I am trying to calculate the population attributable fraction (PAF) for the impact of an exposure on suicide risk across age strata using odds ratios from my case-control study. I am using conditional logistic regression to estimate odds ratios. In such scenario (adjusted odds ratios) the Miettinen formula should be used to obtain PAF:
(RR-1)/RR multiplied by (number of exposed cases/total number of cases)
In my estimation, RR (relative risk) is substituted with odds ratios. and the proportion of exposed cases refer to those in the subpopulation.
I can manually calculate PAF, but how can I obtain the 95% confidence interval?
I used PUNAFCC but got zero results.
I entered these syntaxes
Clogit suicide exposure gender if agegroup==1, group(matchedid) or
PUNAFCC
Unfortunately, I got "zero" for all estimates of PAF. I would much appreciate any advice on how to obtain the 95% confidence interval for PAF in STATA given the above described scenario.
Thank you
Danah
(RR-1)/RR multiplied by (number of exposed cases/total number of cases)
In my estimation, RR (relative risk) is substituted with odds ratios. and the proportion of exposed cases refer to those in the subpopulation.
I can manually calculate PAF, but how can I obtain the 95% confidence interval?
I used PUNAFCC but got zero results.
I entered these syntaxes
Clogit suicide exposure gender if agegroup==1, group(matchedid) or
PUNAFCC
Thank you
Danah
