Hi, Phillippo,
In the multinma package, only the mean difference could be calculated for the continuous outcome. I wonder whether it is possible to add other effect size parameters, such as standard mean difference. This is helpful when an outcome is measured with differential scales. For example, investigators might use 100-VAS or 10-NRS or other Likert scales to measure pain intensity.
It is the same for the binary outcomes. The calculations for relative ratio (RR) might be useful for clinicians. I think adding a function in the relativeeffects command might solve the problem?
Regards,
Hui
Hi, Phillippo, In the multinma package, only the mean difference could be calculated for the continuous outcome. I wonder whether it is possible to add other effect size parameters, such as standard mean difference. This is helpful when an outcome is measured with differential scales. For example, investigators might use 100-VAS or 10-NRS or other Likert scales to measure pain intensity. It is the same for the binary outcomes. The calculations for relative ratio (RR) might be useful for clinicians. I think adding a function in the relativeeffects command might solve the problem? Regards, Hui