Dear Editor

Although there is nothing incorrect in the effort by Karpati et al. [1] to appraise changing neighborhood mortality inequalities in New York City, the effort is compromised by a failure to recognize the statistical tendency whereby the rarer an outcome, the greater the relative difference in experiencing it (and the smaller the relative difference in avoiding it). The authors point out that relative inequalities in mortality in the United Kingdom have been increasing despite absolute improvements in life expectancy at all levels. But in times of declining mortality one should expect to observe increasing relative differences in mortality rates between more and less advantaged groups solely as a result of the mortality decline [2-6]. Whether there occurred in the United Kingdom any change in relative health that was not solely a result of declining mortality has gone unexamined [6]. In the case of New York City, the fact that overall declines in mortality resulted in even a small reduction in relative mortality differences, rather than the increase in relative differences that would more typically occur, should be interpreted (although cautiously [2,6]) as a favorable development.

The patterns in Table 1 of the Karpati article regarding changes in relative differences in death rates from particular diseases must be similarly interpreted in light of the tendency for declines in prevalences to increase relative differences and increases in prevalence to reduce relative differences. But it remains difficult to draw conclusions about meaningful changes in relative differences regarding particular causes – i.e., changes that are not solely a consequence of changes in prevalence – with any degree of confidence [2,6].

The tendency for relative differences to be greatest where an outcome is rarest also suggests that certain of the study’s broader findings concerning relative differences in adverse health outcomes across New York City neighborhoods are pretty much what one should expect. The study found that relative differences in mortality are greater among the young than the old. Such a pattern is what one ought to expect to find (and what one typically in fact finds) simply because mortality is rarer among the young [6]. Absolute mortality differences, as well as relative differences in survival rates, however, tend to be greater among the old [6,7]. The authors also find that while heart disease and cancer show comparatively small age-adjusted mortality rate ratios across neighborhoods, such diseases are among the most important contributors to excess mortality in the poorest neighborhoods. Such a pattern is also to be expected simply because more common outcomes tend to show small relative differences but large absolute differences (as reflected in Table 1); and it is the absolute difference that determines the contribution of a particular diseases to a group’s overall excess mortality (whether measured in absolute or relative terms). One similarly observes that nationally the causes of death with comparatively small black-white mortality ratios tend to be the largest contributors to overall racial disparities in mortality [8].

The authors also employ the relative index of inequality (RII) and years of potential life lost before age 65 (YPLL-65) to appraise changes in the size of inequalities over time. The RII, which is simply a variation on the standard relative difference, is subject to the same interpretive problems as the relative difference with respect to identifying changes in the respective situation of two groups that are not solely a consequence of changes in prevalence of an outcome. YPLL-65, like longevity differences, would seem also to change solely as a result of changes in the overall prevalence of mortality, but with a direction of change that is difficult to predict [2,6]. Thus, neither of these measures obviates the difficult problem or distinguishing changes in the relative health of two groups that are solely a consequence of overall declines in mortality from those that are not. The expanded monitoring of health patterns suggested by the authors should be undertaken in conjunction with an effort to develop tools for addressing such problem.

James P. Scanlan

References

1. Karpati AM, Bassett MT, McCord C. Neighborhood mortality inequalities in New York City, 1989-1991 and 1999-2001. J Epidemiol Community Health 2006;60:1060-1064.

2. Scanlan JP. Can we actually measure health disparities? Chance 2006;19(2):47-51: http://www.jpscanlan.com/images/Can_We_Actually_Measure_Health_Disparities.pdf.

3. Scanlan JP. Measuring health disparities. J Public Health Manag Pract 2006;12(3):294 [Lttr]: http://www.nursingcenter.com/library/JournalArticle.asp?Article_ID=641470 (accessed Nov. 10, 2006).

4. Scanlan JP. Race and Mortality. Society 2000;37(2):19-35: http://www.jpscanlan.com/images/Race_and_Mortality.pdf.

5. Scanlan JP. Divining difference. Chance 1994;7(4):38-9,48 http://jpscanlan.com/images/Divining_Difference.pdf.

6. Scanlan JP. The misinterpretation of health inequalities in the United Kingdom: Paper presented at: British Society for Population Studies Annual Conference 2006, Southampton, England, Sept. 18-20, 2006: http://www.jpscanlan.com/images/BSPS_2006_Complete_Paper.pdf (accessed Nov. 10, 2006).

7. Scanlan JP. Measuring health inequalities: Paper presented at: 5th International Conference on Health Economics, Management and Policy, Athens, Greece, June 5-7, 2006: http://www.jpscanlan.com/images/Measuring_Health_Inequalities.pdf.

8. Wong MD, Shapiro MF, Boscardin WJ, Ettner SL. Contributions of major diseases to disparities in mortality. N Engl J Med. 2002;347:1585 - 92.