Dear all,
I am trying to figure out how to convert odds ratio per unit decrease into beta value per unit increase.
E.g. OR = 2.3 per 1 mg/dL decrease -> Beta = x per 1 mg/dL increase.
I am aware that Beta = ln(OR). But to switch unit direction do I use its inverse, so Beta = ln(1/OR)?
Appreciate any help!
As in all estimates of treatment effect, odds ratios or relative risks reported in meta-analysis should be accompanied by confidence intervals.
beta is not a variable. it is a coefficient in a model which is estimated from the data.
as log odds = alpha+ beta X, the change in log odds going from X = 2 to X=1 is (alpha + beta x2 ) - (alpha+ beta x 1) = beta (it is a reduction in log odds) and the change in log odds going from X = 1 to X=2 is also beta but it is an increase in log odds.
You try an an alternative approach for meta-analysis of paired binomial outcomes that does not rely on the assumption of known within-study variances. Using SAS, Stata, and R, most of the results from the alternative approach are quite different from those produced by the conventional meta-analysis approaches, which use study-level summaries that reduce the data to log-odds-ratio and its estimated SE.
The alternative approach is straightforward and can be easily implemented in two commonly used commercial software packages, SAS and Stata, and one free software package, R.
Additional care is needed for meta-analyses of rare events, when some studies report zero events.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis