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;
40 Cards in this Set
- Front
- Back
two critical questions in epi
|
1) internal validity- is the study believable
2) external- if so can the results be tied with the genersal public |
|
works by taking a national samploe
|
external validity
|
|
what are the 3 alternative explanations that must be eliminated during test process
|
1) confounding
2) bias 3) other things |
|
cruse and adjusted rate make up __________?
|
confounding
|
|
how does one determine whether confoundinng is present or not?
|
compare crude rate and relative risk or odds ratio
|
|
exposure? outcome?
|
e= gender race ethnicity
o= speed (what u want to know) |
|
what are some potential confounders?>
|
age, height, for speed
|
|
nuissance variables?
|
confounding variables
|
|
whiich variables are usually confounders?
|
usually risk factors for the disease
|
|
steps for contolling for confounding?
|
1) randomization
2) restriction 3) matching 4) stratification 5) multivariable analysis |
|
what is randomization?
|
must have sufficient sample size, randomization controlls for both known and unknown
--* limited *-- only experimental |
|
what is restriction?
|
restricts certain categories
|
|
controllling for confounders in the design phase GOAL
|
elemiinater or reduce variation between compared groups
|
|
a drawback of restriction
|
limits generalizability
|
|
what is def of matching when controllling for confounders?
|
distribut confoundiing factors in an identical manner
|
|
2 x 2 data table is used to calculate what of what?
|
odds ratio of total data
|
|
what are the limitations of stratification?
|
difficult to control for many variables simultaneously
|
|
what is a not how to way to control for confounding?
|
multivariate
|
|
what are teh design and analysis factors to confounding?
|
design- randomization, restriction, matching
analysis- stratification, multivariate analysis |
|
the third variable that mixes the effects between exposure and and outcome
|
confounder
|
|
CONFOUNDING-
|
1.make a comparison if crude rate and average of strata look sor a 10 % difference
|
|
this unlike confounding should be reported and described and not controlled
|
effect measure modification
|
|
relative risk
|
used in a cohort study to analyze
|
|
odds ratio?
|
used for case control studies to analyze
|
|
what are both ways to assess effect measure modification?
|
1. visually
2. statistical test--chi squared |
|
witth effect measure stratification
|
u compare the two strata
|
|
A C
B D relative risk? |
A/C & B/D
|
|
odds ratio?
|
a/d / b/c
|
|
null hypothesis
|
equal
|
|
alternative
|
not equal
|
|
if you are controlling for the confounding
|
report the adjusted measure of modification
|
|
if u want to say there is a difference in the strata?
|
report the measure of association for each strata
|
|
the mixing of effects between and exposure of association of each strata
|
effect measure modification
|
|
bias
|
systmeatic error of inference
|
|
bias toward null
|
effects are underestimated
|
|
bias away from null
|
effects are overestimated
|
|
what kind of error can happen in all study types
|
bias
|
|
what is in a case cobntol study
|
selection bias
|
|
giw else can y get selection bias
|
prospectve cohort and experimental atudies
|
|
BIAS dor case control
|
berkson and neyman
|