Clinical Questions - Biostatistics Flashcards
Sensitivity
A/A+C
Specificity
D/B+D
Probability
Probability = predictive value. Look at the stem to determine if it is asking for the PPV or NPV
PPV = A/A+B NPV = D/C+D
Attributable Risk
AR = A/(A+B) - C/(C+D)
Incidence
Number of new cases/population at risk of the disease within a given timeframe
Prevalence
Number of total cases in the population/population at risk of the disease minus dead and cured within a given timeframe
Research study that compares a group of people with a given disease to a group without the disease
Case control study - retrospective observational
Research study with a given risk factor or exposure to a group without the risk factor
Cohort study - prospective or retrospective
Participants are randomly assigned to either the experimental group or the control group
Randomized controlled trial
Relative risk
RR = (A/(A+B)) / (C/(C+D))
Absolute risk reduction
ARR = C/C+D - A/A+B
Number needed to treat
1/ARR
Standard error of the mean
SEM = standard deviation / sq rt of sample size
Confidence interval
Mean +/- Z (SEM)
CI of 90%: Z = 1.645
CI of 95%: Z = 1.96
CI of 99%: Z = 2.57
Z for CI of 90%
1.645
Z for CI of 95%
1.96
Z for CI of 99%
2.57
Individual who experience an adverse outcome have a different likelihood of recalling an exposure than those without the adverse outcome
recall bias
Early detection may be confused with an increase in survival
lead time bias
The association between an exposure and an outcome is distorted by another variable
Confounding bias
Subjects in different arms of a study are systematically treated differently
procedure bias
information is distorted by the way it is gathered
measurement bias
