Public Health and Community Medicine: Population Prevention and Epidemiology
Core concepts
Community medicine asks you to think in populations, not patients. Master measures of frequency (incidence, prevalence) and association (relative risk, odds ratio, attributable risk), then link them to study designs: cohort (incidence, RR), case-control (odds ratio), cross-sectional (prevalence), and randomized trials (causality). Know screening principles: the condition must be important, detectable early, treatable, and the test must have acceptable sensitivity, specificity, and predictive values. Understand prevention levels: primordial, primary, secondary, tertiary. Learn herd immunity thresholds and the distinction between efficacy (trial conditions) and effectiveness (real-world).
Traps in questions
Predictive value depends on prevalence: a highly specific test in a low-prevalence population yields many false positives. Questions confuse incidence with prevalence when duration changes (a treatment that prolongs survival raises prevalence without raising incidence). Screening bias (lead-time and length bias) makes survival look better without changing mortality. Distinguish relative risk from absolute risk reduction when judging clinical importance. Remember that sensitivity and specificity are properties of the test, while predictive values depend on the population.
Worked example
A screening test with 99% specificity in a population with 1% prevalence: of 1,000 people, about 10 have disease and 10 are false negatives (if sensitivity 90%), while about 10 of 990 healthy people are false positives, so positive predictive value is roughly 50%. This shows why specificity matters most in low-prevalence settings.
Rule of thumb
In low-prevalence populations, a positive result needs confirmation; raise the cutoff (higher specificity) to improve positive predictive value.