NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
Structured Abstract
Background:
Historically, the National Center for Health Statistics has published statistics on a variety of health outcomes by population subgroup, supporting efforts to assess and monitor health disparities in the U.S. population. With small sample or population sizes, however, direct estimates from the Center’s data systems may lack precision, limiting the information available for small subpopulations. Small-domain estimation is a methodological approach that can be used to provide more precise model-based estimates of outcomes for small subgroups.
Methods:
This report describes the performance of an updated methodology to estimate health outcomes in small subpopulations using the enhanced modified Kalman filter (eMKF). The eMKF procedure borrows strength across subgroups and over time to produce more precise estimates of health outcomes in small subpopulations. This report evaluates the performance of the eMKF procedure under different scenarios based on simulated data, to quantify differences in accuracy and precision of model-based estimates compared with direct estimates.
Results:
Using simulated data, model-based estimates using the eMKF Bayesian model averaging approach presented marked improvements in root mean squared error compared with direct estimates, with larger improvements seen for smaller sample sizes. Improvements were seen across a wide array of analytic scenarios, including outcomes with higher or lower prevalence, trends that varied from linear to cubic, trends that were shared or varied by group, as well as trends that involved unequally spaced time points. In all cases, relative root mean squared errors were smaller for eMKF estimates than direct estimates. Gains in equivalent sample size of up to 420% were observed.
Conclusions:
Model-based estimates of health outcomes among small subpopulations, where direct estimates may be statistically unreliable, can help inform policies and programs to address disparities in the United States.
Keywords:
disparities, inequities, small-domain estimation, National Health Interview Survey, National Health and Nutrition Examination SurveyContents
Suggested citation:
Rossen LM, Talih M, Patel P, Earp M, Parker JD. Evaluation of an enhanced modified Kalman filter approach for estimating health outcomes in small subpopulations. National Center for Health Statistics. Vital Health Stat 2(208). 2024. DOI: https://dx.doi.org/10.15620/cdc/157497.
- NLM CatalogRelated NLM Catalog Entries
- Evaluation of an Enhanced Modified Kalman Filter Approach for Estimating Health ...Evaluation of an Enhanced Modified Kalman Filter Approach for Estimating Health Outcomes in Small Subpopulations
Your browsing activity is empty.
Activity recording is turned off.
See more...