Muench’s article describes a fascinating episode in the history of epidemiology, in which crucial progress was made in identifying the causes of the ‘Blue Death’ (cholera) in 19th century England. The progress was from employing humoral and miasmic models of relevant epidemiology to a more modern and scientifically sound model. This paper will briefly summarize the first part of the article, and then answer some crucial questions about it.
The article centers on John Snow, a mid-19th century physician also trained as what was then known as a surgeon-apothecary. Snow studied Asiatic cholera, which began to show up in England in 1831. He was trained in the humoral model of disease, which maintained that general health required—and perhaps consisted in—balancing four ‘humors’: blood, phlegm, black bile, and yellow bile (Muench, paragraph 2). This model was not ideally successful in threating cholera, of course. Also prevalent at the time was the miasmic model, according to which diseases such as cholera are found in polluted air. A crucial step in Snow’s medical education or progress was taken when he began to study coal mine workers, and later anesthesia. While we now know that cholera was contracted from highly polluted drinking water, at the time, to reiterate, it was thought that the disease was airborne. Snow helped to show that this was not the case, partly through his experience with the study of anesthesiology. Basically what he learned in studying anesthesiology cast doubt, for him, on the miasmic model (Muench, paragraphs 7-10).
Models are essentially theories, and theories allow scientists (including physicians) to test hypothesis by determining whether the predictions—or statistical correlations—made by the theory or model are correct. The humoral model of disease proposed that cholera was caused by an imbalance in the four ‘humors’. So it predicted that modifying a patient’s situation with respect to one or more of the humors would cure cholera. The miasmic model, by contrast, held that cholera was caused by airborne pollutants of some sort. It therefore predicted that reducing or eliminating exposure to relevantly polluted air would cure the disease.
The general skills necessary for implementing the sort of experimental design that would lead to experiments that improved the measuring and delivery of anesthetic are complex. But perhaps most important was skill in ensuring that only the crucial variable was being measured in a given experiment. In other words, the crucial skill was being able to control for extraneous factors in an experiment. If one does not do this, then the results of an experiment will not verify (or indeed falsify) the hypothesis, model, or theory in question. For the possibility will always be left open that the result of the experiment is due to extraneous factors, rather than the crucial factor involved in the theory or model. So for example, if a test of an anesthetic drug did not control for body weight, then it might be thought that a given amount of gas is sufficient, for people in general, to be able to avoid the pain of surgery—when in fact the experiment showed only that a person of such-and-such body weight would be sufficiently anesthetized by the amount of gas in question.
Experiments are considered strong tests of hypotheses because they are (or should be) conducted in controlled environments. Merely anecdotal evidence, by contrast, again leaves open the possibility that extraneous factors led to the result in question. So it does not provide a strong test. With an experiment one can isolate the crucial factor, associated with the hypothesis or theory, to determine whether it and it alone is responsible for the experimental result obtained.
- Muench, Susan Bandoni. “The mystery of the blue death: A case study in epidemiology and the history of science.” Journal of College Science Teaching 39.1 (2009): 60-66.