PUBH 3131 Lecture Notes - Lecture 5: Confounding, Bar Chart, Line Graph
Document Summary
Differ in distribution of variables in diseased/non-diseased. Associated factors may not be causally but not independent. Might be no formal comparison group if: Can"t tell if it was due to chance (fail to reject) Narrow down to measurable outcome and measurable exposure. If health outcome differs between exposed and non-exposed, exposure and outcome are associated. Hold all but variable of interest constant. Confounding: comparison groups differ by a factor that itself changes the outcome. Can"t know if results due to exposure or confounder. Artifactual results: number due to outside factors. One with true association and one we aren"t interested in. Distortion of a measure of the effect of an exposure on an outcome due to the association of the exposure with other factors that influence the occurrence of the outcome. Crude rates=non adjustment made, observed over entire population. Specific rates=rates given a particular value of some other variable or combination of variables.