Geographic transmission hubs of the 2009 influenza pandemic in the United States

Stephen M. Kissler, Julia R. Gog, Cécile Viboud, Vivek Charu, Ottar N. Bjornstad, Lone Simonsen, Bryan T. Grenfell

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

A key issue in infectious disease epidemiology is to identify and predict geographic sites of epidemic establishment that contribute to onward spread, especially in the context of invasion waves of emerging pathogens. Conventional wisdom suggests that these sites are likely to be in densely-populated, well-connected areas. For pandemic influenza, however, epidemiological data have not been available at a fine enough geographic resolution to test this assumption. Here, we make use of fine-scale influenza-like illness incidence data derived from electronic medical claims records gathered from 834 3-digit ZIP (postal) codes across the US to identify the key geographic establishment sites, or “hubs”, of the autumn wave of the 2009 A/H1N1pdm influenza pandemic in the United States. A mechanistic spatial transmission model is fit to epidemic onset times inferred from the data. Hubs are identified by tracing the most probable transmission routes back to a likely first establishment site. Four hubs are identified: two in the southeastern US, one in the central valley of California, and one in the midwestern US. According to the model, 75% of the 834 observed ZIP-level outbreaks in the US were seeded by these four hubs or their epidemiological descendants. Counter-intuitively, the pandemic hubs do not coincide with large and well-connected cities, indicating that factors beyond population density and travel volume are necessary to explain the establishment sites of the major autumn wave of the pandemic. Geographic regions are identified where infection can be statistically traced back to a hub, providing a testable prediction of the outbreak's phylogeography. Our method therefore provides an important way forward to reconcile spatial diffusion patterns inferred from epidemiological surveillance data and pathogen sequence data.

Original languageEnglish (US)
Pages (from-to)86-94
Number of pages9
JournalEpidemics
Volume26
DOIs
StatePublished - Mar 1 2019

Fingerprint

Pandemics
Human Influenza
Disease Outbreaks
Phylogeography
Electronic Health Records
Population Density
Communicable Diseases
Epidemiology
Incidence
Infection

All Science Journal Classification (ASJC) codes

  • Epidemiology
  • Parasitology
  • Microbiology
  • Public Health, Environmental and Occupational Health
  • Infectious Diseases
  • Virology

Cite this

Kissler, S. M., Gog, J. R., Viboud, C., Charu, V., Bjornstad, O. N., Simonsen, L., & Grenfell, B. T. (2019). Geographic transmission hubs of the 2009 influenza pandemic in the United States. Epidemics, 26, 86-94. https://doi.org/10.1016/j.epidem.2018.10.002
Kissler, Stephen M. ; Gog, Julia R. ; Viboud, Cécile ; Charu, Vivek ; Bjornstad, Ottar N. ; Simonsen, Lone ; Grenfell, Bryan T. / Geographic transmission hubs of the 2009 influenza pandemic in the United States. In: Epidemics. 2019 ; Vol. 26. pp. 86-94.
@article{d360286d5b354054be0daea35f642934,
title = "Geographic transmission hubs of the 2009 influenza pandemic in the United States",
abstract = "A key issue in infectious disease epidemiology is to identify and predict geographic sites of epidemic establishment that contribute to onward spread, especially in the context of invasion waves of emerging pathogens. Conventional wisdom suggests that these sites are likely to be in densely-populated, well-connected areas. For pandemic influenza, however, epidemiological data have not been available at a fine enough geographic resolution to test this assumption. Here, we make use of fine-scale influenza-like illness incidence data derived from electronic medical claims records gathered from 834 3-digit ZIP (postal) codes across the US to identify the key geographic establishment sites, or “hubs”, of the autumn wave of the 2009 A/H1N1pdm influenza pandemic in the United States. A mechanistic spatial transmission model is fit to epidemic onset times inferred from the data. Hubs are identified by tracing the most probable transmission routes back to a likely first establishment site. Four hubs are identified: two in the southeastern US, one in the central valley of California, and one in the midwestern US. According to the model, 75{\%} of the 834 observed ZIP-level outbreaks in the US were seeded by these four hubs or their epidemiological descendants. Counter-intuitively, the pandemic hubs do not coincide with large and well-connected cities, indicating that factors beyond population density and travel volume are necessary to explain the establishment sites of the major autumn wave of the pandemic. Geographic regions are identified where infection can be statistically traced back to a hub, providing a testable prediction of the outbreak's phylogeography. Our method therefore provides an important way forward to reconcile spatial diffusion patterns inferred from epidemiological surveillance data and pathogen sequence data.",
author = "Kissler, {Stephen M.} and Gog, {Julia R.} and C{\'e}cile Viboud and Vivek Charu and Bjornstad, {Ottar N.} and Lone Simonsen and Grenfell, {Bryan T.}",
year = "2019",
month = "3",
day = "1",
doi = "10.1016/j.epidem.2018.10.002",
language = "English (US)",
volume = "26",
pages = "86--94",
journal = "Epidemics",
issn = "1755-4365",
publisher = "Elsevier",

}

Kissler, SM, Gog, JR, Viboud, C, Charu, V, Bjornstad, ON, Simonsen, L & Grenfell, BT 2019, 'Geographic transmission hubs of the 2009 influenza pandemic in the United States', Epidemics, vol. 26, pp. 86-94. https://doi.org/10.1016/j.epidem.2018.10.002

Geographic transmission hubs of the 2009 influenza pandemic in the United States. / Kissler, Stephen M.; Gog, Julia R.; Viboud, Cécile; Charu, Vivek; Bjornstad, Ottar N.; Simonsen, Lone; Grenfell, Bryan T.

In: Epidemics, Vol. 26, 01.03.2019, p. 86-94.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Geographic transmission hubs of the 2009 influenza pandemic in the United States

AU - Kissler, Stephen M.

AU - Gog, Julia R.

AU - Viboud, Cécile

AU - Charu, Vivek

AU - Bjornstad, Ottar N.

AU - Simonsen, Lone

AU - Grenfell, Bryan T.

PY - 2019/3/1

Y1 - 2019/3/1

N2 - A key issue in infectious disease epidemiology is to identify and predict geographic sites of epidemic establishment that contribute to onward spread, especially in the context of invasion waves of emerging pathogens. Conventional wisdom suggests that these sites are likely to be in densely-populated, well-connected areas. For pandemic influenza, however, epidemiological data have not been available at a fine enough geographic resolution to test this assumption. Here, we make use of fine-scale influenza-like illness incidence data derived from electronic medical claims records gathered from 834 3-digit ZIP (postal) codes across the US to identify the key geographic establishment sites, or “hubs”, of the autumn wave of the 2009 A/H1N1pdm influenza pandemic in the United States. A mechanistic spatial transmission model is fit to epidemic onset times inferred from the data. Hubs are identified by tracing the most probable transmission routes back to a likely first establishment site. Four hubs are identified: two in the southeastern US, one in the central valley of California, and one in the midwestern US. According to the model, 75% of the 834 observed ZIP-level outbreaks in the US were seeded by these four hubs or their epidemiological descendants. Counter-intuitively, the pandemic hubs do not coincide with large and well-connected cities, indicating that factors beyond population density and travel volume are necessary to explain the establishment sites of the major autumn wave of the pandemic. Geographic regions are identified where infection can be statistically traced back to a hub, providing a testable prediction of the outbreak's phylogeography. Our method therefore provides an important way forward to reconcile spatial diffusion patterns inferred from epidemiological surveillance data and pathogen sequence data.

AB - A key issue in infectious disease epidemiology is to identify and predict geographic sites of epidemic establishment that contribute to onward spread, especially in the context of invasion waves of emerging pathogens. Conventional wisdom suggests that these sites are likely to be in densely-populated, well-connected areas. For pandemic influenza, however, epidemiological data have not been available at a fine enough geographic resolution to test this assumption. Here, we make use of fine-scale influenza-like illness incidence data derived from electronic medical claims records gathered from 834 3-digit ZIP (postal) codes across the US to identify the key geographic establishment sites, or “hubs”, of the autumn wave of the 2009 A/H1N1pdm influenza pandemic in the United States. A mechanistic spatial transmission model is fit to epidemic onset times inferred from the data. Hubs are identified by tracing the most probable transmission routes back to a likely first establishment site. Four hubs are identified: two in the southeastern US, one in the central valley of California, and one in the midwestern US. According to the model, 75% of the 834 observed ZIP-level outbreaks in the US were seeded by these four hubs or their epidemiological descendants. Counter-intuitively, the pandemic hubs do not coincide with large and well-connected cities, indicating that factors beyond population density and travel volume are necessary to explain the establishment sites of the major autumn wave of the pandemic. Geographic regions are identified where infection can be statistically traced back to a hub, providing a testable prediction of the outbreak's phylogeography. Our method therefore provides an important way forward to reconcile spatial diffusion patterns inferred from epidemiological surveillance data and pathogen sequence data.

UR - http://www.scopus.com/inward/record.url?scp=85051960220&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85051960220&partnerID=8YFLogxK

U2 - 10.1016/j.epidem.2018.10.002

DO - 10.1016/j.epidem.2018.10.002

M3 - Article

VL - 26

SP - 86

EP - 94

JO - Epidemics

JF - Epidemics

SN - 1755-4365

ER -