I am following a cohort of patients over time from which I am planning to sub-sample a sub-group of patients to pilot a new intervention. The plan is to compare the effectiveness of the new intervention on improving specific outcomes for a sub-sample compared to those did not receive the intervention. the original study is an inception prospective cohort with several followup assessment time points (3 evaluations time points).

Which design is better to use?

  1. Nested Case-Control (the intervention group [participants who have completed the intervention] will be compared to control group [those who didn't receive an intervention] and maybe a third group of normal people from the community) in this case the 2 (or 3) study groups will be matched on age and comorbidities.

  2. Case-Cohort design (my understanding for this is that a sub-sample will be randomly selected at the baseline for an intervention and then compared to the rest of the cohort). Both are new to me and not sure if any of these is a superior design over the other one and better to use, and what would be the factors to consider?

  • 3
    $\begingroup$ Just to make sure I understand your descriptions correctly, the only difference is that in the first case we control for age and comorbities as part of the study design by sampling in a particular way, while in the second case we sample randomly and then control for age and comorbidities statistically (as I assume you would)-- correct? $\endgroup$ – Jake Westfall Nov 1 '13 at 22:49

Your Answer

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

Browse other questions tagged or ask your own question.