Tracking Spanish Flu through Austro–Hungarian Press

The area of the red circles on this map is proportional to the smoothed relative number of mentions of influenza in newspapers throughout the Austro–Hungarian lands in the last four months of 1918. In the same vein as Google was detecting influenza epidemics using search engine query data, perhaps these numbers approximate the local timeseries of influenza morbidity.

Austria-Hungary

The 1918–1919 pandemic, called Spanish flu, killed at least five times more people than World War I. Here is a chart of its mortality in England and Wales (source: Edwin O. Jordan, Epidemic Influenza: A Survey, 1927):

weekly-mortality

The three waves occurred at similar times worldwide. I concentrated on the second, most deadly wave in the Austro–Hungarian empire and in the states that emerged in its lands. The full-text search of ANNO (AustriaN Newspapers Online, a digitisation initiative of the Austrian National Library) allowed me to count all the occurrences of the words {Grippe, Influenza}, {chřipka, chřipky, chřipce, chřipku, chřipkou} (in the Czech newspaper Dělnické listy), and {grypa, grypą, grypę, grypie, grypy, influenza, influenzy} (in the Polish newspaper Kurjer Lwowski) by the day of their publication in 27 newspapers from 14 cities and towns. A little side discovery is that the name “Spanish flu” in all three languages first appeared in print at the beginning of July 1918.

Despite the inevitable inaccuracy of this method, among others due to OCR errors and the shadowing of the epidemic by other topics like the surrender of Austria–Hungary, the end of the monarchy, and the independence of national states, noise tends to cancel out. I inspected all 29 mentions of flu in Kurjer Lwowski in the week of Oct 7–13, 1918. Out of them:

  • 19 concerned local events,
  • 4 concerned events elsewhere in Austria,
  • 5 concerned events elsewhere in Europe.

I fitted the Gompertz curve y = c \exp(-e^{-b(t-t_0)}) to the cumulative numbers, keeping in mind the fact that the issues of some newspapers from the end of the year are missing from ANNO (recall the independence). For instance, here is the best fit for the cumulative number of times flu was mentioned in newspapers published in Vienna:

Vienna1

and a full-year chart of mentions of flu per week in the same newspapers:

Vienna2

I tried also to verify Andrew Price–Smith’s claim that Spanish flu tipped the balance of power towards the Entente in the last days of World War I. Unfortunately, many belligerent states do not have online archives of newspapers from 1918, the German archives are unsearchable, and Gallica and the British Newspaper Archive are useless: the mentions of grippe or influenza in French and British press are few and far between outside advertisements, a sign of wartime censorship. However, this table from Epidemic Influenza: A Survey showing that the highest mortality co-occurred in France and Germany makes me doubt Price–Smith’s claim:

highest-weekly-mortality

The map of Austria–Hungary above comes from MPIDR [Max Planck Institute for Demographic Research] and CGG [Chair for Geodesy and Geoinformatics, University of Rostock] 2012: MPIDR Population History GIS Collection (slightly modified version of a GIS-File by Rumpler and Seger 2010) – Rostock (Rumpler, H. and Seger, M. 2010: Die Habsburgermonarchie 1848–1918. Band IX: Soziale Strukturen. 2. Teilband: Die Gesellschaft der Habsburgermonarchie im Kartenbild. Verwaltungs-, Sozial- und Infrastrukturen. Nach dem Zensus von 1910, Wien).

Advertisements
Tracking Spanish Flu through Austro–Hungarian Press

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s