Hi All,
I have a question regarding the 95% CI derivation for the relative risk RR.
I found the following info on the web but would like to know where it came
from; any help with the derivation or a reference showing how to obtain this
would be most appreciated. I would like to understand it a little better
rather than blindly using the formula.
I have proportions p1 and p2 from samples n1 and n2 the relative risk is
defined as:
RR = p1/p2
and the 95% CI is [RR*exp( -Z/S ), RR*exp( Z/S )]
where Z = Z(0.025) fron N(0,1) and S^2 = variance of ln(RR)
so S^2 = V(ln(RR)) = (1-p1)/p1*n1 + (1-p2)/p2*n2 and S=sqrt( V( ln(RR ) ) )
How do we get V(ln(RR)) as defined above?
Many Thanks,
Richard.
_________________________________________________________________
Is your PC infected? Get a FREE online computer virus scan from McAfeeŽ
Security. http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963
|