Public Health, Epidemiology, and Pharmacovigilance
Public Health, Epidemiology, and Pharmacovigilance tests population-level reasoning: measuring disease frequency, interpreting diagnostic and screening tests, detecting safety signals, and reporting adverse drug reactions. The exam emphasizes denominators and bias.
Core rules
- Incidence measures new cases per population at risk over time; prevalence measures existing cases at a point or period.
- Sensitivity detects true positives; specificity detects true negatives. Predictive values depend on prevalence.
- Surveillance systems range from passive spontaneous reporting to active and sentinel systems.
- Pharmacovigilance detects, assesses, and prevents adverse effects; spontaneous reports are underreported and cannot prove causality alone.
- Causality assessment considers temporality, biological plausibility, dechallenge, rechallenge, and alternatives.
Worked example
A screening test has 90% sensitivity and 95% specificity, and disease prevalence is 1%. For 10,000 people, 100 have disease and 90 are true positives. Of 9,900 without disease, 5% (495) are false positives. Positive predictive value = 90 / (90 + 495) = 90 / 585 = 0.154, or about 15.4%. Most positives are false despite good sensitivity and specificity because prevalence is low. This is the base-rate effect.
Traps
Do not interpret sensitivity as the chance that a positive result is correct. Do not compare prevalence and incidence without specifying time. Do not treat spontaneous ADR reports as incidence rates. Do not confuse a signal with proof of causation. Do not ignore reporting bias, confounding, or latency when assessing vaccine and drug safety.