Population Health and Community Medicine
Core vocabulary
Sensitivity is the proportion of diseased patients who test positive; specificity is the proportion of well patients who test negative. Positive predictive value depends on prevalence - a good test applied in a low-prevalence population generates many false positives. Know lead-time bias (screening appears to lengthen survival simply by diagnosing earlier), length bias (slower, milder disease is over-represented among screen-detected cases), and the difference between NNT and NNH.
Evidence and guidelines
USPSTF grade A and B services are covered without cost sharing; C and D are not recommended for routine use. Distinguish screening from diagnostic testing, and disease-oriented outcomes (A1c, LDL) from patient-oriented outcomes (mortality, function, quality of life).
Community medicine
- Herd immunity thresholds vary by pathogen; measles requires roughly 95% coverage with two doses, so a single unvaccinated child in a school matters.
- Reportable diseases include most STIs, tuberculosis, measles, hepatitis, and foodborne illness clusters.
- Screen for social determinants - food insecurity, housing, transportation - and connect patients to resources.
- Outbreak investigation: case definition, line list, attack rate, then intervention.
Traps
- Confusing relative with absolute risk reduction; a 50% relative reduction may be trivial in absolute terms.
- Assuming high specificity rules out disease - in a low-prevalence population a positive result may be more often false than true.
- Giving advice instead of using motivational interviewing, the evidence-based brief intervention.
Worked example
A test with 90% sensitivity and 95% specificity in a population with 1% prevalence: among 10,000 people, 90 true positives and about 495 false positives, giving a PPV near 15%. At 20% prevalence the PPV exceeds 80%.