I'm doing a meta-analysis on prognostic studies. The studies I have included present the following values: AUC, 95% CI, a cut-off value and associated sensitivity and specificity.

Also, is it even possible to compare these if some studies report the values regarding predicting an event and other studies report the values on predicting event freedom?

I would like to reach the optimal cut-off value of the measurement to predict an event. Would it be a possibility to just do a weighted mean?

More Anne Nielsen's questions See All
Similar questions and discussions