Interesting paper; perhaps the most common motivation for moving beyond the exponential as a transition distribution in disease models is that doesn't enforce a certain minimum waiting time, but here it seems the exponential doesn't give enough quick transitions!
One question: I couldn't find any discussion of the role of symptomatic infections and ensuing treatment, so I was wondering if you would care to comment on any possible confounding effect this would have on the relationship between age and distribution of infection durations? At face value I would expect that the sharper curves (lower Weibull shape parameter) in the 0-5 and 5-9 age groups relative to older children and adults might reflect the higher rate of symptomatic infections in these age groups (at PfPR ~ 49% transmission) and hence treated infections (assuming treatment is non-zero)?
Effects of symptomatic cases and/or treatment?
12 August 2015
Interesting paper; perhaps the most common motivation for moving beyond the exponential as a transition distribution in disease models is that doesn't enforce a certain minimum waiting time, but here it seems the exponential doesn't give enough quick transitions!One question: I couldn't find any discussion of the role of symptomatic infections and ensuing treatment, so I was wondering if you would care to comment on any possible confounding effect this would have on the relationship between age and distribution of infection durations?
At face value I would expect that the sharper curves (lower Weibull shape parameter) in the 0-5 and 5-9 age groups relative to older children and adults might reflect the higher rate of symptomatic infections in these age groups (at PfPR ~ 49% transmission) and hence treated infections (assuming treatment is non-zero)?
Competing interests
None.