Donald Olson, MPH, Columbia University, 722 W. 168th St. Kolb Research Annex #724, New York, NY 10032, (212)543-5960, dro7@columbia.edu, Lone Simonsen, PhD, National Institute of Allergy and Infectious Disease, National Institutes of Health, 6700-B Rockledge Drive, Room 3153, Bethesda, MD 20892-7630, Paul Edelson, MD, Center For Public Health Preparedness, Columbia University, 722 W. 168th St., New York, NY 10032, and Stephen S. Morse, PhD, Center for Public Health Preparedness, Columbia University Mailman School of Public Health, 722 West 168th Street, New York, NY 10032.

The influenza pandemic of 1918 caused an estimated 40-50 million excess deaths worldwide, with an unusual focus among young otherwise healthy adults. The epidemiologist Wade Hampton Frost suggested an "early phase" of the pandemic was responsible for regional epidemics of pneumonia and influenza (P&I) mortality, most evident in New York City, six months before the explosive autumn wave. To test this hypothesis we undertook an analysis of the age-specific pattern of mortality in this period. A Serfling regression model was applied to monthly age-specific New York City data from 1907-1921 to generate baselines of mortality; excess mortality was computed for each age group. The age-specific risk of death in the "early phase" and pandemic season were calculated relative to non-pandemic epidemic influenza seasons. Over all age groups, we estimated 29,000 all-cause and P&I excess deaths occurred during the pandemic wave from September 1918 to April 1919. During the preceding season of January to April 1918 the model detected 4,300 all-cause and 2,600 P&I deaths. During the latter season, the age pattern of mortality changed markedly from predominantly among those 65 and over in January to predominantly those under 45 by March. Our analysis confirms the epidemic in early 1918 New York City was consistent with a herald wave of the pandemic. The pattern of epidemic to pandemic age-specific relative risk is consistent with the recycling of an antigenically similar influenza subtype a half century before.

Learning Objectives:

This presentation provides a novel analysis of a previously unrecognized data source, and should provide the attendee with the chance to learn how to
• Recognize the shifting age distribution of epidemic and pandemic influenza mortality,
• Analyze routinely collected mortality statistics to estimate influenza-related mortality
• Identify evidence consistent with influenza herald waves and pandemic recycling.

Keywords: Infectious Diseases, Surveillance

Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.