Epidemiology, Biostatistics, and Research Methods
This domain tests epidemiologic measures, study design, diagnostic test interpretation, and basic inference. Calculation questions reward careful table reading.
Core rules
- Incidence is new cases over time; prevalence is existing cases at a point or period. Prevalence depends on incidence and duration.
- Sensitivity = true positives / all diseased. Specificity = true negatives / all nondiseased. Positive predictive value depends on prevalence.
- Study designs: cross-sectional measures exposure and outcome together; case-control compares diseased and nondiseased retrospectively; cohort follows exposed and unexposed forward; randomized trials allocate intervention.
- Bias includes selection, information, recall, and confounding. A confidence interval estimates precision; a p-value addresses compatibility with the null hypothesis.
- Type I error is a false positive; type II error is a false negative; power is 1 minus type II error.
Worked example
A 2x2 table shows 90 test-positive and 10 test-negative among 100 diseased, and 20 test-positive and 180 test-negative among 200 nondiseased. Reasoning: sensitivity = 90/100 = 90%; specificity = 180/200 = 90%; PPV = 90/(90+20) = 81.8%; NPV = 180/(180+10) = 94.7%. If prevalence falls, PPV falls even though sensitivity and specificity stay fixed.
Traps that cost marks
- Reporting PPV as if it were sensitivity.
- Assuming a p-value is the probability that the null hypothesis is true.
- Confusing incidence with prevalence in a chronic disease.