When conducting a systematic review and meta-analysis on the topic of congenital heart disease, specifically focusing on studies from Africa. However, You encountered a challenge with all of the downloaded articles as they do not report odds ratios but instead provide descriptions of the types or prevalence of the condition. Consequently, confidence intervals (CI) and odds ratios (OR) are not available for analysis. Despite this limitation, and you has instructed me to proceed with writing a meta-analysis using Stata. In such a situation, what alternative approaches can be consider? Additionally, which variables can be use for the analysis? Should this methods be used to generate the OR and CI From the expose and unexpose group or the cases and control ?
When conducting a systematic review and meta-analysis on the topic of congenital heart disease, specifically focusing on studies from Africa. However, You encountered a challenge with all of the downloaded articles as they do not report odds ratios but instead provide descriptions of the types or prevalence of the condition. Consequently, confidence intervals (CI) and odds ratios (OR) are not available for analysis. Despite this limitation, and you has instructed me to proceed with writing a meta-analysis using Stata. In such a situation, what alternative approaches can be consider? Additionally, which variables can be use for the analysis? Should this methods be used to generate the OR and CI From the expose and unexpose group or the cases and control ?
is the last chi-square test for the difference between crude OR and OR(MH) or for the significance of OR(MH)?
I wondered the same thing