1

I have calculated the relative risk & number needed to treat using the log-binomial regression model. NNT is calculated as the reciprocal of the marginal difference.

Does it make sense to report different p-values for the RR and the NNT?

kjetil b halvorsen
  • 63,378
  • 26
  • 142
  • 467
Megan Moreton
  • 95
  • 5
  • https://www.cebm.net/2014/03/number-needed-to-treat-nnt/. It does not seem like NNT is equivalent to RR, so why do you expect the p-values to be equal? – kjetil b halvorsen May 12 '20 at 15:04

0 Answers0