The covid-19 crisis has prompted new appreciation of how the public thinks about numbers during an epidemic. Tracking daily reports of cases and deaths, making sense of how different models predict trajectories and engaging in “social distancing” measures intended to “flatten the curve” testify to how an awareness of disease data has transformed our thinking as well as daily behavior.

The 1918 influenza epidemic was also marked by an unprecedented attention to statistics as the crisis unfolded. Newspapers across the United States provided daily reports on the number of cases and deaths, particularly as numbers accelerated in the acute weeks of the epidemic. Readers of newspapers in the fall of 1918, much like today’s consumers of digitally mediated news, could follow the epidemic at the national level and in their own communities.

Understanding how influenza statistics were reported in 1918 can inform our understanding of the data circulating about covid-19. The key lesson? Context matters. Death toll statistics are much more useful when broken down by categories such as race and age, and the case fatality rate — the number of deaths among those sick or (in the case of covid-19, with positive tests) — is more useful than raw death totals. And, these numbers don’t tell the story of the people battling disease and treating patients with the pressure of the media spotlight on their every action.

Washington newspapers at the peak of the influenza epidemic anticipated what we see now in reports on covid-19. On Oct. 14, for example, a front-page article in the Evening Star included a table showing the daily number of influenza cases and death since Oct. 1, adding up to a total of 12,847 cases and 544 deaths in just two weeks. On the same day, a front-page headline in the Washington Times declared 96 influenza deaths in 24 hours was “the biggest toll recorded.”

These reports appeared day after day in the weeks that followed, recording the number of cases and deaths as a narrative of escalating loss. The numbers were reported by the District Health Office, but the newspapers emphasized their significance with banner headlines, boxed tables and lists of individual deaths, all of which marked a significant break from the usual reporting practices.

The daily toll for October 1918 can be illustrated using a chart familiar to anyone tracking the first weeks of covid-19. On Oct. 1, the city health office reported 162 cases and 23 deaths. After a brief lull, both numbers began to increase significantly. One week later, the health officer reported more than 50 deaths daily. The peak number of cases, 2,174, was reported on Oct. 8, just a few days after an order closing schools and businesses was implemented.

After Oct. 8, the number of cases began to decline, yet deaths continued to increase in the days that followed, reaching the peak number, 92, on Oct. 12. Although the number of deaths reported each day declined for the rest of the month, the rate was more gradual, compared to the decline in cases. On Oct. 31, two days after announcing that closing orders would gradually come to an end, the city health office reported nine deaths, less than one-half the toll reported on the first days of the month.

These numbers were published by the Evening Star on Nov. 4, confirming that readers had access to recent data about the number of cases and deaths.

The numbers revealed the severity of the pandemic — disabusing anyone of the belief that the sudden and remarkably lethal influenza outbreak bore any relation to the seasonal flu. For the decade preceding the 1918 influenza, Washington recorded an average of 80 deaths per year from influenza (or grippe, as it was registered early in the century). On Oct. 10, 12 and 13, 1918, the deaths reported in a single day exceeded those yearly averages. The total number of influenza deaths in the last three months of 1918, nearly 2,000 deaths, was higher than the total deaths from influenza reported since the beginning of the 20th century.

The data also exposed structural inequalities along racial and economic lines. In October 1918, Washington recorded 344 “Colored” deaths and 1,127 “White” deaths from influenza. Yet, those raw totals were deceiving. The nearly 10 percent case fatality rate for African Americans was far higher than the less than six percent rate for whites. As demonstrated by historian Vanessa Northington Gamble, the disparity stemmed from structural conditions, exacerbated and reinforced by the influenza epidemic, and responsible for higher death rates among African American populations before, during and after 1918. Although the city health officer documented the number of deaths by race, newspapers rarely commented on the differential impact of the disease along racial lines, just as the initial reports on covid-19 often failed to recognize the differential impact along race and class lines.

Yet these numbers do tell another, perhaps more comforting, story of survival amid devastating loss. The nearly 25,000 cases of influenza reported by physicians and hospitals in just one month, October 1918, represented nearly six percent of city’s population. But having influenza was far from a sentence of death — though the banner headlines and hundreds of deaths might have made it seem to be the case. In reality, the vast majority of those who became ill with influenza survived.

In fact, based on official reports from physicians and hospitals, more than nine out of every 10 victims survived falling ill with influenza during this deadly epidemic. And this number is slightly misleading — in a positive way. As city health officials later acknowledged, the reported cases represented a fraction of the total number of cases, which they estimated to be closer to 50,000 cases of influenza, making up 10 percent of the total population.

The officially reported totals never acknowledged the number of Washingtonians recovering or never becoming ill, who nonetheless had influenza. Given that the case reports understated the actual number ill with influenza, the survival rate was certainly higher than official calculations. Even in the deadliest epidemic in world history, in other words, only about 10 percent of the District population fell ill and more than 90 percent of the sick did survive.

The toll of the 1918 epidemic can be counted in total numbers, as fatality rates and as stories of individual loss. City health officers emerged from the influenza epidemic with renewed confidence in their ability to use data to anticipate and manage disease outbreaks. The impact of these daily reports on the population is more difficult to determine from the newspapers. Oral histories with influenza survivors describe the personal experience of hearing about numerous deaths or seeing frequent funerals, but rarely address how the news was reported (which is partly a consequence of survivors usually having been quite young in 1918).

The history of the 1918 influenza epidemic is also a story of survival, even in the midst of unprecedented losses. As we enter into the next stage of covid-19, which may include a spike in cases as states start to reopen and social distancing eases, these numbers from the 1918 influenza provide a context for understanding how history can shape our expectations for the future, and they also offer useful reminders about the need to understand numbers in context — a statistic that looks bad as a raw number may, in fact, signal a ray of hope when understood in light of the more complete data.