On 11/21/2011 5:34 AM, wanderley bernardo wrote:
> How can we give an evidence based answer about: Why 20% loss of
> follow-up in RCT is too much?
It's difficult to summarize here because of a few formulas, but I show a
mathematical derivation of the impact of loss to follow-up on my new
website:
--> http://www.pmean.com/11/dropouts.html
The answer is indeed that it depends, but I show what "it" depends on. I
also should note in contrast to some of the other responders that a 1%
dropout rate would not be a problem except in some rather unrealistic
settings.
Steve Simon, [log in to unmask], Standard Disclaimer.
Sign up for the Monthly Mean, the newsletter that
dares to call itself average at www.pmean.com/news
|