What do you think about when someone mentions Benjamin Franklin? Do you think of the statesman, the inventor, the man with the kite in the thunderstorm, or the first Postmaster General? Among his many activities and accomplishments, Benjamin Franklin also managed to include a little bit of epidemiology when he wrote the introduction to a pamphlet about variolation in 1759. Epidemiology is the study of “that which comes upon the people.” Two forms of epidemiology are descriptive epidemiology and analytical epidemiology. Analytical epidemiology is done through the use of statistics to research diseases and interventions based on the observations done through descriptive epidemiology. Franklin performed descriptive epidemiology in showing the number of cases of smallpox, the number of deaths attributable to smallpox, and similar descriptive numbers of people who received variolation in colonial Boston.
At the time that the pamphlet was written, the only protection against smallpox was variolation, the practice of purposely infecting a person with the smallpox virus under controlled conditions in order to confer immunity via a milder form of the disease. Jenner’s breakthrough use of the cowpox virus did not come along until 1796 and would not be widely used for a few years after that. While it had been used for many centuries in places like China and Asia Minor, variolation did not arrive into Europe until the early 1700s and to the Americas in 1721 with Rev. Cotton Mather and Dr. Zabdiel Boylston.
Franklin had a personal stake in controlling smallpox. His son, Francis Folger Franklin, died of smallpox at the age of four in 1736. Franklin noticed that variolation caused fewer deaths than naturally contracted smallpox did. To encourage the use of variolation in the American colonies, Franklin wrote the introduction to a pamphlet written by English physician William Heberden in 1759. The pamphlet encouraged the use of a controlled infection with smallpox in order to “save thousands” from smallpox. In that pamphlet, Franklin did a bit of descriptive epidemiology: he placed a table in that pamphlet describing the observation that fewer people died from variolation than from acquiring smallpox “the common way.” Here’s what he reported about a Boston epidemic of 1753-54:
In Whites, he observed that 452 (8.9%) of 5,059 died after acquiring smallpox through natural means. In Blacks, 62 (12.8%) of 485 died the same way. Overall, 9.3% of the people observed to have acquired smallpox by natural means died. Franklin then observed that 23 (1.2%) of 1,974 and 7 (5.0%) of 139 Blacks died after variolation. Overall, 1.4% of those who received smallpox through variolation died.
It is easy to see that 1.4% is better than 9.3% in terms of death from smallpox acquired in a controlled way versus smallpox acquired naturally, but is this just a chance observation, something that randomly happened? Analytical epidemiology of these numbers gives us the answer. Using biostatistics, we find that people who obtained smallpox “in the common way” had a 6.5-fold higher risk (also known as a relative risk) of dying from smallpox than those who “received the distemper by inoculation.” The same biostatistics tell us that, if we were to observe this situation 100 times, the risk of dying from naturally acquired smallpox would be between 4.5 and 9.4 times higher than dying through variolation 95 out of those 100 times. This range is known as a 95% confidence interval, another statistical value. The relative risk is the ratio of the probability of death (1.4%) in the people who received variolation to the probability of death (9.3%) in the people who did not receive variolation. A relative risk of 1.0 means that the risk of dying is the same in the two groups. The fact that the 95% confidence interval does not include 1.0 means that these findings were not due to chance: the differences were the result of a real influence from variolation on death rates. There was also no significant difference between Whites and Blacks, meaning that variolation could be used for both groups.
In his effort to show evidence that variolation was better than acquiring smallpox “the common way,” Franklin performed a bit of observational epidemiology that was backed then, as it is now, by biostatistics. He did this almost 100 years before John Snow, considered the father of epidemiology, used observational epidemiology and a map to control the 1854 cholera outbreak in London. Today, descriptive epidemiology is performed by epidemiologists using large and complex disease surveillance systems. Their findings are then put to the test through statistical software and, when needed, analytical studies in the field. All of these findings, descriptive and analytical, are then used to control and prevent diseases.