruel fate, thy name is allergy! But there is hope. What if you had a friend nearby? Could they help you out with this perilous predicament? If you look at the flowchart above it would seem not. However, that is one of the beautiful things about an algorithm; it can always be modified to suit the situation. Consider such an alteration below:
A prime example of how such decision trees can be used in Medicine is found in the American Heart Association’s ACLS protocols. The letters A-C-L-S are part of an acronym which stands for Advanced Cardiac Life Support. These algorithms are an attempt to guide a clinician toward the proper care of a patient in actual or potential cardiac arrest. They were first published in 1974 and, in theory at least, are a distillation of what science says is the best way to bring someone back from the dead. They can be quite elaborate. For example, what follows is just one of the many ACLS algorithms an emergency physician is expected to keep floating around in his head on a day-to-day basis:
The primary advantage of these flowcharts is they allow a physician to keep a lot of information in his head so he can apply it under extremely stressful situations, when there isn’t enough time to think things through. A tremendous disadvantage, however, is that algorithms can also allow a physician to project an appearance of competence when he really doesn’t know what he’s doing. And this is what Dr. G was saying, albeit in an indirect fashion.
Was this assessment of your humble servant justified? Dr. Cranky thinks not. Medicine is far too complicated to learn in just a few years. All the knowledge accumulated in medical school, even after supplemented by an additional four years of postgraduate training, is just not enough. Disease exists in infinite variations and textbook descriptions are far from the norm. True understanding is a life-long process and takes decades of dedicated work. That is why we call it the practice of medicine.
In real-life, medical students and residents learn the healing arts by starting with simple algorithms such as the itch example your host provided above. This is borne of necessity, since young doctors-in-training are expected to be functional from the moment they set foot on the wards. You have to start somewhere, and genuine understanding at this point is impossible. With each new success and failure budding Asclepians build on the foundation such decision trees provide and refine them. A good medical educator, such as the Big C, shepherds his young charges, encourages the expansion of their knowledge and teaches them why each decision point is important.
This is not to say, surprisingly enough, that Dr. G didn’t have a good point; he just directed it toward the wrong people. You can’t expect a young intern, fresh out of medical school, to understand all the intricacies of something as complex as renal pathophysiology. However, your earnest author is in concordance with his former tormentor when it comes to older doctors who practice via algorithms. Dr. Cranky can’t stress enough that an such decision sets should be used as a starting point only. Every day your favorite mentor encounters colleagues in the great house of Medicine who do things in a cookbook fashion, without any real understanding. These people have perverted a wonderful tool into a crutch for the indolent.
To be a true physician you must know more than just what to do in any given situation. It is critical that you look deep, examine the underlying mechanisms and understand why!