Web Exclusives: Diseases
The Science of Simulating Disease Spread
Part 3: The Future of Infectious Disease Modeling
I have one of the best jobs ever. I manage scientific research programs for the National Institute of General Medical Sciences at the National Institutes of Health. Some people use the term "faceless bureaucrat," but I don't see my position that way at all. I work for the American people, to support basic research that will improve and protect their health. I spend my days thinking about interesting scientific problems, working with very smart, wise, and kind people, and—yes, it's true—doing paperwork. OK, two out of three isn't bad.
A major part of my job is managing the Models of Infectious Disease Agent Study (MIDAS) program. I facilitate MIDAS's research and ensure that it benefits people. Sometimes that entails creating a space where MIDAS researchers can have good conversations and really dig into their ideas thoroughly. It's very exciting to listen to these conversations, and I always learn something new. Working with a group of really good scientists is like standing at the very top of a mountain and looking out at all the vistas. There are so many possibilities and ideas!
MIDAS has worked a lot on influenza, most recently on the H1N1 influenza pandemic. We also build models of methycillin-resistant Staphlococcus aureus (MRSA), cholera, dengue fever, malaria and other diseases.
We have also worked on a number of policy-related issues, including:
- How much of the population needs to be vaccinated to stop disease transmission?
- What are the highest priority groups for vaccination to stop the disease? What are the highest priority groups to prevent serious disease and death?
- What is the economic impact of interventions?
- What information can we use to figure out where we are on the epidemic curve?
There are many questions still to be answered. Like any area of science, modeling is a continuous process—you never get to the end. There are always more interesting and important questions, and much to do to improve the MIDAS models.
We need to figure out how to model the effects of climate. For example, flu is always worse in the winter than in the summer, partly due to humidity and temperature. Likewise, we need to include knowledge about the evolution of disease-causing agents—like changes in the strains of the flu virus or drug resistance in tuberculosis or Staph. We want to study how disease outbreaks vary geographically. For example, not all places in the United States were equally affected by H1N1. Some cities have very high rates of MRSA, and some are hardly affected. What accounts for this variation?
A huge challenge for the future is too much information! There is a huge amount of data out there about pathogen genomes, immunity, human behavior, climate, evolution, ecology, transportation systems and animal diseases. All of this, and a great deal more, has an impact on the spread of human infectious diseases. How will MIDAS incorporate this amount of information into its models?
Even as there is too much information, there is also too little. Nobody knows how many people in the United States were infected with this year's flu virus—and that's a critical piece of information for the accuracy of models.
The information problem is even worse in some places where infectious diseases are often endemic and where nutrition, access to health care and co-infections are important considerations. If we are to model infectious diseases to protect everybody, we need more information.
MIDAS models are rapidly becoming more sophisticated, informative and useful. We're continuing to learn and grow. It's really interesting. But it also really matters. Not just to me, but to everyone. Our goal is to make all our lives healthier and better.
Read Part 1: Why We Model Infectious Diseases
Read Part 2: Modeling the 2009 H1N1 Pandemic
This article series by Irene Eckstrand first appeared on Year of Science 2009 as blog posts celebrating science and health.