A Spatial Variant of the Basic Reproduction Number for the New Orleans Yellow Fever Epidemic of 1878 Andrew Curtis and Jacqueline W. Mills University of Southern California Jason K. Blackburn California State University, Fullerton A spatial variant of the basic reproduction number (R 0 ), here defined as the number of subsequent deaths attributed to an initial mortality, can be used to identify geographic variation within an epidemic. A spatial R 0 was calculated at the neighborhood level, here defined by a 50-m buffer surrounding an index case, for mortality data from the 1878 yellow fever epidemic of New Orleans. The highest number of secondary mortalities linked to a neighborhood index case was twelve, with a further eighty-seven extrapolated morbidity cases. Results also highlight the importance of multideath residences and cultural contacts in neighborhood-level disease spread. Key Words: basic reproduction number, medical geography, New Orleans, yellow fever. W hen the structure of an epidemic is con- sidered, and especially when the suscep- tibility of an area to a major disease outbreak is predicted, it is important to understand the re- lationship between those susceptible to the dis- ease, those infected with the disease, and those who have recovered. In addition, the basic reproduction number (R 0 ) provides an indication of whether the introduction of disease will result in a localized burnout or signal the beginning of an epidemic that could move through all geo- graphic scales. This study uses data from the 1878 yellow fever epidemic of New Orleans to calculate a spatial variant of the R 0 , here defined as the number of mortalities that can be linked to a first disease occurrence in a neighborhood. This spatial variant of the R 0 adds further insight into understanding the spatial and temporal complexity of urban epidemics. Even though the data analyzed in this study are from the late 1800s, this insight may prove beneficial in under- standing how to control the spread of emerging and reemerging pathogens in the Americas. Two classic concepts concerning the structure of a (mosquito-vectored) epidemic are the threshold of the disease (Ross 1909, 1911) and the associated relationship between Susceptible, Infected, and Removed people, commonly re- ferred to as the SIR model (Kermack and McK- endrick 1991). In the creation of the traditional epidemiological S-shaped curve only S and I are used, with relatively few I leading to slow pro- gression of the disease at the base of the curve, whereas after mass exposure the disease again slows at the top of the curve because the number of S has been considerably reduced. R may occur either because of previous exposure (both during the epidemic or during an earlier outbreak) or through vaccination. A further epidemiological concept used to characterize the likelihood of disease outbreak is the basic reproduction num- ber (R 0 ), often defined as the number of second- ary infections resulting from the introduction of an index or first case (Anderson and May 1991). This definition usually does not include a mea- sure of space or spatial scale, even though it has been noted that both local and global diffusion patterns can affect the shape of the epidemio- logical curve, with the traditional S curve actually containing multiple peaks as new susceptible populations are found spatially. In other words, there are mini-epidemics existing within the epidemic (Bailey 1975; Watts et al. 2005). Public health reports of the 1800s provide a rich source of epidemic data. In many cases these reports contain location-specific disease events suitable for investigating the spatial and cultural interconnections within an epidemic. The Offi- cial Report of the Deaths from Yellow Fever as Re- ported by the New Orleans Board of Health (ORD The Professional Geographer, 59(4) 2007, pages 492–502 r Copyright 2007 by Association of American Geographers. Initial submission, August 2005; revised submissions, June and October 2006; final acceptance, April 2007. Published by Blackwell Publishing, 350 Main Street, Malden, MA 02148, and 9600 Garsington Road, Oxford OX4 2DQ, U.K.