Study Designs

Experimental & Observational, (Analytical, Descriptive)Study Designs

https://www.cebm.net/wp-content/uploads/2015/09/observational-studies-and-more-Carl-Heneghan.pdf
https://www.slideshare.net/ParthaSarkar7/observational-analytical-study?from_action=save
https://www.slideshare.net/ParthaSarkar7/observational-analytical-study?from_action=save
https://www.slideshare.net/ParthaSarkar7/observational-analytical-study?from_action=save
  • COHORT STUDY
    • compare the exposure (exposed vs unexposed)
    • to find probability of getting disease (or disease incidence).
    • Risk Ratio of ‘disease’ incidence.
  • CASE-CONTROL STUDY
    • compare the disease (diseased vs healthy)
    • to find risk factor frequency.
    • Odd Ratio of ‘exposure/risk factor‘.
  • NESTED CASE CONTROL STUDY (NCC)
    • 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.
https://www.slideshare.net/ParthaSarkar7/observational-analytical-study?from_action=save
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).
https://www.stomponstep1.com/odds-ratio-relative-risk-calculation-definition-probability-odds/
https://www.stomponstep1.com/odds-ratio-relative-risk-calculation-definition-probability-odds/
https://www.slideshare.net/terryshaneyfelt7/what-does-an-odds-ratio-or-relative-risk-mean?from_action=save
https://www.slideshare.net/terryshaneyfelt7/what-does-an-odds-ratio-or-relative-risk-mean?from_action=save
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
https://www.pinterest.com/pin/720435271610625102/
https://www.slideshare.net/ParthaSarkar7/observational-analytical-study?from_action=save
https://www.slideshare.net/ParthaSarkar7/observational-analytical-study?from_action=save
https://www.slideshare.net/ParthaSarkar7/observational-analytical-study?from_action=save
https://www.slideshare.net/ParthaSarkar7/observational-analytical-study?from_action=save
https://www.slideshare.net/ParthaSarkar7/observational-analytical-study?from_action=save
https://www.cebm.net/wp-content/uploads/2015/09/observational-studies-and-more-Carl-Heneghan.pdf

REFERENCES