Epi/Biostats Flashcards
Cross-sectional Study
observational
assesses frequency of disease at a particular point in time
measures disease prevalence
Case-control Study
observational
compares group with disease to a group without disease to look for exposure
measures odds ratio
Cohort Study
observational
compares group with exposure that without
measures relative risk
prospective OR historical
Twin Concordance Study
compares frequency of disease in MZ or DZ twins
nature vs nurture
Adoption Study
compares siblings with bio vs adoptive parents
nature vs nurture
Phase I clinical trial
small # healthy volunteers
safety, tox, PK, PD
Phase II clinical trial
small # of pts with disease
efficacy, dosing, AE
Phase III clinical trial
Large # of pts, random assignment
compare to standard therapy
Phase IV clinical trial
post-marketing surveillance
AE
Sensitivity
True Positive Rate=TP/(TP+FN)
probability of a positive test when the disease is present
high = low FN rate
SN-N-OUT (highly SeNsitive test, when Negative, rules OUT disease)
Specificity
True Negative Rate=TN/(TN+FP)
probability of a negative test when the disease is absent
high=low FP rate
SP-P-IN (highly SPecific test, when Positive, rules IN disease)
confirm after + screening test
Positive predictive value
PPV=TP/(TP+FP)
probability that a person with a + test actually has the disease
varies directly with disease prevalence
Negative predictive value
NPV=TN/(TN+FN)
probability that a person with a - test actually does not have the disease
varies indirectly with disease prevalence
Incidence
# of new cases/# at risk during a specified time period
Prevalence
# of existing cases/total# at a point in time
Odds ratio
odds that the group with the disease was exposed to a RF divided by the odds that the group without the disease was exposed
Relative Risk
Risk of developing disease in the exposed group divided by risk in the unexposed group
equals OR at low prevalence
Attributable Risk
Difference in risk between exposed and unexposed groups
the proportion of disease occurrences that are attributable to the exposure
Relative Risk Reduction
the proportion of risk reduction attributable to the intervention as compared to a control
RRR=1-RR
Absolute Risk Reduction
difference in risk attributable to the intervention as compared to a control
Number Needed to Treat
NNT=1/ARR # of patients who need to be treated for 1 pt to benefit
Number Needed to Harm
NNH=1/AR # of patients who need to be exposed to a RF for 1 pt to be harmed
Precision
consistency, reproducibility, reliability
absence of random variation (SD)
increases power
Accuracy
trueness, validity
absence of systematic error or bias
Selection Bias
error in recruiting and assigning subjects
may correct with randomization
Recall Bias
awareness alters recall by subjects
common in retrospective studies
Measurement Bias
distorted info gathering
correct with standardized testing
Procedure Bias
subjects in different groups are not treated the same
can correct with blinding and placebo use
Observer-expectancy bias
when the researcher believes the treatment is efficacious
use blinding and placebo
Confounding bias
when a factor is related to both exposure and outcome but not in same causal path
can match or randomize
Lead-time bias
early detection is confused with increased survival
Normal distribution
68% between 1 SD either side of mean
95% between 2 SD
99.7% between 3 SD
Type I error (alpha)
stating that there is a difference when none exists
incorrectly reject the null hypothesis
False-positive error
alpha level
Type II error (beta)
the null hypothesis is not rejected when it is false
stating that there is not a difference when one exists
false-negative error
Power
1-beta
probability of rejecting the null when it is false
increases with sample size, expected effect size, and precision
Confidence Interval
range of values within which the true mean falls
CI=mean+/-Z(SEM)
95% CI for p=0.05
t-test
difference between means of 2 groups
ANOVA
differences between means of 3 or more groups
Chi-square
differences between 2 or more proportions of CATEGORICAL outcomes