I'm conducting a quantitative synthesis of several studies reporting ORs and 95% CIs for a health outcome. Some studies report summary 2x2 table data, while others only report adjusted or unadjusted effect estimates without details about the numbers of cases or participants in each exposure group. In cases where studies don't report 2x2 summary data, is anyone aware of ways to recalculate ORs with respect to a new reference group? I would like to maximise the number of studies I can include in my synthesis but also recognise the limitations of the available data.
I haven't tried doing this myself, though I am aware that I can recalculate ORs using a new reference category from the one reported. It's the standard errors/95% CIs that I am most unsure about!
Thank you so much! Yaning