The Laws of Medicine (Ted)

In this enjoyable book, Dr. Siddhartha Mukherjee establishes his three “laws of medicine.” Medicine is what Dr. Mukherjee calls an “uncertain science,” that is to say, it is not absolute. There are no “universal equations.” This is one profession that will not soon be “taken over by robots.” Medicine is a delicate profession, in that it can only be made “perfect” by the careful observations of imperfect doctors.

Law One A strong intuition is much more powerful than a weak test. Dr. Mukherjee reveals that test can only be effective when paired with knowledge of the patient. A test done without prior knowledge of a patient is extremely ineffective. Dr. Mukherjee illustrates this with an illustration that is pertinent today. In the example situation in which we have a test for Ebola with only a 5 percent false-positive (tested positive incorrectly) rate. Say we test every individual coming through an airport (0.1 percent of people actually have Ebola). Dr. Mukherjee reveals that in this situation 98 percent of tests would be false positive! I wonder what percentage of Covid-19 cases are false-positive.

Law Two “Normals” teach us rules; “outliers” teach us laws. Usually when a new medicine fails testing, most think of why it didn’t work. But Dr. Mukherjee encourages us to investigate why it did work in outlier cases. The importance of this idea is illustrated in history. When Tycho Brahe proposed a modified geocentric universe, he thought he had finally reached the conclusion. But the model could not explain the unusual movement of Mars. This problem was investigates by Johannes Kepler. The “outlier” would eventually give rise to a heliocentric model of the universe in which the planets move in ellipses.

Law Three For every perfect medical experiment, there is a perfect human bias. Many times, a perceived result may be purely psychological: [W]hen you enroll a patient in a study, you inevitably alter the nature of the patient’s psyche and, therefore, alter the study. The device used to measure the subject transforms the nature of the subject. Therefore, a good doctor must know any biases present in a study when prescribing a medicine to a patient. Any perceived uncontrolled variables must be investigated to determine if it will affect a certain patient. The control of biases is important in this “uncertain science.”

In this quick read, Dr. Mukherjee presents the many perplexing questions behind the “uncertain science” of medicine.