I am trying to solve a problem about a case-control study, where I need to calculate the odds ratio the association of X and Y, stratified by two age groups. First, I calculated the crude OR for each group, which is ok and gives me similar results (1.30-1.45). Nevertheless, when calculating the OR ADJUSTED for age, I performed a weighted analysis by Mantel Haenszel method and it gives me a very high value (18), which does not make sense to me. I am not sure if I used the wrong equation or did not calculate correctly. Could this mean that this is not a confounder, but an effect modifier? , or could it be that it is not homogenous and therefore I cannot perform Mantel Haenszel? Besides, Which would be the statistical test to perform?

Also, on another note, how could I use odds ratio to assess the confounding effects of age?

Thank you very much :)

  • $\begingroup$ Welcome to Cross Validated! Woudl you be able to share your data and show your calculations? $\endgroup$ Mar 26, 2022 at 22:04
  • 1
    $\begingroup$ Please clarify your specific problem or provide additional details to highlight exactly what you need. As it's currently written, it's hard to tell exactly what you're asking. $\endgroup$
    – Community Bot
    Mar 26, 2022 at 23:35


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.