Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
27 Cards in this Set
- Front
- Back
is confounding a non causal relationship
|
yes!!!!!!! but it IS causally related to the outcome just not in the cause pathway of exposure to disease outcome-- it's own pathwya
|
|
what three conditions must be met in order for something to be confounding
|
associated with the epxosure of interest (causally or not)
causally related to the outcome And is NOT part of the exposure outcome causal pathway |
|
if you control for your factor and the OR goes down, then
|
your variable was confounding
|
|
what are some ways that you can control for confounding
|
conduct RCT
matching statification adjustment |
|
what is another name for interaction
|
effect modification
|
|
define effect modification
|
refers to the situation where the effect of the risk factor differs depending on the presence or absence of a third factor.
|
|
how can you determine if something is an effect modifier?
|
if when you ask if the assocation is found in all groups? If your answer is it depends, then you have an effect modifier.
|
|
what is adjustment
|
use of stats to control for confounding
|
|
what are two ways to stratify by stats
|
adjustment based on stratification
adjusment based on multiple regression |
|
what are two types of stratifciation adjustments
|
direct method
mantel haenszel |
|
what are four ways to use multiple regression
|
mutliple linear
logistic cox poisson |
|
how do you use antel haenszel
|
adjust OR
weighted average of OR from strata formed by suspected confounder |
|
T or false. regression coefficient epxresses the strength of the association between predictor and slope
|
true!
|
|
T or F. correlation coefficient expresses the strength of the association (slope)
|
FALSE only tells you how good of a line is it
|
|
what is the letter for regression coeffiecient
|
B1
|
|
Y =
|
B0 +B1(X) + B2X + B3X
|
|
put it in words-- if you talk about beta for age
|
controlling for all other variables, for every 5 years incrase in age, leukocyte count decreases
|
|
T or false. regression coefficient epxresses the strength of the association between predictor and slope
|
true!
|
|
T or F. correlation coefficient expresses the strength of the association (slope)
|
FALSE only tells you how good of a line is it
|
|
what is the letter for regression coeffiecient
|
B1
|
|
Y =
|
B0 +B1(X) + B2X + B3X
|
|
put it in words-- if you talk about beta for age
|
controlling for all other variables, for every 5 years incrase in age, leukocyte count decreases
|
|
define adjusted rate
|
weighted average of age specific rates in the study population using as weight the number of individuals in an arbitrary standard population
|
|
what is logistic regression
|
multiple regression where the dependent variable is the log(Oddds)
|
|
In logistic regression what is the ajusted OR =
|
e^beta1
|
|
what is required about the variable in logistic regression
|
must be binary
|
|
what is Cox proportional hazards regression
|
log RR
multiple regression applied in the context of survival analysis. |