A district health officer in Tamil Nadu is evaluating the impact of a new hypertension screening program launched 6 months ago. Before the program, the recorded prevalence of hypertension was 15% (based on hospital records). After 6 months of intensive community screening, the prevalence has risen to 22%. The officer suspects this reflects successful case detection. What is the most appropriate next step to confirm whether the increase in prevalence is due to improved detection versus a true increase in incidence of hypertension?

Explanation

## Understanding the Epidemiological Dilemma The officer faces a classic interpretation challenge: - **Prevalence increased from 15% to 22%** after screening - This could mean: - **True increase in incidence** (new cases arising) - **Detection bias** (existing undiagnosed cases now identified) - **Combination of both** ## Why Incidence Comparison Is the Answer **Key Point:** Prevalence is a **snapshot** that rises when: 1. Incidence increases (more new cases) 2. Detection improves (hidden cases become visible) 3. Duration increases (people live longer with disease) To distinguish detection bias from true disease increase, one must measure **incidence** in comparable populations. **High-Yield:** - **Incidence in screened population** = newly detected cases (includes both pre-existing undiagnosed + truly new) - **Incidence in unscreened population** = only newly occurring cases (no detection of pre-existing) - **Difference** = magnitude of detection bias ## The Correct Methodological Approach ```mermaid flowchart TD A[Prevalence increased 15% → 22%]:::outcome --> B{What caused the rise?}:::decision B -->|Detection bias| C[Pre-existing undiagnosed cases now found]:::outcome B -->|True incidence increase| D[New cases occurring at higher rate]:::outcome B -->|Both| E[Detection + new cases]:::outcome C --> F[Compare incidence: screened vs unscreened]:::action D --> F E --> F F --> G[Screened incidence >> Unscreened incidence = Detection bias dominant]:::outcome F --> H[Screened incidence ≈ Unscreened incidence = True increase dominant]:::outcome ``` **Clinical Pearl:** If the screened population has much higher incidence than the unscreened population, the prevalence rise is largely due to detection of pre-existing cases. If incidence rates are similar, the rise reflects true new disease occurrence. ## Why This Matters for Public Health If the increase is **purely detection bias**, the screening program has: - Successfully identified hidden cases (good for individual management) - Not prevented new cases (incidence unchanged) If the increase is **true incidence rise**, the program must also address: - Underlying risk factors - Prevention strategies - Community education