Public Health, Social Medicine, and Medical Law/Ethics
This domain combines epidemiology and biostatistics with the Japanese health system and the legal duties of a physician. Calculations appear every year, so practise them to the point of automaticity.
Core rules
- Study design: cohort studies yield incidence and relative risk; case-control studies yield odds ratios and suit rare outcomes; randomisation controls confounding; ecological studies risk the ecological fallacy.
- Screening: sensitivity = TP/(TP+FN), specificity = TN/(TN+FP), positive predictive value = TP/(TP+FP). PPV falls sharply as prevalence falls - the central point of mass screening.
- Bias: lead-time and length-time bias inflate survival in screened populations; the healthy worker effect; recall bias; confounding versus effect modification.
- Japanese systems: universal health insurance; late-stage elderly system (75+); long-term care insurance (40+); specific health checkups (特定健診) targeting metabolic syndrome; occupational health under the Industrial Safety and Health Act.
- Law and ethics: 医師法 requires notification of unusual death (異状死) within 24 hours; rules on death certificates, confidentiality, and informed consent; Mental Health Act involuntary admission categories; Organ Transplant Act; Infection Control Act categories 1–5.
- Four principles: autonomy, beneficence, non-maleficence, justice.
Worked example
A test with 90% sensitivity and 95% specificity is applied to 10,000 people with 1% prevalence. Reasoning: 100 diseased, 90 detected; of 9,900 healthy, 5% false positive = 495. PPV = 90/585 ≈ 15%. Most positives are false positives, so a positive result needs confirmatory testing and counselling about anxiety and cost.
Traps
- Reporting relative risk from a case-control study.
- Assuming a sensitive test is a good diagnostic test regardless of prevalence.
- Confusing the duty to notify an unusual death with an obligation to report to police.