Stats Flashcards
how to calculate relative risk?
risk of exposed having disease
over
risk of nonexposed having disease
a/a+b divided by c/c+d
what is the formula for odds ratio?
ad/bc
What is present with right or positive skewed data?
mean is larger than the median
how do you calculate positive and negative predictive value?
true positives over total positive tests
true negatives over total negative tests
what is the formula to calculate number needed to harm?
1/attributable risk
in words, what is a type 1 error and type 2 error?
Type 1 is when the study shows a difference, but there is no difference
type 2 is the study shows no difference but there is difference
power of a study increases when what happens to alpha and sample size?
increase
what do you want to be high for screening tests and what do you want to be high for confirmatory tests?
snesitivty
specificty
what is precision and accuracy in a study?
precision is reproducibility, are the darts close to eachother
accuracy is does the data represent reality, are the darts close to the center
what is the formula for number needed to treat?
1/absolute risk reduction
what can we conclude if the 95% confidence interval crosses 1
not signifcant
simply explain the four phases new drugs must go through before going to market?
1: healthy volunteers
2. small amount of people with disease
3. large scale trial
4 genera surveillance
formula for sensitivty and specficity?
disease and positive test/all positive tests
no disease and negative test/all negative tests
how much of the data is in 1, 2 and 3 standard deviaitons?
68
95
99.7
