1. Keppel, Kenneth G. PhD
  2. Pearcy, Jeffrey N. MS

Article Content

Response to Scanlan Concerning: Measuring Relative Disparities in Terms of Adverse Events


In his letter and his articles, Scanlan raises many issues relevant to the measurement of disparities in health. Our article addresses the implications of one of Scanlan's observations-that reductions in the relative disparity between blacks and whites in infant survival correspond to increases in the relative disparity in rates of infant mortality.1 The choice to measure a disparity in terms of favorable or adverse events affects the magnitude and direction of relative measures of disparity between groups, comparisons of relative measures of disparity across different indicators, and conclusions about changes in relative measures of disparity over time. In the interests of consistency, we argue in favor of expressing indicators in terms of adverse events when relative measures of disparity are used.2


In his letter, Scanlan expresses the concern that "the rarer the outcome the greater the relative disparity in experiencing it and the smaller the disparity in avoiding it." This observation does not necessarily apply to changes in disparity where the distribution of events for two groups is changing. Over time, as long as the proportional improvement for the less favored group exceeds the proportional improvement for the more favored group, the relative disparity between these two groups will be reduced. We recommend making these comparisons on the basis of adverse events.


Scanlan also expresses concerns about the basis for conclusions about changes in disparity over time. Elsewhere, we have argued that differences between groups should be interpreted not only based on absolute and relative measures of disparity but also in terms of the frequencies on which these measures are based.3 Similarly, changes in disparity should be interpreted from each of these perspectives. We agree that questions about what it means to reduce disparity, and particularly what it would mean to eliminate disparity, need additional attetion.




1. Scanlan JP. Divining difference. Chance. 1994;7:38-39, 48. [Context Link]


2. Keppel KG, Pearcy JN. Measuring relative disparities in terms of adverse events. J Public Health Manag Pract. 11(6);479-483. [Context Link]


3. Keppel K, Pamuk E, Lynch J, et al. Methodological issues in measuring health disparities. Vital Health Stat. 2005;2(141). [Context Link]