1
$\begingroup$

I am working in R tryign to estimate the population attributable fraction (PAF) for a binary outcome and 2 exposure variables that I had to coerce to become binary. I have used the "epi.2by2" function within epiR, but that gives me the simple calculation that I could have done easily by hand. Is there a way to adjust the PAF for an external fixed variable? Similar to how odds ratios can be adjusted. This is for a cohort count dataset, which is hard to find worked examples for (usually Case-Control).

Improve this question
$\endgroup$

1 Answer 1

Reset to default
2
$\begingroup$

"[R package] attribrisk Schenck, Atkinson, Crowson, and Therneau (2014) estimates PAFs in matched and unmatched case-control designs."

. . .

"The new R package graphPA described here also estimates PAFs for cross sectional, case-control and cohort study designs under random samples."

source: https://cran.r-project.org/web/packages/graphPAF/vignettes/graphPAF_vignette.pdf

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.