Epidemiologic Modeling with FluSurge for Pandemic (H1N1) 2009 Outbreak, Queensland, Australia

Philip R.A. Baker, Jiandong Sun, James Morris, and Amanda Dines
Author affiliations: Queensland Health, Brisbane, Queensland, Australia (P.R.A. Baker, J. Sun, A. Dines, J. Morris); Queensland University of Technology, Brisbane (P.R.A. Baker, J. Sun); and University of Queensland, Brisbane (A. Dines)
Suggested citation for this article

Abstract
At the beginning of the pandemic (H1N1) 2009 outbreak, we estimated the potential surge in demand for hospital-based services in 4 Health Service Districts of Queensland, Australia, using the FluSurge model. Modifications to the model were made on the basis of emergent evidence and results provided to local hospitals to inform resource planning for the forthcoming pandemic. To evaluate the fit of the model, a comparison between the model's predictions and actual hospitalizations was made. In early 2010, a Web-based survey was undertaken to evaluate the model's usefulness. Predictions based on modified assumptions arising from the new pandemic gained better fit than results from the default model. The survey identified that the modeling support was helpful and useful to service planning for local hospitals. Our research illustrates an integrated framework involving post hoc comparison and evaluation for implementing epidemiologic modeling in response to a public health emergency.

Influenza pandemics can result in substantial excess illness and death (1,2). In the past century, 3 influenza pandemics have occurred, commonly referred to as the 1918 Spanish flu, the 1957 Asian flu, and the 1968 Hong Kong flu. It is estimated that 40–50 million persons died during the 1918 pandemic, which is considered to be one of the most severe disease events in history (1,2). The following 2 relatively mild pandemics caused approximately 2 million (1957) and 1 million (1968) deaths, respectively (1,2).
It is widely foreseen that excess illness and deaths in a future pandemic may place serious demands on and even exhaust the available hospital resources in a community. For example, modeling studies consistently predict that current intensive care unit (ICU) services in several industrialized countries could be overwhelmed during a future event of pandemic influenza (3–5). The prediction of the expected impact of an emerging pandemic would enable appropriate preparation to be made without diversion of excess resources and thus have the potential to reduce pandemic- and nonpandemic-related illness and death.
Since April 2009, a new variant of influenza virus A (H1N1) initially discovered in Mexico and the United States has caused a wave of pandemic influenza. On May 8, 2009, the first case of pandemic (H1N1) 2009 influenza in Australia was confirmed in Queensland (6). During the initial "Delay" and "Contain" phases of the Australian Health Management Plan for Pandemic Influenza (7), during April 26–June 22, 2009, a total of 593 laboratory-confirmed cases were notified in Queensland. Among the patients these cases represent, 16 hospitalizations and no deaths were reported (8). However, the reported number of cases, hospitalizations, and deaths may only represent a small fraction of the true numbers because not all persons who are infected seek medical care and have a specimen collected. Further, not all specimens will have positive results and be reported (9).
To assist hospital planners in their preparation for pandemic (H1N1) 2009, at the emerging stage of the pandemic in 2009, regional epidemiologists made predictions of the potential need for general hospital resources and ICU services for 4 Health Service Districts in Queensland, Australia, by using the FluSurge model developed by the Centers for Disease Control and Prevention in the United States (10,11). A follow-up survey was conducted in early 2010 to evaluate the application of these predictions. The aims of the studies were to describe the modeling work, to explore the fitness of the predictions to the actual hospital data, and to evaluate the application and usefulness of the modeling to hospital planners.
full-text ►
Modeling with FluSurge for Pandemic, Australia | CDC EID: - Enviado mediante la barra Google