Measuring Disparities in Healthy People 2010
*Kenneth G Keppel, National Center for Health Statistics
Keywords: health disparities, measurement, rate differences
Consensus has been reached on the measurement of disparities in Healthy People 2010. Disparities are measured in relative terms, from the best groups rate, with all indicators expressed in terms of adverse events. Increases and decreases in disparity can, therefore, be measured. In order to eliminate disparities, we must also reach consensus about how to measure the absence of disparity. If we can't define what "parity" is, then we don't really know what "disparity" is. Tests of statistical significance are not very helpful. When the size of the group or the size of the sample is small, we are forced to accept the null hypothesis even when relative differences are large.