Pandemic flu may be well mitigated until vaccine is available

virtual community interacted the way people normally do: within households, schools and workplaces, and the community at large. All three models were set up to have attack-rate patterns similar to those of past U.S. flu pandemics. Predicting the spread of an infectious disease such pandemic influenza requires much more than simply connecting dots on a map. Instead, Halloran and colleagues rely on a tool called stochastic modeling to take into account real-world unpredictability, as well as many factors about the disease and the affected population. In constructing these models, the researchers begin with assumptions about how people interact and how the virus spreads. They also introduce and evaluate the effectiveness of various intervention strategies. The study assessed the effectiveness of two broad categories of intervention: medical (the use of surveillance to identify cases and to use antiviral agents to treat flu patients and prevent the disease among their close contacts) and non-pharmaceutical (social distancing, such as school closures, voluntary quarantine, and travel restrictions). The researchers tested combinations of targeted antiviral treatment and prophylaxis (prevention) and general social-distancing strategies with five intervention scenarios of different levels of stringency and disease transmissibility. The researchers calculated the transmissibility of the virus by the average number of secondary cases infected by each primary case at the beginning of the epidemic, and by the speed at which such cases arose. Such parameters help predict whether an epidemic will turn into a full-fledged pandemic.

In the absence of any intervention, all three computer models produced similar illness-attack rates, ranging from nearly 47 percent to nearly 60 percent of the population having symptomatic influenza. The least-stringent scenario had the following characteristics: interventions were not implemented until 1 percent of the population had developed symptomatic influenza, schools were closed, only 60 percent of clinical cases were treated with antivirals and their contacts prophylaxed, compliance with quarantine was only 30 percent and compliance with social distancing was only 60 percent. Even with these limitations, at the lower transmissibility of the flu virus, the combined intervention strategies tested by the three computer-modeling groups achieved a reduction in influenza cases of between 83 percent and 94 percent. “We ran this simulation with the assumption that the pandemic was as virulent and lethal as the 1918 pandemic,” Longini said. “Even when modeling the situation of pandemic flu, with a modest compliance range in social-distancing measures, and modest ability to identify and treat and prophylax with antivirals, the interventions were similarly