Study Designs

Experimental & Observational, (Analytical, Descriptive)Study Designs
    • compare the exposure (exposed vs unexposed)
    • to find probability of getting disease (or disease incidence).
    • Risk Ratio of ‘disease’ incidence.
    • compare the disease (diseased vs healthy)
    • to find risk factor frequency.
    • Odd Ratio of ‘exposure/risk factor‘.
    • A case-control study nested within a cohort study.
    • exposure of interest is only measured among the cases and the selected controls.
  • Because the number of cases is probably fairly small, can match multiple controls to a given case to increase the power.
  • NCC design is often used when the exposure of interest is difficult or expensive to obtain and when the outcome is rare.
    • NCC is more efficient than the full cohort design. Cases of a disease that occur in a defined cohort are identified and, for each, a specified number of matched controls is selected from among those in the cohort who have not developed the disease by the time of disease occurrence in the case.
Probability vs. Odds
  • Probability is essentially the same as percentage.
  • We tend to compare the number of occurrences of a certain outcome to the number of total events measured.
  • Probability ranges between zero and one. 
  • Odds is a ratio of the likelihood of an event happening compared to the likelihood of an event not happening.
  • Odds can be zero or any positive number (not just values between 0 and 1).
How big should the RR or OR ?
  • In RCT we are satisfied with a small increase (or decrease) in risk.
  • Cohort study: RR > 3 (for a minor adverse event)
  • Case-control: OR > 4