Which of the following is a common issue in incidence and prevalence measurement?

• A. Hospital admissions are comprehensive for all populations
• B. Hospital records are designed for treatment, not research
• C. Numerators are always perfectly recorded
• D. Denominators never undercount minority groups

Answer: (B)

Data source design and data quality strongly influence estimates of how many people have a disease and how many new cases occur. When we measure incidence and prevalence, we need reliable counts of cases (numerators) and a well-defined population at risk (denominators). Hospital records fit the bill in terms of containing diagnostic information, but they are created for clinical care and billing, not for research or population surveillance. This means many people with the condition may not be captured if they aren’t hospitalized, are treated in outpatient settings, or are coded differently, leading to incomplete or biased counts. Diagnoses can be miscoded or inconsistently recorded, and patterns of who enters the hospital can change over time or vary by population, further distorting the measurements. Because hospital data aren’t designed to reflect true population incidence and prevalence, they represent a common source of measurement bias in these estimates.