metamedian (version 1.2.1)

metamedian_survival: Meta-Analysis of the median survival times

Description

This function implements the Wald approximation-based approach described by McGrath et al. (2025) to meta-analyze median survival times. This approach considers that each study reports a Kaplan-Meier estimate of median survival in (each group of) each study along with confidence intervals. This approach performs an inverse-variance weighted meta-analysis of the median survival time (for one group studies) or the difference/ratio of median survival times between groups (for two-group studies). The within-study standard errors are estimated by assuming the confidence intervals around the median survival estimates are Wald-type intervals.

Usage

metamedian_survival(data, estimand, pool_studies = TRUE, ...)

Value

an object of class "rma.uni". For additional details, see rma.uni.

Arguments

data

data frame containing the study-specific summary data. For one-group studies, this data frame can contain the following columns:

med.g1	median.
med.ci.lb.g1	lower confidence interval bound around the median.
med.ci.ub.g1	upper confidence interval bound around the median.
med.ci.level.g1	level of the confidence interval

For two group studies, this data frame can also contain the following columns for the summary data of the second group: med.g2, med.ci.lb.g2, med.ci.ub.g2, and med.ci.level.g2.

estimand

character string specifying the estimand in the meta-analysis. The options are "median_g1" (median in group 1), "median_g2" (median in group 2), "difference_median" (difference of medians across groups), and "ratio_median" (ratio of medians across groups).

pool_studies

logical scalar specifying whether to meta-analyze the studies. If this argument is set to FALSE, function will not meta-analyze the studies and will return a list with components yi containing the study-specific outcome measure estimates and sei containing the study-specific within-study standard error estimates. The default is TRUE.

...

optional arguments that are passed into the rma.uni function for pooling. See documentation of rma.uni.

References

McGrath S., Kimmelman J., Ozturk O., Steele R., and Benedetti A. (2025). Meta-analysis of median survival times with inverse-variance weighting. arXiv e-prints arXiv:2503.03065.

Examples

## Median survival in the comparator group
metamedian_survival(data = dat.lung, estimand = "median_g2")

## Difference of median survival
metamedian_survival(data = dat.lung, estimand = "difference_median")

## Ratio of median survival (logarithm scale)
metamedian_survival(data = dat.lung, estimand = "ratio_median")