American Journal of Epidemiology Advance Access originally published online on October 15, 2008
American Journal of Epidemiology 2008 168(10):1214-1216; doi:10.1093/aje/kwn317
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
LETTERS TO THE EDITOR |
RE: "AN OVERVIEW OF METHODS FOR MONITORING SOCIAL DISPARITIES IN CANCER WITH AN EXAMPLE USING TRENDS IN LUNG CANCER INCIDENCE BY AREA-SOCIOECONOMIC POSITION AND RACE-ETHNICITY, 1992–2004"
Public Health Sciences Section, Division of Community Health Sciences, School of Clinical Sciences and Community Health, University of Edinburgh, Teviot Place, Edinburgh EH89AG, United Kingdom
(e-mail: raj.bhopal{at}ed.ac.uk)
The appropriate measurement of ethnic disparities/inequalities in health and disease is crucially important, not just for monitoring but also for choosing priorities for interventions. I would like to add to the debate that Harper et al. (1) and Messer (2) have started, not least to offer a European perspective.
Our modern-day, multiethnic world has set lofty goals of racial equality and—even more importantly, because they are not merely aspirations—ethical and just policies, that is, equity (3). Our goals, in the United States and Europe alike, are backed by legislation and national policy. These goals are immensely difficult to achieve, and the absence of both clarity in conceptual reasoning and high-quality data are twin obstacles, arguably making the task impossible. The underpinning questions of why we have set these goals, and the consequent health-care actions required, are as important as the technical questions of measurement (3).
Harper et al.’s analysis of 7 methods (1) shows that they lead to contradictory conclusions on whether disparities in lung cancer have narrowed or not. In fact, relative and absolute approaches inevitably contradict each other when the populations have the same proportionate reductions in risk. Having demonstrated that relative and absolute approaches lead to very different priorities for action, I wrote in 1988 that both relative measures and absolute measures were essential, with the latter deserving primacy (4). These ideas were developed with Senior (5), with the conclusion that the usual epidemiologic method of contrasting minority populations with the white/European-origin population, particularly using relative measures, was seriously misleading and ethnocentric. I proposed that in all ethnicity and health research with a policy or health-services intention (as is the case with analysis and reduction of disparities), investigators present their data as shown in Table 1 (6). This approach was applied to a national health needs assessment, producing the first rounded picture of ethnic variations in health and disease in the United Kingdom (7).
|
With Donaldson (8), I made the argument that the white population should not be the automatic reference group, as is almost invariably the case in practice—though recent work, including Harper et al.’s (1), has been veering away from this. I have suggested that the group with the lowest disease rate should be the reference group (3). This approach would reduce the stigma associated with tackling conditions for which there is an excess in ethnic minority populations, for it would show where their status was better than that of the majority. Obviously, this recommendation would need to be modified if such a group were small and if the disease rates were imprecise.
I have stressed the dangers of using broadly defined ethnic groups (9). Major variations are missed, and false conclusions of "no difference" are all too common. To take an obvious example—one that is seldom acted upon in the United States—populations of Chinese origin have comparatively low rates of cardiovascular disease, while populations of Indian origin have comparatively high rates (7). By combining these groups as "Asian," both important findings are usually lost. There are also substantial variations within populations defined as Indian or Chinese (10). Harper et al.’s work (1) is highly advanced from a technical point of view but crude in relation to the ethnic groupings used. Administrative categories, created for nonmedical purposes, should not be permitted to dominate epidemiology, which needs to be guided by causal models, as Messer proposes (2) and Harper et al. (1) agree with.
The next step is to think about how we can intervene to reduce inequalities, and that requires attention to conditions that are major burdens as reflected in absolute measures. Focusing policy on conditions for which there is a relative excess can widen, not narrow, ethnic-group inequalities. Furthermore, equal reductions in absolute terms across ethnic groups will increase inequalities in relative terms, not reduce them. I have illustrated this elsewhere (3), but the basic idea is given in Table 2.
|
I welcome these papers, not just for their own merits, but because they open up a relatively neglected area of debate, analysis, and research. We have been sketching out the territory for some 20 years here in the United Kingdom.
 |
ACKNOWLEDGMENTS |
|---|
Conflict of interest: none declared.
|
References |
|---|
 TOP References |
|---|
- Harper S, Lynch J, Meersman SC, et al. An overview of methods for monitoring social disparities in cancer with an example using trends in lung cancer incidence by area-socioeconomic position and race-ethnicity, 1992–2004. Am J Epidemiol (2008) 167(8):889–899.
[Abstract/Free Full Text] - Messer LC. Invited commentary: measuring social disparities in health—what was the question again? Am J Epidemiol (2008) 167(8):900–904.
[Abstract/Free Full Text] - Bhopal RS. Ethnicity, Race, and Health in Multicultural Societies: Foundations for Better Epidemiology, Public Health, and Health Care (2007) New York, NY: Oxford University Press.
- Bhopal RS. Health care for Asians: conflict in need, demand and provision. A prerequisite for health. In: Equity: The Proceedings of the 1987 Summer Scientific Conference of the Faculty of Community Medicine (1988) London, United Kingdom: Faculty of Community Medicine [Royal College of Physicians] and World Health Organization.
- Senior PA, Bhopal R. Ethnicity as a variable in epidemiological research. BMJ (1994) 309(6950):327–330.
[Free Full Text] - Mackintosh J, Bhopal R, Ahmad N. Step by Step Guide to Epidemiological Health Needs Assessment for Ethnic Minority Groups (1998) Newcastle, United Kingdom: University of Newcastle. 1–99.
- Gill PS, Kai J, Bhopal RS, et al. Black and minority ethnic groups. In: Stevens A, Raftery J, Mant J, et al, eds. Health Care Needs Assessment: The Epidemiologically Based Needs Assessment Reviews. Abingdon, United Kingdom: Radcliffe Medical Press Ltd; 2007:227–239. (http://www.hcna.bham.ac.uk/series/bemgframe.htm).
- Bhopal R, Donaldson L. White, European, Western, Caucasian, or what? Inappropriate labeling in research on race, ethnicity, and health. Am J Public Health (1998) 88(9):1303–1307.
[Abstract/Free Full Text] - Bhopal R, Unwin N, White M, et al. Heterogeneity of coronary heart disease risk factors in Indian, Pakistani, Bangladeshi, and European origin populations: cross sectional study. BMJ (1999) 319(7204):215–220.
[Abstract/Free Full Text] - Ramaiya KL, Swai AB, McLarty DG, et al. Prevalences of diabetes and cardiovascular disease risk factors in Hindu Indian subcommunities in Tanzania. BMJ (1991) 303(6797):271–276.
[Abstract/Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. Bhopal Medicine and public health in a multiethnic world J Public Health, September 1, 2009; 31(3): 315 - 321. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||