More secure bases for decision needed during pandemics
Knowledge of how an infection spreads is inadequate, which makes efforts in a pandemic insecure. The outbreak of swine flu in 2009 led to different measures in different countries. Researchers at Linköping University, among other places, are now trying to find better models for decision making.
Schools were closed in the US, but not in Canada. Sweden and a handful of other countries instituted mass vaccinations, while most other countries refrained from it. The 2009 swine flu revealed the prevailing insecurity and lack of knowledge of how to best confront pandemics. Decision makers in various countries came to completely different conclusions as to what should be done.
Toomas Timpka and his colleagues have developed what they call a protocol for improving and structuring the handling of information and decisions during large influenza outbreaks.
“Back during the avian flu, we saw that information systems which supported the necessary quick decision-making processes were missing,” he says.
It’s an issue of such basic knowledge as people’s daily contact patterns, how they move about, what transportation they use, who they meet, and what type of contact they have.
“We’re missing data here, but we still simulate and count on that,” Timpka says.
The problem with influenza outbreaks is, of course, that decisions must be made quickly. To initiate a study that yields results two years later is of no help in acute situations. Other bases must therefore be found and systematised for decision-making.
PROSPER, the tool presented by Timpka and his colleagues, is made roughly out of three ingredients: 1. a mathematical model for short-term predictions (nowcasting) used by people like meteorologists; 2. simulations of how people live and move around, which show how an infection spreads; and 3. calibrations of existing epidemiological knowledge of how the latest influenza outbreaks looked.
“We’re not really contributing anything new, but we’re putting together and structuring existing information systems,” Timpka says. He thinks that the response to the swine flu would have been different if such a model had been used at the time.
“After early analyses, there were a lot of us who were sceptical that the flu would pick up speed, and it didn’t.” The knowledge base for the decision to vaccinate was dubious.
Moreover, the definition itself of pandemic is outdated, something that explains why the swine flu was given the label despite very few lives being lost.
“According to the World Health Organization’s definition, a pandemic breaks out if a virus causes confirmed cases in every continent,” says Timpka, an epidemiologist and informatician.
“That definition may have worked before, but with today’s travel patterns most viruses spread quickly to every continent. The definition needs to be revamped.”
PROSPER stands for: the Protocol for a Standardized Information Infrastructure for Pandemic and Emerging Infectious Disease Response. The PROSPER model is presented in an article in PLoS ONE.
Text: Anika Agebjörn
Last updated: 2012-12-10