Skip Over Navigation Links

Web Exclusives: Diseases

The Science of Simulating Disease Spread
Part 2: Modeling the 2009 H1N1 Pandemic
Irene Eckstrand
Posted January 4, 2010

2009 was a newsworthy year. Barack Obama was sworn in as President of the United States. The global economy struggled through a major recession. The world celebrated the 200th anniversary of Charles Darwin's birth. And we experienced the first influenza pandemic since 1968.

Irene Eckstrand
Irene Eckstrand gives an overview of the advantages of computer modeling for infectious disease spread.

A pandemic happens when an infectious disease spreads globally, as HIV/AIDS has done for the past few decades. The most famous influenza pandemic happened in 1918 in the midst of World War I. The war caused the deaths of 16 million people, military and civilian, but over 40 million people died of the Great Influenza of 1918. Truly, pandemics can change the course of history.

Influenza pandemics can start when a new virus strain, to which humans have little or no immunity, emerges and starts spreading from person to person. In April of 2009, when public health officials saw early signals in a flu outbreak in Mexico, they rapidly worked to develop a safe and effective vaccine. At the same time, they began to collect and share information about the disease and the people affected. As a result, the H1N1 pandemic of 2009 will likely become the best studied pandemic in history.

Modelers Assist During the Outbreak

The researchers involved in the Models of Infectious Disease Agent Study (MIDAS) along with a global network of scientists with expertise in mathematical, statistical and computational modeling, have worked with the incoming data to understand how H1N1 has been spreading, who's most at risk, the potential impact on health and what interventions might be effective in protecting people from illness and death. The pandemic has given modelers an opportunity to apply their ideas to a real situation and see what they could do to help.

It was clear in the summer of 2009 that there would not be enough vaccine to prevent a wave of flu when schools opened in the fall. The flu was not severe enough to justify closing schools or workplaces. However, certain groups of people (like pregnant women and people with respiratory disease) were at risk of severe disease and death. Was there any way to protect them?

Several MIDAS modelers set out to study this problem, starting by "doing the math"-literally. The first model was a set of equations that suggested that it would save lives and be less costly if vulnerable people had quick and easy access to antiviral drugs to take at the first sign of illness. But given a limited supply of antiviral drugs, was this possible? To answer that question, MIDAS went to a computer model that would include detailed information about vulnerable groups, antiviral drug distribution and behaviors. The results suggest this could work, but the results aren't all ready yet.

Learning from Real Pandemics

Disease models—or any model for that matter—will never predict the future, but they do help us understand trends in disease spread. And, the H1N1 outbreak has taught us a lot.

We learned that very early data based on just a few cases can be misleading. The initial information on H1N1 suggested that it was capable of spreading more quickly than it really was.

We also learned that for this particular disease, closing schools would cost a lot but not be very beneficial. We studied various strategies for distributing scarce vaccines and antiviral drugs to protect the most people.

We learned that even though there are fewer cases of H1N1 than expected, it is still important for people to get vaccinated to protect themselves from potential outbreaks this winter or next year.

And we will continue to learn as the H1N1 outbreak unfolds.

Working with Policymakers

Many MIDAS modelers offer their work to people who make real decisions to protect everyone's lives and health. Throughout the H1N1 pandemic, these collaborations have been critical. Policymakers need to decide things like how many doses of vaccine will be ordered, who will be vaccinated first and whether closing schools is a good idea. Models can often help them think through these difficult questions and make better decisions.

People who make decisions have been able use our modeling results to provide better recommendations that help protect us, our children and our businesses. Modeling infectious diseases has increased the ability of our public health system to protect all of us. That's really important.

Read Part 1: Why We Model Infectious Diseases

Read Part 3: The Future of Infectious Disease Modeling

This article series by Irene Eckstrand first appeared on Year of Science 2009 as blog posts celebrating science and health.

Learn about related research

This page last reviewed on May 5, 2016