Control of confounding

9 important questions on Control of confounding

What's the difference between prospective and retrospective cohort study?

Prospective cohort identified in the present and followed in the future, retrospective they're assembled in the past and followed until present. Prospective study can collect wider range of risk factors and demographic data, but they're much more expensive and time demanding in comparison to retrospective study. On the other hand retrospective study are more limited in terms of data retrieval and range of possible risk factors that can be studied.

Advantages of cohort study?

-Can assess incidence (absolute risk)
-Follows clinicians logic: if exposed gets sick?
-Can assess relationship between exposure and many diseases
-No bias in exposure, as outcome(disease) is not known yet (from exposure to disease)

Disadvantages of cohort studies?

-Susceptible to confounding and other biases: people exposed different from non-exposed ones, making comparison difficult ( information bias, selective dropout, changes in methods exposure assessment)   
-Low incidence or long exposure period make cohort study difficult: better suited for prevalent diseases and rare exposure
-Preliminary study results necessary to get relevant determinants of outcome:what if they're not available?
-Exposure in the past might affect exposure of study population during classification (selection bias)
-Prospective more time and money consuming than retrospective, but have larger scope
-Incidence(risk), AR, RR can be estimated
  • Higher grades + faster learning
  • Never study anything twice
  • 100% sure, 100% understanding
Discover Study Smart

Characteristics of case-control study?

-Longitudinal study, from Disease to Exposure
- 2 groups: cases(disease), control
-Cases(preferably new) selected on base of diagnosis
-Controls selected to make the groups comparison possible
-Exposure occurs before sample becomes cases and controls
-retrospective study seeking for potential risk factors(exposure)
-Calculation of OR (estimation of RR based on that)

Advantages of case-control study?

-faster and cheaper than cohort
-smaller study population give often more complete info
-Suited for rare diseases, not rare exposures

Disadvantages of case-control study?

-More risk of bias(information, selection biases and confounders)
-Only estimation of RR (by means of OR) --> no info on incidence!

Bias(systemic error) in associations: definition?

Process at any stage of inference tending to produce results that depart systematically from the true values

Bias(systemic error) in associations: types?

-Selection bias
-Information bias
-Confounding
-Effect modification

Information Bias/Measurement Bias

Occurs when the methods of measurements are dissimilar among group of patients.
For instance in cohort study is present when members of the cohort are not assessed similarly for outcome (differential misclassification and non-differential misclassification).
In case-control study, instead it might occur when cases are more likely to remember whether they were exposed (=recall bias).

The question on the page originate from the summary of the following study material:

  • A unique study and practice tool
  • Never study anything twice again
  • Get the grades you hope for
  • 100% sure, 100% understanding
Remember faster, study better. Scientifically proven.
Trustpilot Logo