Hello Allstat-ers
I’m undertaking a published data meta-analysis looking at continuous data. We want to look at the mean change between baseline and 1 year (as opposed to the data at 1 year), but we only have the mean and standard deviation for each group at baseline and 1 year.
I can get an estimate of the mean change in each group by calculating mean (post) – mean (baseline).
But how do I estimate the standard deviation for the mean change?
I’ve found a formula for estimating the variance of the difference, but I am not entirely convinced of its validity, and what you use for the arbitrary value r.
variance (Diff) = variance (baseline) + variance (post) - 2r (sqrt (variance(baseline) x variance(post)))
Ideally, obtaining the individual patient data so we can calculate these figures correctly would be the best way forward, but this is unlikely to be possible. Any help gratefully received.
Regards
Roberta
---------------------------------
Yahoo! Messenger NEW - crystal clear PC to PC calling worldwide with voicemail
|