Metaprop - different output when using Metaprop and forest(metaprop)

Hi,
I am performing MA of proportions and my metaprop outputs (for heterogeneity & summary estimate& weighting) are different in the print(metaprop) results and the forest(metaprop) results despite using the same code. Does anyone know why? both codes below.

With this code, the forestplot also will not print the I2 and tau etc but if I just do: forest(meta_Bylog) is does print I2, tau, heterogeneity - any ideas why?

Thanks very much

meta_Bylog <- metaprop(
n,
N,
studlab = paste(Study),
data = By1log_data,
method= "Inverse", #or could be "GLMM"#
sm = "PLOGIT", #or "PLO" or "PRAW" or "PFT" (for double arcsine transformation-will not need 0.5 increment)#
incr = 0.5,
allincr = FALSE,
method.ci = "AC",
level = 0.95,
level.comb = 0.95,
comb.fixed = FALSE,
comb.random = TRUE,
overall = TRUE,
w.random = TRUE,
method.tau = "REML"
method.tau.ci = "QP",
overall.hetstat = TRUE,
method.bias = "linreg",
backtransf = TRUE,
pscale = 100,
title = "Reproducibility",
complab = "Compare",
outclab = "Compare2",
keepdata = gs("keepdata"),
warn = gs("warn"),
)
print(By1log_meta)

versus the forest:
forest.meta(
meta_Bylog,
sortvar = Year,
studlab = TRUE,
comb.random = TRUE,
overall = TRUE,
text.random = "Overall Proportion",
col.random = "blue",lwd = 1.5,
study.results = TRUE,
xlab = "Percentage of Patients",
smlab= "Reproducibility\n ",
xlim =c(10,100),
align = "left",
leftcols = c("Study", "Year", "event", "n"),
leftlabs = c("Study", "Year", "n", "N"),
rightlabs = c("% reproducibility", "95% CI", "Weight"),
just = "center",
print.tau2.ci = TRUE,
print.I2 = TRUE,
print.I2.ci = TRUE,
print.tau2 = TRUE,
print.pval.Q = TRUE,
col.square = "grey",
col.square.lines = "grey",
col.diamond = "maroon",
col.diamond.lines = "black",
colgap = unit(6, "mm"),
digits = 1,
digits.se = 1,
digits.weight = 1,
digits.tau2 = 4,
digits.pval = 4,
digits.mean = 1,

)

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.