Reader Comments
Post a new comment on this article
Post Your Discussion Comment
Please follow our guidelines for comments and review our competing interests policy. Comments that do not conform to our guidelines will be promptly removed and the user account disabled. The following must be avoided:
- Remarks that could be interpreted as allegations of misconduct
- Unsupported assertions or statements
- Inflammatory or insulting language
Thank You!
Thank you for taking the time to flag this posting; we review flagged postings on a regular basis.
closeReferee comments: Referee 2
Posted by PLOS_ONE_Group on 18 Mar 2008 at 10:34 GMT
Referee 2's review:
The authors have done quite some work in revising their paper. However, of the four points I have raised only one has been addressed sufficiently.
(1) No observational evidence to support the chosen parameter values. The authors have added more references but still they do not mention how their parameter values compare to actual observations. The authors' response reveals they use an inconsistent set of parameters for generation time.
(2) No comparison of simulation results with observations. The authors have simply inserted a sentence in the conclusion stating that their results are in any case similar to observations --- even though there is no comparison of their simulation results with observation in the entire manuscript! Depending on one's view, the authors' reply is surprising or shocking. How can a reader judge that the model is actually simulating an epidemic that resembles an influenza pandemic, and this sentence in the conclusion is correct?
(3) Indicating the uncertainty in simulation outcomes. This point has been addressed.
(4) Indicate whether the code is available, whether the model is transportable to other countries. The authors state they provide a link to the code. But there is no such link in the manuscript! The authors reemphasize the computational efficiency, which is completely unrelated to the comment.
Although I am overall sympathetic to this manuscript, its major weakness is that it doesn't present any evidence at all that the simulation results represent an influenza pandemic. I believe that leaving out this evidence is a serious omission that could have been easily repaired; evidently the authors have decided not to do so.
**********
N.B. These are the comments made by the referee when reviewing an earlier version of this paper. Prior to publication the manuscript has been revised in light of these comments and to address other editorial requirements.
RE: Referee comments: Referee 2
Marta_Ciofi replied to PLOS_ONE_Group on 18 Mar 2008 at 17:32 GMT
1. Though there are not many available sources reporting data on latent and infectious periods, we adopted the parameters reported in Ferguson et al., 2005 and Ferguson et al., 2006, for individual based models. As suggested by the reviewer, we now compared the results of our simulations with observed data (see point #2). As regards the parametrization of SEIR models, as reported in other published papers (see Ferguson et al., 2006), we kept fixed the latent period as in the IBM (1.5 days), and chose the infectious period (constant over time) giving the same real time growth rate as observed in the IBM. Detailed information on this procedure are reported in SI text, Section 2.3.
2. Simulation results have been compared with historical data. Since no incidence or mortality weekly data from past pandemics are available for Italy, we compared model results with weekly deaths reported in UK during the 1918-1919 pandemic (Ferguson et al, 2006). Results are reported in Section 2.4. of Supplementary information; they show a good fit of the moderate and severe scenarios, with the first and second wave of the pandemic, respectively. Moreover, we compared cumulative ARs by age obtained in a baseline simulation (R0=1.6) with those reported for the 1918-1919 US epidemic (Glezen, 1996). This comparison shows a good agreement of ARs by age-group in individuals up to 40 years of age, further supporting the choice of incubation and infectious periods used in the model.
3. According to the referee, this issue was properly addressed.
4. The software implementing the algorithm described in this work is available at the URL http://mpa.fbk.eu/static/.... It is available as a gzipped archive containing the source code in C. It contains a README file with information for compiling it (under Unix/Linux machines) and a short description of program parameters, input and output files. It requires GNU Scientific Library (http://www.gnu.org/softwa...) and Sqlite (http://www.sqlite.org/).