1
$\begingroup$

You conduct a case-control study of elevated cholesterol and myocardial infarction (MI). Of 20 MI cases, 10 had elevated cholesterol. Of 30 healthy controls, 10 had elevated cholesterol. These results give an odds ratio (OR) of 2.0, with a 95% confidence interval (CI) $ = [0.6-6.4]$.

Interpret the CI?

$\endgroup$
  • 4
    $\begingroup$ This looks like standard bookwork (especially given the start 'You conduct...' - I don't!). If this is for some subject, or even just for your personal study, please add the self-study tag and read its tag wiki info $\endgroup$ – Glen_b -Reinstate Monica Oct 20 '13 at 22:01
1
$\begingroup$

The OR in your sample is 2. You don't know the exact odds ratio in the corresponding population, but at least you can be 95% confident that it is somewhere between 0.6 and 6.4. Since the interval contains the value 1 (no relation between MI and elevated cholesterol), you cannot claim that there is truely such relation.

$\endgroup$
  • $\begingroup$ observed OR is 2 statistically significant different than the null value(using 2 sided type 1 error rate of 0.05) $\endgroup$ – Malai Oct 20 '13 at 20:02
  • 1
    $\begingroup$ In fact, the results are not statistically significant. The P value with Fisher's exact test is 0.26, and 0.38 with a chi-square test (with Yate's correction). $\endgroup$ – Harvey Motulsky Nov 20 '13 at 1:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.