I am conducting a meta-analysis and have one study which reports relative risks, and several studies reporting odds ratios. I would like to convert the relative risks to odds ratios and obtain corresponding confidence intervals and standard errors.

I have data on the prevalence of the outcome in the control group, the relative risk and 95% CIs.

I came across this formula for deriving the odds ratio from the relative risk (from Convert hazards ratio to odds ratio), based on Grant et al, BMJ 2014:

OR = ((1 - p) * RR) / (1 - RR * p)

where RR is the relative risk, OR is the odds ratio, and p is the control event rate.

To calculate the corresponding confidence intervals and standard error, would the below be correct:

OR_lowCI = ((1 - p) * RR_lowCI) / (1 - RR_lowCI * p)

OR_upperCI = ((1 - p) * RR_upperCI) / (1 - RR_upperCI * p)

log_OR_SE = (ln(OR_upperCI) - ln(OR_lowCI))/3.92

Many thanks in advance for any advice.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.