Abstract:
|
Rates, defined as the number of events divided by a count for the population at risk, are widely disseminated by the National Center for Health Statistics (NCHS). The NCHS Data Presentation Standards Workgroup examined statistical principles for rates and developed data presentation standards for rates, which are expected to be applied to all NCHS data systems starting in 2022. In this presentation, we evaluate the proposed presentation standards for rates (such as rates for certain ambulatory care visits per 100 population) from the National Ambulatory Medical Care Survey (NAMCS) via a simulation study. Several methods to calculate confidence intervals (CIs) for rates derived from NAMCS were compared. The impact of the survey design and sample size of the numerator and the denominator on the CI coverage and relative width of the CI were evaluated. The transparent criteria of the presentation standards for rates ensure rate estimates disseminated by NCHS meet certain thresholds of statistical reliability.
|