Nonpharmaceutical Influenza Mitigation Strategies, US Communities, 1918–1920 Pandemic

We studied nonpharmaceutical interventions used to mitigate the second, and most deadly, wave of the 1918–1920 influenza pandemic in the United States. We conclude that several small communities implemented potentially successful attempts at preventing the introduction of influenza.

sites first by consulting Jordan's 1927 text, Epidemic Influenza (1). We then verified and modified this list by reviewing 240 federal, 92 state (from 40 states), and 25 special local reports and documents. We conducted in situ archival research at 34 locations and examined >1,400 newspaper and contemporary medical and scientific journal articles for the 1918-1920 period.
The communities we identified were diverse and had unique characteristics. Fletcher, Vermont (population 737), was simply too small to suggest that its success resulted from anything more than remote location, good fortune, or the ways in which the virus skipped some communities altogether for unknown reasons (8)(9)(10). The Trudeau Tuberculosis Sanitarium (9) and the Western Pennsylvania Institution for the Blind (10) were already de facto quarantine islands because of the era's prevailing views toward confinement of the contagious and the disabled. Princeton University provided a good example of how a social institution with some measure of control over its population might implement NPIs to protect itself (11).
The US Naval Base at Yerba Buena Island in San Francisco Bay (12) and the mining town of Gunnison, Colorado (13), also offer potential lessons for contemporary pandemic influenza preparedness planning. Under the direction of public health officers, the still-healthy island and mountain town essentially cut off all contact with the outside world to shield themselves from the incursion of influenza. The 2 sites saw almost no cases of infection and thus experienced no deaths, for 2 and 4 months, respectively.
Most important, these communities enacted a policy we have termed protective sequestration, or the measures taken by the authorities to protect a defined and stillhealthy population from infection before it reaches that population. These measures include the following: 1) prohibitions on members of the community from leaving the site; 2) prohibitions against visitors from entering a circumscribed perimeter; 3) typically placing in quarantine those visitors who are allowed to enter for a period of time before admission; and 4) if available, taking advantage of geographic barriers, such as an island or remote location.
Several themes emerged from our historical research. First, coordination among public agencies is essential to any effective public health response. Despite some tension among city, county, and state officials in Gunnison, their relatively smooth cooperation may have played a role in their implementing and maintaining strict public health measures. Second, neither Gunnison nor Yerba Buena could have escaped the flu without full cooperation from the local population. Gunnison's low population density and self-sufficient ranching lifestyle made it easier for residents to bide their time (Figure 1). At Yerba Buena, the military chain of command mandated the cooperation of sailors and allowed the island's commander to close off the base from the outside world with little interference (Figure 2). Finally, these communities had the advantage of early warnings to prepare their populations. Both tracked influenza's westward movement from August to September and, unlike communities along the East Coast, could implement protective health strategies before cases appeared at their doorsteps.
One would like to think that the 6 communities we identified fared better than others because of the NPIs they enacted. While we cannot prove this for any of them, the case is perhaps strongest for Yerba Buena and possibly Gunnison. Further complicating our task, in addition to the uneven quality and quantity of information available for study, is that some of these communities were sparsely populated and geographically isolated, and all of them were subject to the vagaries of how the influenza virus affected populations. Indeed, these communities represent the exception rather than the rule in terms of how most American communities experienced the influenza pandemic of 1918-1920 (14,15). This leads to several intriguing questions regarding what these escape communities can teach us about pandemic preparations today, let alone the question of whether such measures can even be replicated.

Conclusions
First, protective sequestration, if enacted early enough in the pandemic, crafted so as to encourage the compliance of the population involved, and continued for the lengthy time period in which the area is at risk, stands the best chance of guarding against infection. Second, available data from the second wave of the 1918-1920 influenza pandemic fail to show that any other NPI (apart from protective sequestration) was, or was not, effective in preventing the spread of the virus. Despite implementing several NPIs, most communities sustained considerable illness and death. We could not assess how the timing of NPI implementation across the nation affected disease mitigation efforts nor whether these NPIs lessened what might have been even higher rates had these measures not been in place in various locations. Moreover, we could not locate any consistent, reliable data supporting the conclusion that face masks, as available and as worn during the 1918-1920 influenza pandemic, conferred any protection to the populations that wore them (16). In fact, evidence suggests that in most American communities NPIs did not prevent the spread of virus in 1918. What remains unclear is the extent to which they may have been partially effective in reducing spread or mitigating community impact.
However inconclusive the data from 1918 are, the collective experiences of American communities from the pandemic are noteworthy, especially in light of the fact that, if faced with a pandemic today, we would likely rely on many of these same NPIs to attempt to mitigate the spread of the infection until pharmacological supplies of vaccine and antiviral agents were available (17)(18)(19). It is true that the United States of today is a much different nation than it was in 1918, with a larger, more mobile, and more complex society. It is equally true that the communities we examined were all small and relatively isolated (or isolatable). Nevertheless, in the event of another influenza pandemic, many specific subcommunities (e.g., military installations, college and university campuses, nursing homes) may wish to consider protective sequestration measures as potential means to prevent or delay the onset of epidemic influenza in their populations.