Behavioral Science Flashcards
(41 cards)
Case control study
compares to group without disease
ODDS RATIO
Cohort study
compares to group without exposure
RELATIVE RISK
Phase I drug trial
healthy volunteers
Phase II drug trial
small # of patients with disease
Phase III drug trial
large # of patients randomly assigned to treatment or placebo
Phase IV drug trial
postmarket surveillance after treatment approved
Sensitivity
= TP/(TP+FN)
= 1 - false negative rate
when NEGATIVE, rules OUT DISEASES
SCREENING
Specificity
= TN/(TN+FP)
= 1 - false postiive rate
when POSITIVE, rules IN disease
CONFIRMATION (after positive screening test)
PPV
= TP/(TP+FP)
directly related to high pretest probability (prevalence)
NPV
= TN/(TN+FN)
Incidence
new cases/ #people at risk
Prevalence = pretest probability
existing cases/ #people at risk
OR
odds that diseased/non diseased exposed to RF
= (a/c)/(b/d)
RR
risk of developing disease
= (a(a+b))/(c/(c+d))
AR (attributable risk)
difference in risk
= a/(a+b) - c/(c+d)
RRR (relative risk reduction)
proportion of risk reduction attributable to intervention
= 1 - RR
ARR (absolute risk reduction)
= c/(c+d) - a/(a+b)
NNT
=1/ARR
NNH (number need to harm)
number of patients exposed for 1 pt to be harmed
=1/AR
precision
consistency, reproducibility, reliability
decreases with random error
accuracy
validity
decreases with systematic error
confounding bias
e.g. pulmonary disease more common in coal workers than general pop; however, people who work in coal mines also smoke more
positive skew
mean>median>mode
negative skew
mean<mode