Hey there, I‘m currently working on a surivival meta-analysis and I use SPSS for my statistics. The problem is that I have a lot of data regarding mean OS / PFS, but no belonging standard deviations… does anyone have an idea how to solve this problem? I mean I can‘t make a meta-analysis with only the „mean“ data… thank you already very much!
Confidence intervals for mean survival time are asymmetric so cannot come from standard deviations of estimates of the mean. The upper confidence limit will be much further from the mean that the lower limit. This is due partially to skewness of the distribution of event times.