Poster containing information about influenza from the Treasury Department.

Revealing Data: The 1918 Flu Never Topped War

By Ashley Bowen ~

A photograph of a nurse wearing a surgical mask and a list of tips "to prevent influenza"
To Prevent Influenza!, 1918
National Library of Medicine #101580385

Late February and early March marks the 100th anniversary of the Spanish flu’s appearance in the United States. Although the 1918 flu killed more people worldwide in a year than World War I did in four years, the epidemic has been comparatively forgotten. Confronted with the concurrent anniversaries of the 1918 flu and World War I, I wanted to learn more about how physicians thought about the flu while it was spreading around the world. How did they allocate their attention between these two major health crises: the flu, killing millions of men, women, and children around the world and the war, killing millions of soldiers and civilians across Europe?

Poster containing information about influenza from the Treasury Department.
Influenza spread by droplets sprayed from nose and throat, 1918
National Library of Medicine #101453499

Using the archives of three journals digitized for the Medical Journals Backfile Digitization Project, a joint project of the National Library of Medicine and the Wellcome Trust, I investigated how extensively medical journals published about the war and the flu from 1916–1920.

The three journals in this research represent a spectrum of types of titles. The British Medical Journal (BMJ), a major medical journal with over 1,000 articles, letters, or other items published each year. The Edinburg Medical Journal (EMJ), a widely respected, but much smaller, medical journal that published only about 150 or so articles, letters, or other items each year. Finally, the Indian Medical Gazette (IMG), an English-language medical journal published in India, published about 250 or so articles, letters, reviews, notes, and more each year.

For this study, I collected some basic quantitative data about the contents of each journal and then compared the journals to each other. This was made possible because, as part of the Backfile Digitization Project, the NLM has made available to download complete runs of several journals as .txt files for researchers interested in text mining, the analysis of large datasets using automated tools. In short, a way to process the thousands of articles these three journals published without having to read each individual article.

Unsurprisingly, all three publications contained far more articles about the war than about influenza, consistent with what historians have argued for a long time.

A graph plotting mentions of war and flu in three medical journals.
From 1916 to 1920 mentions of war in three prominent medical journals far exceeded mentions of influenza.

Two things did surprise me, however. First, the Indian Medical Gazette published a higher percentage of articles mentioning influenza in 1919 than the other titles. In retrospect, I probably should not have been surprised about the amount of space the IMG gave over to the flu. Modern scholars have described India as the “focal point” of the pandemic. Between 10–20 million people in India died of influenza during the 1918–1919 outbreak.

The second surprise was learning that articles about the flu peaked in all three journals in 1919 while the epidemic was still raging, though with less intensity, around the world. Knowing that the epidemic continued through at least the middle of 1919, I expected that most articles about the disease would appear in 1920, well after the worst of the epidemic and with enough time to account for the pace of scholarly publishing and physicians to write their findings.

This data doesn’t explain why the number of pieces about the flu spiked in 1919 but it does suggest that medical interest in the flu did not extend much beyond the worst of the epidemic. Without additional information and a bit more context, it is difficult to know more. For example, the data only reveal the final editorial decision—about what to publish and when—but cannot reveal any of the behind-the-scenes information we need to determine how representative these published articles were compared to all the manuscripts submitted. Next steps for this research might include splitting the articles by month, rather than year, to identify trends in publishing relative to the flu’s mortality. Alternatively, a researcher might start to critically read articles published in the IMG in 1919, like one editorial titled “The History of Influenza,” to begin to evaluate medical thinking about the epidemic while it still circulated in India.

Black and white photograph of a group of men and women.
Great Britain, British Red Cross Society Voluntary Aid Detachment Hospital, Newton-Abbot, England: Nurses and influenza patients, ca. 1918
National Library of Medicine #101405268

The trench warfare that defined World War I, identification of shell shock, and rapid advances in military medicine all make for a more heroic narrative than influenza, a deadly disease about which doctors could do little in 1918. The flu produced no heroes (though many doctors, nurses, and volunteers certainly provided heroic service to the sick and dying) so it is perhaps unsurprising that medical men did not want to dwell on the disease. It may have defined their practice at home in 1918–1919 but it offered up few opportunities to advance medical knowledge and, by extension, to publish novel findings.

A few quick notes about this work. I only looked at the words “influenza” and “war.” I am certain that I missed articles that used “grippe,” “flu” (rather than influenza), or simply referred to the war by mentioning that someone sustained an injury “in France.” I also didn’t filter out any articles that aren’t explicitly about influenza or war injuries/diseases (for example, in a few places a disease’s death toll is compared to the influenza even if the article isn’t about that). Despite these limitations, this data can add nuance to arguments about how quickly the 1918 influenza became a “forgotten pandemic.”

Complete runs of over 40 historic medical journals are available to download for free from PMC’s text mining collections FTP site. The files open as a series of .txt files ready for analysis. If you make use of these text mining collections, we would like to hear about your project and findings!

We hear about data every day. In historical medical collections, data abounds, both quantitative and qualitative. In its format, scope, and biases, data inherently contains more information than its face value. This series, Revealing Data, explores how, by preserving the research data of the past and making it publicly available, the National Library of Medicine (NLM) helps to ensure that generations of researchers can reexamine it, reveal new stories, and make new discoveries. As the NLM becomes the new home of data science at the National Institutes of Health (NIH), Circulating Now explores what researchers from a variety of disciplines are learning from centuries of preserved data, and how their work can help us think about the future preservation and uses of the data we collect today.

Through 2018, Circulating Now will periodically publish posts featuring NLM collections that illuminate the medical history of The Great War, which lasted from August 1914 to November 1918.

Ashley Bowen, PhD is a contract guest curator for the Exhibition Program in the History of Medicine Division of the National Library of Medicine.


  1. Unfortunately the Edinburgh Medical Journal files are not (yet?) available on the regular Pub-Med Central site. Is digitisation still underway and if so when will it all be available? The EMJ is a wonderful historic resource and it would be great to have access to it!

    1. Yes, unfortunately it’s only currently available in the .txt mining collections (so you’ll need to use another program to search it. I used AntConc). My sense is the title will be on PMC in a few months.

  2. I am UK based amateur researcher. Can you please advise about copyright for images-thanks.Usages for presentations and publications-non commercial.

    1. Hello James, Thanks for reading Circulating Now.
      The Library generally does not own rights in its collections and cannot grant or deny permission to publish or otherwise distribute material in its collections. Patrons must make their own rights assessment in light of the intended use.
      “Usages for presentations and publications-non commercial” may fall under the fair use doctrine.
      For more information about how to make rights determinations consult our Patron Guide to Copyright and Historical Materials.
      If you have a question about the copyright status of a particular item in NLM collections you can contact our reference services at NLM Customer Support.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.