I am using DLNM to find the association of temperature with cardiovascular diseases. I know we can use
cbind(predslgam$allRRlow, predslgam$allRRhigh)["-1",]
for calculating the risk ratio and confidence interval, but how to calculate the P-value and standard error?
Best,
Hasan
From the CI. See any standard standard statistics book on CI and it's relationship with p-values. David Booth
Hi,
RR value is considered clinically significant if it is less than 0.50 when the risk is at least halved or more than 2.00; that is doubled.
To calculate p-value, from the 95% C.I calculate Standard Error and then Z value, subsequently p-value.
Go through this article in BMJ
How to obtain the P-value from a confidence interval
BMJ 2011; 343 doi: https://doi.org/10.1136/bmj.d2304 (Published 08 August 2011)Cite this as: BMJ 2011;343:d2304
- Badges
- Science topic