Question

Process for determining causality (sample size, bias, confounders)? Methods for controlling confounders (and when to use them)?

Answer 1

PROCESS OF DETERMINING CASUALTY:-
(A) Sample size:- this is calculated based on Z-score.
This is a constant value needed for this equation.

Sample size= (Z-score)^2 * StdDev*(1-StdDev) / (margin of error)^2

(B) Bias:- it is a systematic error in the design or data collection that results in a mistaken estimation of the true effect of the exposure and the outcome.
E.g. -selection bias
        - information bias

(C) Confounding:- it is a situation in which association between an exposure and outcome is impaired by the presence of another variable.
This variable called confounder.
A confounder fulfill below conditions:-
        -It is a risk factor for the disease
        -It is not in the causal pathway between exposure and disease.
        -Confounding masks the true effect of a risk factor on a disease due to the presence of another variable.

METHODS OF CONTROLLING CONFOUNDERS:-
-Restriction
-Stratification
-Matching
-Regression
-Randomization