[R-meta] Asymmetric CI for certain estimates in a meta-analysis

Samantha Marinello @m@r|n23 @end|ng |rom u|c@edu
Sun Jun 20 22:24:42 CEST 2021

Hi everyone,

I have a question related to using estimated effects with asymmetric
confidence intervals in a meta-analysis.

In our study, we have extracted 26 estimates and confidence intervals. For
a few of the extractions, the confidence intervals are asymmetric for for
the following reasons: (1) the CIs are bootstrapped and (2) the outcome is
a ratio of ratios. My understanding is that both fixed and random effects
models assume the estimates follow a standard normal distribution and that
the "meta" package in R uses the extracted CIs to estimate SEs and then
re-calculates symmetric confidence intervals.

Is there anything we can do to account for the fact that some of our
estimates have asymmetric confidence intervals or do we need to report this
issue as a limitation of our study?

Thank you in advance for your help!

Kind regards,

*Samantha Marinello*, MS
PhD Candidate

Division of Health Policy & Administration

School of Public Health

University of Illinois at Chicago

1603 W. Taylor Street

Chicago, IL 60612

	[[alternative HTML version deleted]]

More information about the R-sig-meta-analysis mailing list