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;
25 Cards in this Set
- Front
- Back
stats and epi show what?
|
associations
|
|
stats are good at showing ___ but bad at shoing ___?
|
chance; bias
|
|
two situations that lead to false associations?
|
multiple comparisons and disease clusters
|
|
multiple comparisons and disease clusters are examples of what?
|
ad hoc application of stats
|
|
if rr is <1, was disease reduced?
|
yes
|
|
if ci includes 1, is this statistically significant?
|
no
|
|
third variables ethat are linked to exposure and outcome?
|
confounding factors
|
|
systematic error that gets into a study and is one sided?
|
bias
|
|
ultimate way to deal with bias? confounding?
|
blinding; randomization
|
|
confounder must be linked to ___ and ___
|
exposure and disease
|
|
examples of observational studies? interventional/experimental studies?
|
case control and cohort studies; clinical trials
|
|
cohort study?
|
start with disease free; can be prospective or retrospective
|
|
cohort studies won't work if disease is ___
|
rare
|
|
in retrospective, do some people have disease or are they all disease free?
|
some have disease
|
|
problems with retrospective study?
|
confounding factors
|
|
a and d or cohort studies?
|
a: little bias, establish rr, ethics and standard of care can prevent others
d: expensive, continued exposure, takes a long time |
|
case control study?
|
take cases and controls and look back to see if there were any exposure differences
|
|
can you calculate incidence in a case control study? prevalence?
|
no; no
|
|
what do you calculate in a case control study?
|
odds ratio
|
|
what do you calculate in a cohort study?
|
rr
|
|
a and d of case control study?
|
a: cheap, easy, good for rare disease
d: bias, bad if exposure is rare, no info on incidence |
|
when do you do a nested case control study?
|
when the study is too expensive or difficult
|
|
what is nesteed case control study?
|
start with disease free; sample all who develop disease as cases and some who don't for later use as controls
|
|
rr formula?
|
(a/a+b)/(c/c+d)
|
|
odds ratio formula?
|
ad/bc
|