Epidemiology/ Biostatistics I Flashcards
What does a case-control study compare?
Group of people with a disease to group of people without; looks for prior exposure/ risk factor. (p.50)
How are the results of a case-control study measured?
Odds Ratio (p.50)
What does a cohort study compare?
Group with a given exposure/ risk factor to group without this exposure to see if exposure increases liklihood of disease (p.50)
How are the results of a cohort study measured?
Relative Risk (p.50)
How are the results of a cross sectional study measured?
Disease prevalance; causality is not established (p.50)
What is sensitivity?
Proportion of all people with a disease who test positive; probability that the test detects a disease when it is present (p.51)
How is sensitivity measured?
TP/ (TP + FN) or 1- false negative rate (p. 51)
When are tests with high sensitivity most useful?
In screening for diseases with low prevalance (p.51)
What is specificity?
Proportion of all people without the disease who test negative; probability that a test indicates non-disease when disease is absent (p.51)
How is specificity measured?
TN/ (TN + FP) or 1- false positive rate (p.51)
When are tests with high specificity most useful?
As confirmatory tests after a positive screening test (ex. HIV testing) (p.51)
What can Positive Predictive Value measure?
Probability that a person actually has the disease given a positive test result (p.51)
How is Positive Predictive Value measured?
TP/ (TP + FP); it varies with prevalance, the higher the prevalance the higher the PPV
What can Negative Predictive Value measure?
Probability that a person actually is disease free given a negative test retult (p.51)
How is Negative Predictive Value measured?
TN/ (FN + TN); varies inversely with prevalance, high prevalance yields low NPV
What is an odds ratio?
Used in case control studies; odds that the group with the disease was exposed to a risk factor divided by the odds that the group without the disease was exposed; in 2x2 square odds ratio = (a/c)/(b/d) (p.52)
What is relative risk?
Used in cohort studies to determine risk of developing the disease in the exposed group divided by risk in the unexposed group; in 2x2 square: RR= (a/(a+b)/(c/(c+d)) (p.52)
What is attributable risk?
The difference in risk between exposed and unexposed groups; the proportion of disease occurances atrributable to the exposure; in 2x2 table AR= (a/(a+b))- (c/(c+d)) (p.52)
What is absolute risk reduction?
Absolute resuction of risk associated with a treatment compared to a control (p.52)
How do you calculate Number Needed to Treat?
1/ absolute risk reduction (p.52)
How do you calculate Number Needed to Harm
1/ attributable risk (p.52)