Exploring Chicago’s Spanish Flu of 1918

Circulating Now welcomes guest blogger E. Thomas Ewing, Professor of History and Associate Dean in the College of Liberal Arts and Human Sciences at Virginia Polytechnic Institute and State University. Professor Ewing’s recent research in the historical collections of the National Library of Medicine includes new methods of analyzing textual information. Professor Ewing is the Principal Investigator and Project Director of “An Epidemiology of Information: New Methods for Interpreting Disease and Data,” funded by the National Endowment for the Humanities through the Digging into Data Challenge. NLM was pleased to be a co-sponsor of the recent Epidemiology of Information research symposium, in cooperation with the Office of Digital Humanities of the National Endowment for the Humanities, the National Science Foundation, and Virginia Polytechnic Institute and State University. 

“Methods and Mediums” for Understanding Public Health Responses to a Pandemic

On September 23, 1918, Dr. John Dill Robertson, the Health Commissioner for the city of Chicago, issued the following pronouncement:

This brave proclamation vastly underestimated the severity of the epidemic that was just beginning to sweep the entire nation, including cities like Chicago. By October 1, the number of deaths per day would exceed 400, compared to a daily rate of less than 50 when Dr. Robertson made his bold claim. Two weeks later, at the peak of the epidemic, the number of reported deaths from influenza and pneumonia exceeded 2000 per day. The death rate gradually declined, and by early November, the number of deaths reported each day had returned to approximately 300, still higher than when the epidemic began, but far below the peak levels in the middle of October (John Dill Robertson and Gottfried Koehler, “Preliminary Report on the Influenza Epidemic in Chicago,” The American Journal of Public Health, Vol. 8, No. 11 (November 1918), pp. 849-856.)

Despite this seeming contradiction between his initial optimism and the actual impact of the Spanish influenza epidemic, Dr. Robertson remained convinced that his approach to the disease had been correct.

In his retrospective report on the influenza epidemic, Dr. Robertson devoted particular attention to the ways that his office “conducted what may be termed an intensive campaign of publicity and education” before and during the epidemic influenza. In this report, Dr. Robertson identified a range of “methods and mediums” used to ensure that “knowledge and light are agencies of wonderful might and influence” in the “never ending fight against preventable disease.” Although not everyone will “heed the warnings and advice of health officials,” this report asserted, “it still must be remembered that unless the light be spread, darkness will prevail.” (John Dill Robertson, A Report on An Epidemic of Influenza in the City of Chicago in the Fall of 1918 Educational Series No. 15 Department of Health City of Chicago, 1918, pp. 50-52.)

Dr. Robertson’s response to the Influenza Epidemic of 1918-1919 provides a case study for exploring the relationship between the statements of public health officials and the ways that newspapers shaped coverage of the influenza within particular communities.
This research is part of a larger project, entitled the Epidemiology of Information, which seeks to understand how newspapers shaped public opinion during the 1918 influenza pandemic using collections from around the country, including those held by the National Library of Medicine, Library of Congress, and the online Influenza Encyclopedia.

In his report, Robertson described how his office provided “daily information as to the status of the outbreak, together with advice and warning as to the things that should be done to curb the spread of the disease and to protect the public health.” Newspapers also printed daily warnings, often in the form of “Do’s” and “Don’ts,” which were distributed by the Department of Health.

The report cited photographs of street cleaners, nurses, and even the Health Commissioner wearing protective masks as specific examples of how newspapers served this goal of disseminating health warnings. The Health Department also provided articles, at least once a week, to 136 daily, weekly, and monthly publications in the city, including those published in foreign languages, “with the purpose of securing city wide co-operation in enforcing the protective measures adopted by the Department of Health.” In early October, when headlines began “to take an alarming aspect,” with predictions of very high number of cases and deaths based on reports from eastern cities and from military camps, the Health Department took the unusual action of publishing a three-quarter page ad, which featured the U.S. Public Health Service guidelines about “Spanish Flu,” with the intention of preventing “undue alarm” among the people (“Spanish Influenza. ‘The Flu’,” Chicago Daily Tribune, October 3, 1918, p. 24). These examples confirmed Dr. Robertson’s belief that his office took all necessary steps not only to control the disease, but to shape public perceptions of the disease in ways that prevented this perceived threat of “undue alarm.”

To explore the nature of Dr. Robertson’s statements about influenza as the disease approached Chicago, reached peak levels, and then gradually diminished, I have extracted direct quotations from Dr. Robertson that were published in one newspaper, the Chicago Daily Tribune, between September 19 and October 26, 1918. In total, this newspaper published direct quotes from Dr. Robertson in fifteen articles, with just the quoted statements adding up to more than 1000 words.

One way to look at this large amount of data is to visualize it in a tag cloud which highlights words that appear most frequently and therefore provide insight into the kind of language used by Dr. Robertson. Words such as influenza, disease, spread, epidemic, and symptoms suggest the centrality of medical terminology in these quotes, while other terms, such as theaters, close, schools, sweeping or street refer to potential public health measures to safeguard public spaces.

Yet this illustration also provides a misleading impression of the overall tone and purposes of Dr. Robertson’s comments. Many of the terms that appear in this tag cloud that suggest a level of caution are in fact negated in his actual statements. While terms like warning, deaths, careful, prevent, and dangerous, appear prominently in the tag cloud, a closer look at the actual texts indicate that, in most cases, Dr. Robertson was actually using these words in more reassuring and even encouraging ways. For example:

On September 21, Dr. Robertson’s statement about the potential spread of influenza from army camps to the city concluded with this reassuring statement:

On October 3, on the same day that Dr. Robertson declared that “You can keep well if you do your part,” a statement from the Health Commissioner included this bolded, and very encouraging, exhortation:

Finally, on October 7, one full week before the death rates reached its highest peak, Dr. Robertson was quoted as saying:

As this limited analysis suggests, developing new methods for analyzing newspaper reports about the spread of the Spanish flu has great potential to address a key historical question: How do public officials respond to an unprecedented outbreak of contagious diseases in particular contexts? For scientists in the field of epidemiology, these responses can be evaluated in terms of the relative success in achieving the goals of containing the spread of disease, ensuring treatment for infected patients, providing preventive measures where possible, and managing public perceptions in ways that maintain an appropriate balance between the extremes of provoking exaggerated panic or fostering a false sense of security. Historians exploring these questions will focus on the tension between the actions and intentions of public health officials and the perceptions and responses of the population, with the goal of understanding how these tensions shaped attitudes and behaviors among a range of participants.

This study of the reported statements of Dr. Robertson can address questions relevant to understanding both the historical context of and public health responses to disease outbreaks. Further research is needed to match up the tone of these statements with the actual course of the disease in Chicago and throughout Illinois. By accumulating larger quantities of text, other forms of data mining, such as topic modeling, network analysis, and tone classification, become more useful for both historical and epidemiological analysis. These methods must always be accompanied by a variety of methods that can explore the range of media through which public health officials communicated with the public regarding disease outbreaks. Through this kind of integrated and innovative analysis, history becomes more useful to researchers and to the public concerned about disease outbreaks and public health responses.

Dr. Robertson’s efforts to control perceptions of the 1918 influenza remain relevant to the present day, as public health officials use new forms of social media to provide information about disease outbreaks. The Department of Health of Human Services, through the Centers for Disease Control and the National Institutes of Health, provides lists of “flu basics” for the general public, a map tracking flu activity, and updates disseminated through smart phone apps, twitter, facebook, and other forms of social media. Public health agencies now face the same challenge encountered by Dr. Robertson almost a century ago: meeting the public’s right to accurate and timely health information at a time when reports of unprecedented numbers of cases and deaths threatened to spread panic and undermine public health measures. Dr. Robertson was too optimistic in his early predictions of the impact of the disease, and we can learn from that. His efforts to adjust his message to fit the changing circumstances of the disease provides a historical perspective on contemporary efforts by public health agencies to anticipate the spread of diseases and gauge the need for appropriate interventions.

Tom Ewing is an Associate Dean for Graduate Studies, Research, and Diversity in the College of Liberal Arts and Human Sciences and a professor in the Department of History at Virginia Tech.