I am hoping to undertake my first meta-analysis, where the exposure variable it mean number of decayed, missing or filled teeth)and the outcome is BMI category (underweight, normal weight, overweight, obese). I understand that RevMan requires the exposure variable to be categorical. Is there a way of performing meta-analysis with my data? Regards, Louise.
You have to define a measure of association and its standard error (such as mean difference in number of decayed teeth between normal and overweight subjects) that is common or can be defined across studies
You have to define a measure of association and its standard error (such as mean difference in number of decayed teeth between normal and overweight subjects) that is common or can be defined across studies
Hi, Louise. You may consider the dose-response meta-analysis. We have the developed a model for it named REMR. Please see the following link of the article: Article The robust-error meta-regression method for dose-response me...
. Please feel free to contact us if there is any questions.
Meta-analysis can be performed with studies having similar design, aims, and outcomes that could be calculated for the same effect size. There are many kinds of effect sizes for continuous variable, such as Cohen's d, Hedge's g, Pearson r..etc. If included studies adopted BMI as a categorical variable using same standard, it would be better to adopt effect sizes for caregorical variable. Softwares like CMA or STATA can be used.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis