ROUND TABLE DISCUSSION
Combining forces against inequity and poverty rather than splitting hairs
Paula Braveman
Professor of Family and Community Medicine, University of California, San Francisco, Box 0900, 500 Parnassus Avenue, Room MU-306E, CA 94143-0900, USA
Davidson Gwatkins excellent lead article raises many points with important policy implications. I shall comment on just two: equity targets, and health "equity" versus "poverty" approaches.
Equity targets focus attention and increase accountability
Gwatkin aptly points out the need to move from the standard practice of setting population health goals in terms of averages to specifying goals systematically and explicitly in distributional terms. He demonstrates concretely how sizeable improvements can be manifest in an indicators average level but be concentrated in the better-off groups. In the present global circumstances there is a particular risk of this occurring, since knowledge of preventive practices and ability to apply them, as well as access to expensive medical technologies, are likely to become increasingly concentrated in the better-off groups in the absence of concerted public policy action.
What stands in the way of countries adopting equity targets? From a research perspective, a lack of reliable data on health indicators disaggregated by social group immediately comes to mind, as does lack of consensus about indicators. Progress towards targets must be measurable in terms of specified indicators; otherwise there is no accountability or way to assess whether policies are likely to be generally on or off track in leading to greater equity. Governments routinely report on scores of health indicators. Should there be an equity target for all of them or only for selected ones, and if the latter, which ones? The literature suggests that disparities in health status associated with socioeconomic differences narrow during adolescence, rise again in young adulthood, then fall again among the elderly (13). Should all periods in the life-course be reflected? All health conditions? The literature also indicates that associations between socioeconomic status and health vary with both the health indicator and the socioeconomic measure considered. Should we select the indicators likely to reveal the widest gaps or use other criteria, such as the modifiability of disparities, or the total burden of ill-health (considering both prevalence and severity of health consequences)?
These are difficult questions, but a far more difficult one lies in the political realm: how to build a broad societal consensus. Societies may generally be less tolerant of social inequalities in health than in wealth but they are likely to vary both in the levels of health inequality they will tolerate, and in what the better-off are willing to pay to reduce inequality. Society as a whole including the better-off, who have greater political power must feel invested in improving the health of all its members, including those previously left out. Towards that end, equity targets should supplement rather than replace average targets. And there must be adequate public discussion and consensus-building to provide the basis both for setting targets and for formulating actions to achieve them.
Health equity targets may indeed provide a useful mechanism to help build and maintain societal consensus. Both the initial target-setting process and periodic reviews of progress towards achieving targets could provide a focus for public discussion of societal views and policy options; clearly, all social groups including advocates for the disadvantaged and all relevant sectors must be included. Gwatkin has rightly noted that effective response to social inequalities in health will often require action outside the health sector. One of the key functions of the health sector should be to provide information on an ongoing basis to reflect the health consequences (overall and for more and less advantaged groups) of actions in all sectors. This role should not depend on whether health services themselves are likely to be of major importance in an effective response to observed health inequalities.
"Equity" and "poverty" approaches
I also would like to comment on the "poverty" versus "equity" distinction, and to second Gwatkins conclusion that the practical differences between these approaches are far smaller than the commonalities. I would like to underscore the wide consensus among those working under the "equity" banner that absolute measures of the health of different social groups, not only relative differences between them, are essential for assessing health equity (4, 5). Correspondingly, in focusing on improving the health of the poor, one is implicitly trying to close a gap, by bringing the poor up to a level of health experienced by better-off groups. The levels achieved by the better-off suggest what is possible and, by extension, what is acceptable.
Among the few differences not discussed explicitly in the lead article is the inherently broader focus of the "equity" criterion, since it concerns social disadvantage not only because of poverty but for any reason, such as ethnicity, gender or location. In practice, disadvantage in other dimensions is often intertwined with socioeconomic disadvantage, and those who focus on "poverty" often address issues such as ethnicity, gender or other factors which make the ill-effects of poverty worse. But in some settings an exclusive focus on poverty will not take adequately into account the health disadvantages suffered by, for instance, ethnic minorities or girls and women. Proponents of the "poverty" focus may well note, however, that from a policy perspective, mixing concern for the health effects of multiple types of social disadvantage can produce a message that is too abstract, complex, and diffuse.
An "equity" focus also seems more encompassing in that it is concerned with gaps even when the worse-off are not in absolute poverty. As Gwatkin notes, in most developing countries, such a large proportion of the population lives in absolute poverty, and overall resources are so constrained, that concerns about "equity" and "poverty" are likely to be similar in practice. On the other hand, in most industrialized and some middle-income countries, a large population segment are "near" or "working" poor rather than absolutely poor, making an exclusive focus on the absolutely poor too narrow. However, a "poverty" focus could encompass concern with both relative and absolute poverty or deprivation, and so include the "near poor" or "working poor".
A concern for health differences across the socioeconomic gradient, rather than only at the poverty line, also distinguishes the "equity" approach from a focus on "poverty". However, advocates for the "poverty" approach should also find relevant the argument that living in an unequal society may be damaging to the health of everyone in it, not only its most disadvantaged members. Evidence to support this view has been accumulating for some time.
Yet another distinction between a "poverty" and an "equity" focus is that the latter makes explicit that an ethical value, namely social justice, is involved. By contrast, the need for a healthy workforce and social stability could be reasons for improving the health of the poor without invoking equity. However, in most circumstances (outside of development organizations), appealing to the self-interest of the better-off groups will be more effective in leading to policies which improve the health of the poor than appealing to an abstract notion of equity. Furthermore, despite the theoretical concerns, most individuals focusing on the "health of the poor" do so out of a commitment to equity.
Very much in line with Gwatkins conclusions, I believe that in many cases the differences between a "poverty" and an "equity" approach reflect rhetoric more than substance, and tactics rather than long-term strategy or underlying values. Both approaches call for action to improve the health of the disadvantaged. We should combine forces rather than split hairs. 
1. Ford G et al. Patterns of class inequality in health through the lifespan: class gradients at 15, 35 and 55 years in the west of Scotland. Social Science and Medicine, 1994, 39: 10371050.
2. Hart CL, Smith GD, Blane D. Inequalities in mortality by social class measured at 3 stages of the lifecourse. American Journal of Public Health, 1998, 88: 471474.
3. West P. Health inequalities in the early years: is there equalization in youth? Social Science and Medicine, 1997, 44: 833 858.
4. Mackenbach JP, Kunst, AE. Measuring the magnitude of socio-economic inequalities in health: an overview of available measures illustrated with two examples from Europe. Social Science and Medicine, 1997, 44: 757771.
5. Wagstaff A, Paci P, Van Doorslaer, E. On the measurement of inequalities in health. Social Science and Medicine, 1991, 33: 545557.