Which estimates to have as default in OUTBREAK ANALYSIS ?
In version 2.0 a new /OA (Outbreak analysis) parameter will be added to analysis.
The question is what should be shown as default (other aspects will come as further options): e.g.: . tab ill vanilla milk coffee /oa Would make a table for ill by all other variables, and then show what:
A: N total number of observations for exposed and unexposed: Ill number of persons with outcome Attack Rate + CI (percent with outcome) Odds Ratios + CI
B: N total number of observations for exposed and unexposed: Ill number of persons with outcome Attack Rate (percent with outcome) RR + CI
Please discuss default and desired options
Regards Jens Lauritsen EpiData Association
I prefer B of the two options. Add the /o option to provide a separate table of odds ratios, or replace the RR with OR (I prefer the replacement, since the user should know what the study design is). Add the /ci option to provide CI for the attack rates (not something I routinely look for). Add the /t option for chi-squared and p, Fisher exact p.
Jamie
epidata-list@lists.umanitoba.ca wrote:
In version 2.0 a new /OA (Outbreak analysis) parameter will be added to analysis.
The question is what should be shown as default (other aspects will come as further options): e.g.: . tab ill vanilla milk coffee /oa Would make a table for ill by all other variables, and then show what:
A: N total number of observations for exposed and unexposed: Ill number of persons with outcome Attack Rate + CI (percent with outcome) Odds Ratios + CI
B: N total number of observations for exposed and unexposed: Ill number of persons with outcome Attack Rate (percent with outcome) RR + CI
I favor option b. As risk (as opposed to the odds or rate) is the parameter of interest in an outbreak with a brief time-to-event period, seeing the risk is the first datum of interest for me. And I always like to see a CI with *any* estimate as it gives an indication of the precision of the estimate above and beyond giving information about significance.
Richard Herrell NIMH
epidata-list@lists.umanitoba.ca wrote: In version 2.0 a new /OA (Outbreak analysis) parameter will be added to analysis.
The question is what should be shown as default (other aspects will come as further options): e.g.: . tab ill vanilla milk coffee /oa Would make a table for ill by all other variables, and then show what:
A: N total number of observations for exposed and unexposed: Ill number of persons with outcome Attack Rate + CI (percent with outcome) Odds Ratios + CI
B: N total number of observations for exposed and unexposed: Ill number of persons with outcome Attack Rate (percent with outcome) RR + CI
Please discuss default and desired options
Regards Jens Lauritsen EpiData Association
_______________________________________________ EpiData-list mailing list EpiData-list@lists.umanitoba.ca http://lists.umanitoba.ca/mailman/listinfo/epidata-list
--------------------------------- Sponsored Link
Talk more and pay less. Vonage can save you up to $300 a year on your phone bill. Sign up now.
participants (1)
-
epidata-list＠lists.umanitoba.ca