Public Health: Population Measures and Study Design
What the candidate must understand
Public health questions test measurement and prevention. Know incidence, prevalence, mortality rates, standardized rates, and the difference between crude and adjusted measures. Understand study designs, including cohort, case–control, and cross-sectional studies, and the measures each yields: relative risk, odds ratio, and prevalence ratio. Know the levels of prevention, screening principles, vaccination schedules and contraindications, and the structure of outbreak investigation. Be able to interpret confidence intervals and to distinguish statistical significance from clinical importance.
Traps that appear in questions
- Calculating relative risk from a case–control study; use the odds ratio instead.
- Confusing incidence, which measures new cases over time, with prevalence, which measures existing cases.
- Ignoring confounding when a crude association is presented.
- Recommending screening for a condition without an accepted screening test or effective early treatment.
- Misreading a confidence interval that includes 1 as evidence of no effect rather than inconclusive evidence.
Worked example: relative risk
In a cohort, 20 of 1,000 exposed and 10 of 1,000 unexposed develop disease. Risk in exposed = 2%, risk in unexposed = 1%, relative risk = 2.0, and attributable risk = 1%. The rule of thumb: cohort studies give risk ratios; case–control studies give odds ratios, which approximate the risk ratio only when the disease is rare.
Rule of thumb
Before computing any measure, identify the study design and the denominator; most errors in public health questions are denominator errors.