epi/biostats Flashcards
cross-sectional study
collects data from group to assess disease frequency, risk factors at a particular time. can show dz prevalence
case-control
compares a group with dz to group w/o dz, looks for prior event or risk factor. -> odds ratio
cohort
prospective or retrospective. compares a group with a risk factor/exposure to a group without to see if exposure is associated with incr. dz prevalence. -> relative risk
twin concordance
compares frequency of monozygotic and dizygotic twins developing same dz. -> heritability vs. environmental factors
adoption study
compares siblings raised by biological vs. adoptive parents. -> heritability vs. environmental factors
phase I trial
small n of healthy volunteers. -> safety, toxicity, pharmacokinetics, pharmacodynamics
phase II trial
small n of pts w/dz. -> treatment efficacy, optimal dosing, adverse effects
phase III trial
large n of pts, randomly assigned to tx vs. placebo. -> compares tx to standard of care
phase IV trial
postmarketing surveillance of pts after tx is approved. detects rare or long-term adverse effects.
sensitivity
= true positive rate. important for screening tests. = TP/(TP+FN) = 1-false neg rate. SN-N-OUT = highly sensitive, when negative, rules out dz
specificity
= true negative rate. important for diagnostic tests. = TN/(TN+FP) = 1-false positive rate. SP-P-IN = highly specific, when positive, rules in dz
positive predictive value
=proportion of positive results that are true positives. = TP/(TP+FP). PPV varies w/prevalence/pretest probability
negative predictive value
=proportion of negative results that are true negatives = TN/(TN+FN). NPV varies inversely w/prevalence/pretest probability.
incidence vs. prevalence
incidence = new cases (incidents). prevalence looks at all current cases. prevalence = incidence (almost) for short duration dz, e.g. common cold. prevalence + pretest probability (almost)
odds ratio
case controls. odds that the dz group has target exposure divided by odds that dz free group was exposed
relative risk
cohorts. risk of dz in exposed group divided by risk of dz in unexposed group.
attributable risk
difference in risk vetween exposed and unexposed groups
relative risk reduction
proportion of risk reduction attributable to intervention as compared to the control
absolute risk reduction
difference in risk (not proportion) attributable to the intervention as compared to a control
number needed to treat
number of patients who need to be treated for 1 patient to benefit = 1/ARR
number needed to harm
number of patients who need to be exposed to a risk factor for 1 patient to be harmed = 1/AR
type 1 error (alpha)
false positive error
type 2 error (beta)
false negative error
t test
compares mean of 2 groups
