Novel measurement of spreading pattern of influenza epidemic by using weighted standard distance method: retrospective spatial statistical study of influenza, Japan, 1999–2009Reportar como inadecuado

Novel measurement of spreading pattern of influenza epidemic by using weighted standard distance method: retrospective spatial statistical study of influenza, Japan, 1999–2009 - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

International Journal of Health Geographics

, 11:20

First Online: 19 June 2012Received: 21 February 2012Accepted: 22 May 2012DOI: 10.1186-1476-072X-11-20

Cite this article as: Shobugawa, Y., Wiafe, S.A., Saito, R. et al. Int J Health Geogr 2012 11: 20. doi:10.1186-1476-072X-11-20


BackgroundAnnual influenza epidemics occur worldwide resulting in considerable morbidity and mortality. Spreading pattern of influenza is not well understood because it is often hampered by the quality of surveillance data that limits the reliability of analysis. In Japan, influenza is reported on a weekly basis from 5,000 hospitals and clinics nationwide under the scheme of the National Infectious Disease Surveillance. The collected data are available to the public as weekly reports which were summarized into number of patient visits per hospital or clinic in each of the 47 prefectures. From this surveillance data, we analyzed the spatial spreading patterns of influenza epidemics using weekly weighted standard distance WSD from the 1999-2000 through 2008-2009 influenza seasons in Japan. WSD is a single numerical value representing the spatial compactness of influenza outbreak, which is small in case of clustered distribution and large in case of dispersed distribution.

ResultsWe demonstrated that the weekly WSD value or the measure of spatial compactness of the distribution of reported influenza cases, decreased to its lowest value before each epidemic peak in nine out of ten seasons analyzed. The duration between the lowest WSD week and the peak week of influenza cases ranged from minus one week to twenty weeks. The duration showed significant negative association with the proportion of influenza A-H3N2 cases in early phase of each outbreak correlation coefficient was −0.75, P = 0.012 and significant positive association with the proportion of influenza B cases in the early phase correlation coefficient was 0.64, P = 0.045, but positively correlated with the proportion of influenza A-H1N1 strain cases statistically not significant. It is assumed that the lowest WSD values just before influenza peaks are due to local outbreak which results in small standard distance values. As influenza cases disperse nationwide and an epidemic reaches its peak, WSD value changed to be a progressively increasing.

ConclusionsThe spatial distribution of nationwide influenza outbreak was measured by using a novel WSD method. We showed that spreading rate varied by type and subtypes of influenza virus using WSD as a spatial indicator. This study is the first to show a relationship between influenza epidemic trend by type-subtype and spatial distribution of influenza nationwide in Japan.

KeywordsWeighted standard distance Influenza Spatial compactness AbbreviationsWSDWeighted Standard Distance

IDSCInfectious Disease Surveillance Center

IDWRInfectious Diseases Weekly Report

IASRInfectious Agents Surveillance Report.

Electronic supplementary materialThe online version of this article doi:10.1186-1476-072X-11-20 contains supplementary material, which is available to authorized users.

Download fulltext PDF

Autor: Yugo Shobugawa - Seth A Wiafe - Reiko Saito - Tsubasa Suzuki - Shinako Inaida - Kiyosu Taniguchi - Hiroshi Suzuki


Documentos relacionados