Modelling NPIs

A new study out of Australia gives us some insite into why non-pharmaceutical interventions (NPIs) might be so important during a pandemic. The study doesn't have a summary for laymen so I've pulled out one paragraph and bolded some of the findings.


"For epidemics with an unmitigated R0 of 1.5, case isolation, school closure or community contact reduction made significant reductions in the final attack rate, reducing it from 33% to 6%, 13% and 16% respectively. School closure combined with any of the other interventions reduced the cumulative attack rate to below 10%, which may be deemed to be the threshold below which an epidemic does not occur. School closure combined with case isolation reduced the final attack rate to 8%. For epidemics with an R0 of 2.5, only the combination of all the modelled intervention measures appeared capable of controlling the epidemic, reducing the final attack rate from 65% to 3%. School closure combined with case isolation more than halved the final attack rate (to 30%). To achieve the large reductions in attack rates for R0 values of 2.0 and 2.5, combinations of interventions needed to operate for unfeasibly long periods of time (greater than 5 months)."

Basically, under various scenarios, there's a strong possibility that combining more than one NPI could drastically reduce the number of people that fall ill. One of the challenges would be how long we need to maintain the interventions. As the section above points out, 5 months or longer might be necessary but not feasible.

To read the full article, visit:

http://www.plosone.org/article/info:doi/10.1371/journal.pone.0004005