Public Health and Epidemiology
Core concepts
Study design, measures of association, outbreak investigation, and health systems are tested. Know incidence versus prevalence, relative risk versus odds ratio, and the interpretation of confidence intervals.
Question traps
- Incidence vs prevalence. Incidence = new cases over time; prevalence = all existing cases at a point. Prevalence rises with longer duration and better survival; incidence is unaffected by duration.
- Relative risk vs odds ratio. Relative risk is used in cohort studies; odds ratio in case-control studies. An OR approximates RR only when the outcome is rare.
- Confounding vs bias. Confounding is a third variable associated with both exposure and outcome; bias is a systematic error in design or measurement. Randomisation addresses confounding; blinding addresses bias.
- Confidence intervals. A 95% CI that excludes 1 (for ratios) or 0 (for differences) indicates statistical significance. A wide CI means imprecision, not necessarily absence of effect.
- Outbreak investigation. The first steps are to confirm the diagnosis, define the case, and describe the outbreak by person, place, and time before hypothesising.
Worked rule of thumb
For any measure-of-association question, first identify the study design. Cohort → relative risk; case-control → odds ratio; cross-sectional → prevalence. This single step eliminates most distractors.
Health systems
Know the levels of prevention and the concept of herd immunity threshold, which depends on the basic reproduction number.