Epidemiology and Biostatistics
Core concepts
Epidemiology studies the distribution and determinants of health states in populations. Master measures of frequency (incidence, prevalence), measures of association (relative risk, odds ratio, attributable risk), and measures of impact. Understand study designs: descriptive (case reports, cross-sectional) and analytic (cohort, case-control, experimental trials). Know sensitivity, specificity, positive predictive value, and screening principles. Biostatistics requires descriptive statistics (mean, median, mode, range, standard deviation) and inferential concepts (confidence intervals, p-values, type I/II errors, sampling methods).
Traps in questions
- Confusing incidence (new cases over time) with prevalence (all existing cases at a point).
- Reversing sensitivity (true positives among diseased) and specificity (true negatives among non-diseased).
- Mixing relative risk (cohort) with odds ratio (case-control).
- Misreading PPV/NPV, which depend on prevalence.
- Confusing type I error (false positive) with type II (false negative).
Worked example / rule of thumb
If a screening test has 90% sensitivity and 95% specificity, a positive result in a low-prevalence population may still have low PPV. Rule of thumb: sensitivity and specificity are properties of the test; PPV and NPV depend on prevalence. For study design, ask: did exposure precede outcome? (cohort) or were cases and controls selected by outcome? (case-control).
Key teaching points
Incidence measures risk; prevalence measures burden. Association does not prove causation. Always check denominators and bias (selection, information, confounding).