Digital data reveal new pandemic dynamics in 17th-century Venice

The EPFL researchers targeted an epidemic of the bubonic plague that brought death rates of nearly 35%. © Antonio Zanchi: The Virgin appearing to the Plague Victims

The EPFL researchers targeted an epidemic of the bubonic plague that brought death rates of nearly 35%. © Antonio Zanchi: The Virgin appearing to the Plague Victims

Researchers at EPFL have used digitized historical records to provide novel insights into the spread of the bubonic plague in Venice, Italy.

The COVID-19 pandemic has been characterized by a great deal of fear and uncertainty, as reliable data required to make key healthcare and policy decisions are often difficult and costly to obtain. Nevertheless, the data that are available have helped countries to understand and manage the spread of the novel coronavirus to a degree that would have been unimaginable to those who lived through the so-called “Second Pandemic” of the bubonic plague.

It was the dynamics of an epidemic of this “Second Pandemic” in 17th-century Venice that the EPFL researchers in the College of Humanities Digital Humanities Lab (DHLab), led by Frédéric Kaplan, and the School of Life Sciences Digital Epidemiology Lab, led by Marcel Salathé, chose to study. They based their research on newly collected and digitized daily death records, or necrologies, from the city’s Patriarchal Archives. They used data science techniques to analyze the spread of the bubonic plague, which is caused by the bacterium Yersinia pestis, in the Italian city between 1630 and 1631.

“It took more than five years to collect, annotate and model the data, but the results demonstrate that scientifically precious datasets can be extracted from century-old archival documents. These “big data of the past" can change our views on currently studied phenomena,” says Kaplan.

Notably, the team identified two distinct peaks of deaths, for which they propose explanations based on computational models of disease dynamics. Their work was published in October in the open-access Nature Research journal, Scientific Reports.

A peak and a tail

The researchers found that the Venice necrologies – which they note probably represent only a surviving portion of the total daily death records – showed more than 43,000 deaths, likely due in large part to the plague, within just two years; a figure consistent with the nearly 35% mortality rate at the time. They used data science techniques and computational models to simulate the disease’s rampant spread throughout the population.

Interestingly, the team observed that the deaths appeared to follow a novel pattern: a first peak in 1630 that reached over 400 deaths per day at its worst, followed by a less acute, but longer-lasting, peak in 1631. They note that this is the first description of such a “long tail of high mortality” in the literature on the subject.

The researchers hypothesize that this “two-stage” phenomenon may be due to some deaths from other causes having been mistakenly attributed to the plague, or to the occurrence of two distinct plague epidemics – for example, one bubonic and the other pneumonic. They also note that the observed pattern could stem from variations in Venice residents’ behavior in reaction to the outbreak, which could have affected the disease transmission rate.

This work, the investigators conclude, underscores the importance of digitized records for understanding the societal impacts of historical phenomena, and for digital data collection as a tool for studying global patterns of disease spread as well as local dynamics. 

“It’s deeply fascinating to look at epidemic curves of disease outbreaks hundreds of years ago,” says Salathé. “This work is a first step towards obtaining a more detailed epidemiological understanding of these outbreaks.”

References

Lazzari, G., Colavizza, G., Bortoluzzi, F. et al. A digital reconstruction of the 1630–1631 large plague outbreak in Venice. Sci Rep 10, 17849 (2020). https://doi.org/10.1038/s41598-020-74775-6
 

 

Author: Celia Luterbacher
Source: EPFL