PCD logo

The Rate Stabilizing Tool: Generating Stable Local-Level Measures of Chronic Disease

PEER REVIEWED

Two scatterplots compare the root mean square error associated with the unsmoothed/crude estimates with the estimates obtained from the Rate Stabilizing Tool’s 2 approaches. In both cases, the scatterplots indicate that the nonspatial and spatial smoothing approaches of the RST produce estimates with lower root mean square errors than the unsmoothed/crude estimation approach.

Figure 1.
Comparison of the root mean square error (rMSE) of the age-standardized rates from the 2 smoothing approaches (A, nonspatial vs crude estimates and B, spatial vs crude estimates) of the Rate Stabilizing Tool to the unsmoothed rates estimated directly from the raw data in the simulation study.

Return to Article

Figure 2 contains various maps of the region surrounding Charlotte, North Carolina. First, it shows a map of the unsmoothed/crude mortality rates. This map contains many estimates in the lowest and highest quintiles of the distribution, which could be due to small population sizes. Next, the figure displays maps of the estimated mortality rates from the nonspatial and spatial smoothing approaches of the Rate Stabilizing Tool. Here we find that some of the extreme values from the first map have been brought toward the mean. Furthermore, these maps highlight regions with unreliable rate estimates. Finally, the figure contains 2 additional maps that indicate whether the Rate Stabilizing Tool’s estimates are deemed to be significantly higher or lower than the rate in the state of North Carolina as a whole.

Figure 2.
Illustration of the functionality of the Rate Stabilizing Tool using heart disease mortality data from the region surrounding Charlotte, North Carolina. A, Age-standardized heart disease death rates by census tract using 3 methods, with hatch marks indicating unreliable rates based on the 2 Bayesian smoothing approaches. B, census tracts with death rates that are significantly higher or lower than the state rate using the 2 smoothing methods.

Return to Article

Top

Error processing SSI file

The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.