Calculating sample size based on incidence rate?

#1
I would like to calculate the necessary sample size for a retrospective cohort analysis. I will be comparing male with female and my outcome will be extra-articular manifestations (EAMs) in patients with spondyloarthritis (more specifically, axial spondyloarthritis and psoriatic arthritis).

This is the paper I would like to extract data from: Incidence of extra-articular manifestations in ankylosing spondylitis, psoriatic arthritis and undifferentiated spondyloarthritis: results from a national register-based cohort study | Rheumatology | Oxford Academic (oup.com)

They provide incidence rates for the different subtypes of EAMS (acute uveitis anterior, psoriasis and inflammatory bowel disease). Here is a small extract of the provided table:

data_paper.png

I figured I would use something like incidence in males / total males and take this as the "incidence" of the outcome in the sample size calculation. In the example above, for AU, this would be 7.8% (435/5571) for males and 5.3% (146/2746) for females. Using enrollment ratio 2:1, alpha 0.05 and power 80%, then according to this calculator, I would require 3372 patients. However, I feel as if this is not the correct way to do this. Does anyone have advice for me?
 

Attachments