I am comparing the Standardised Mortality Ratios across 10 groups, and have used the ems package to calculate the group-level SMR and confidence intervals .
I would like to do an across group (all 10 groups) comparison to derive a P value, as well as some specific pairwise comparisons. From a theoretical perspective I know above can be done using the provided means and confidence intervals, but I can't figure out if there's a function to do it in R, and wonder if someone could help.
I've included the dput formatted output from ems here (I'm not sure why it doesn't come up as a neat table?)
structure(list(Variables = c("Overall", "Group",
NA, NA, NA, NA), Levels = structure(c(NA, 1L, 2L, 3L, 4L, 5L), levels = c("1",
"10", "2", "3", "4", "5", "6", "7", "8", "9"), class = "factor"),
N = c(455240, 44167, 62167, 40959, 35790, 49904), Observed = c(31182,
3226, 4057, 3029, 2397, 3413), Expected = c(33633.08248,
3637.17687, 4276.20381, 3261.36463, 2722.91475, 3684.30542
), SMR = c(0.92712, 0.88695, 0.94874, 0.92875, 0.88031, 0.92636
), lower.Cl = c(0.91874, 0.86175, 0.92536, 0.90222, 0.85105,
0.90102), upper.Cl = c(0.93559, 0.91289, 0.9727, 0.95607,
0.91057, 0.95241)), row.names = c(NA, 6L), class = "data.frame")
