library(stats)
power.prop.test(p1 = .45, p2 = .25, power = .80, sig.level = 0.05, alternative = c("two.sided"), strict = FALSE)

 Two-sample comparison of proportions power calculation 

          n = 88.0928
         p1 = 0.45
         p2 = 0.25
  sig.level = 0.05
      power = 0.8
alternative = two.sided

Why does my sample size calculation differ from the one in this NEJM-published paper?

Why prop.power.test function calculate the power without the Yates continuity correction? but prop.test implements this correction?

I can achieve the same results as this paper by using this online calculator. What could be the reason? Can I achieve this result in r?
Thank you very much for helping!!!!