Modeling Contagion

Posted on November 9, 2011 by


Image Credit: Anna Tanczos. Wellcome Images. This digital graphic of the H1N1 virus particle shows the viral surface proteins hemagglutinin (H) and neuraminidase (N) in red and yellow, respectively. The genetic material that lies at the center of the H1N1 particle is single-stranded RNA, shown in purple.

Many of you have probably come across the recent Steven Soderbergh thriller, Contagion, complete with an all-star cast of Kate Winslet, Matt Damon, Jude Law and Lawrence Fishburne, furiously racing against time to beat a novel viral pandemic that transmits alarmingly fast!

The trailer for Contagion got me really excited – here was a film that promised to show off some ‘real’ science! No ill-will intended to sci-fi movies and their aficionados; I was just ready to see some credit go to good science, and more plausible scientific achievements. The merits and demerits of the film itself have been reviewed in the popular media, but what struck me most was that the creators of the story and film had gone to great lengths to portray the science as accurately as possible. Indeed, they had closely consulted with Dr. W. Ian Lipkin, Professor of Epidemiology at Columbia University’s Mailman School of Public Health, and a renowned expert in the field. A+ on that effort!

So, could a similar real-world pandemic ever striking us to that magnitude? Although the extent of the pandemic might have been exaggerated in the film, the depicted networks for detection and investigation of fast-spreading infectious diseases really do exist.

National and international agencies such as the Centers for Disease Control and Prevention (CDC), World Health Organization (WHO), and other local health agencies such as our county-level health departments, monitor disease outbreaks closely; moreover, several research projects use existing data to project the severity of future epidemics. One such research effort takes place in Pittsburgh, at the University of Pittsburgh’s Graduate School of Public Health.

In 2004, the Modeling of Infectious Disease Agent Study, or MIDAS, was established by the NIH as a national collaborative team of researchers who would develop new models of infectious disease dynamics. This study was designed to better prepare national agencies to detect and respond to new infectious disease agents, and to the re-emergence of known infectious diseases. The aim is to model a disease outbreak and vary parameters to include different scenarios, such as travelers introducing an infectious agent to a previously uninfected population (as in the film!) These models can also allow us to predict the efficacy of treatments such as vaccines, and contrast disease transmission in the presence or absence of an intervention.

In 2009, Pitt’s Graduate School of Public Health became home to a MIDAS National Center for Excellence – one of two centers to be established nationwide; the other center runs at the Harvard School of Public Health. Dr. Donald Burke, a global health expert, Dean of the Pitt Graduate School of Public Health, UPMC-Jonas Salk Chair in Global Health, Associate Vice Chancellor for Global Health, and Director of the Center for Vaccine Research, serves as the Principal Investigator for this Center. The Pittsburgh Center comprises research teams that are expert in disciplines such as public health, computational biology, medicine, law and policy. The center is divided up into six projects that together allow for the development of computational tools and modeling for the general areas of parameter estimation, behavior, disease evolution, environmental effects, policy preparedness and applied modeling.

MIDAS researchers are currently focused on diseases such as tuberculosis, malaria, MRSA, dengue fever and influenza. In 2011, MIDAS teams in Pittsburgh were involved in studies on issues such as the economic cost of school closure in Pennsylvania during the H1N1 flu epidemic in 2009; the role of social networks in the transmission of a community outbreak of H1N1; the implications of inequitable access to vaccines in the spread of a flu epidemic; the economic implications of a vaccine against hookworm; several other vaccine-related studies in the US and abroad, among others.

A great example of the use of MIDAS-generated simulations comes from recently published research on the effect of inequitable access to vaccines during an epidemic.  This study modeled a virtual Washington, D.C. population of 7,414,562 individuals using information from the census, like demographics, location, income status, health status, and employment status. The simulation then tested the outcome of an H1N1 flu outbreak in the absence of vaccines, and in the case of inequitable distribution of vaccines by socioeconomic status. The results showed that when lower socioeconomic sections of the population receive lower access to vaccines, even if there is increased access for the wealthy, a greater number of the overall population become infected with the infectious agent.  Inequitable access models also resulted in greater numbers of infected individuals at the peak of the epidemic. These models give us valuable insights into how to manage future vaccine distribution efforts. Currently, the United States utilizes MIDAS models for determining how to slow down flu transmission at the early onset of the outbreak, to deal with the delay in drug supply; the WHO uses MIDAS research to make decisions on global influenza medicinal stockpiles.

Keep an eye out for MIDAS-generated research in the news, as this network of Pittsburgh scientists with diverse expertise – ranging from biology to computer science – work together to generate research with far-reaching applications!  And so, yes, Steven Soderbergh and his team were absolutely correct in casting the best artists in the industry to represent the fictional equals of the outstanding scientists doing important science to benefit  local, national and global communities!

Posted in: In depth