Epidemiology and Cultural Psychiatry
What the exam tests
This domain tests measures of frequency and association, study design, bias, screening statistics and culturally shaped presentations. Expect calculations and questions about whether a finding is causal, cultural or an artefact.
Core rules
- Incidence is new cases per population at risk per time; prevalence is all existing cases at a point or period. Prevalence rises with longer duration and better survival.
- Relative risk compares incidence between exposed and unexposed groups; odds ratio approximates it in rare outcomes.
- Sensitivity detects true cases; specificity detects true non-cases; positive predictive value depends heavily on prevalence.
- Bias includes selection, recall, Berkson and lead-time bias. Association is not causation.
- Cultural psychiatry distinguishes idioms of distress, explanatory models and culture-bound syndromes from universal psychopathology; use a cultural formulation rather than pathologising difference.
Worked example
A screening test has 90% sensitivity and 90% specificity in a population with 10% prevalence. Of 1000 people, 100 have the disorder; 90 are true positives and 10 are false negatives. Of 900 without it, 810 are true negatives and 90 are false positives. PPV is 90/(90+90), or 50%. The reasoning is that even a good test produces many false positives when prevalence is low; confirmatory assessment is essential.
Traps that cost marks
Do not present PPV as a fixed property of the test. Do not confuse incidence with prevalence. Do not label a culturally normative experience as psychosis. Avoid ecological fallacy: group-level associations do not apply automatically to individuals.