The Geographic Synchrony of Seasonal Influenza: A Waves across Canada and the United States

Abstract

BackgroundAs observed during the 2009 pandemic, a novel influenza virus can spread globally before the epidemic peaks locally. As consistencies in the relative timing and direction of spread could form the basis for an early alert system, the objectives of this study were to use the case-based reporting system for laboratory confirmed influenza from the Canadian FluWatch surveillance program to identify the geographic scale at which spatial synchrony exists and then to describe the geographic patterns of influenza A virus across Canada and in relationship to activity in the United States (US).Methodology/Principal FindingsWeekly laboratory confirmations for influenza A were obtained from the Canadian FluWatch and the US FluView surveillance programs from 1997/98 to 2006/07. For the six seasons where at least 80% of the specimens were antigenically similar, we identified the epidemic midpoint of the local/regional/provincial epidemics and analyzed trends in the direction of spread. In three out of the six seasons, the epidemic appeared first in Canada. Regional epidemics were more closely synchronized across the US (3–5 weeks) compared to Canada (5–13 weeks), with a slight gradient in timing from the southwest regions in the US to northeast regions of Canada and the US. Cities, as well as rural areas within provinces, usually peaked within a couple of weeks of each other. The anticipated delay in peak activity between large cities and rural areas was not observed. In some mixed influenza A seasons, lack of synchronization sub-provincially was evident.Conclusions/SignificanceAs mixing between regions appears to be too weak to force a consistency in the direction and timing of spread, local laboratory-based surveillance is needed to accurately assess the level of influenza activity in the community. In comparison, mixing between urban communities and adjacent rural areas, and between some communities, may be sufficient to force synchronization.

Distribution of influenza and other acute respiratory viruses during the first year after the 2009–2010 influenza pandemic in the English‐ and Dutch‐speaking Caribbean countries

“…Virological data indicate that the influenza A(H1N1)pdm09 may continue to be a common virus in circulation in the Caribbean with displacement of the previous strain of influenza A(H1N1) noted. Approximately one‐third of influenza A(H1N1)pdm09 cases reported gastrointestinal symptoms, which is consistent with data previously reported . Influenza A(H3N2) was the most frequently confirmed influenza virus, and overall influenza activity peaked in October 2010 with cases confirmed in five Caribbean countries.…”

“…Approximately one-third of influenza A(H1N1)pdm09 cases reported gastrointestinal symptoms, which is consistent with data previously reported. 14,22,23 Influenza A(H3N2) was the most frequently confirmed influenza virus, and overall influenza activity peaked in October 2010 with cases confirmed in five Caribbean countries. Respiratory syncytial virus was the most frequently confirmed virus among patients with ARI in the first year after the influenza A(H1N1)pdm09 pandemic with a majority of cases occurring in children under 5 years of age.…”

Deviations in influenza seasonality: odd coincidence or obscure consequence?

“…It has been shown that there is a remarkable degree of synchronization of influenza outbreaks at a regional level [20,[31][32][33] as well as between countries [24,34] in temperate climates. Investigators relate synchronization to globalization [35,36], social mixing patterns [37], and transportation networks [30,38].…”

Shifting patterns of seasonal influenza epidemics

“…Influenza diffuses through complex spatiotemporal patterns that appear to change across scales [62,61,32,15,31,17,21,23,39,65,49,22]. A hierarchical spread combining a wave-like behavior on the small scale and long-range seeding events synchronizing distant populations was reported by several studies [62,61,17,65,50]. This is generally explained by the multiscale nature of mobility patterns of individuals [12], with commuting mainly acting on the short distance whereas air travel flows are responsible for non-local diffusion [12,61,17,22].…”

“…in Israel [32] and in France [15,21,23] where no dominant transportation mode was identified [23]. Spatial patterns of influenza spread also depend on the dominant viral strain circulating [61,50,19,40,43], with a tendency of B-dominated influenza seasons to be characterized by slower and later epidemics compared to A-dominated epidemics [50,19].…”