Boston Review  February/March 2000

Justice is Good for Our Health

How greater economic equality would promote public health.

Norman Daniels, Bruce Kennedy, Ichiro Kawachi

We have long known that the more affluent and better-educated members of a
society tend to live longer and healthier lives: René Louis Villermé made
this point as early as 1840, and it has been shown to hold for just about
every human society. Recent research suggests that the correlations between
income and health do not end there. We now know, for example, that
countries with a greater degree of socioeconomic inequality show greater
inequality in health status; also, that middle-income groups in relatively
unequal societies have worse health than comparable, or even poorer, groups
in more equal societies. Inequality, in short, seems to be bad for our health.

Moreover, and perhaps more surprisingly, universal access to health care
does not necessarily break the link between social status and health. Our
health is affected not simply by the ease with which we can see a
doctor--though that surely matters--but also by our social position and the
underlying inequality of our society. We cannot, of course, infer causation
from these correlations between social inequality and health inequality
(though we will explore some ideas about how the one might lead to the
other). Suffice to say that, while the exact processes are not fully
understood, the evidence suggests that there are social determinants of
health.

These social determinants offer a distinctive angle on how to think about
justice, public health, and reform of the health care system. If social
factors play a large role in determining our health, then efforts to ensure
greater justice in health care should not focus simply on the traditional
health sector. Health is produced not merely by having access to medical
prevention and treatment, but also, to a measurably greater extent, by the
cumulative experience of social conditions over the course of one¹s life.
By the time a sixty-year-old heart attack victim arrives at the emergency
room, bodily insults have accumulated over a lifetime. For such a person,
medical care is, figuratively speaking, "the ambulance waiting at the
bottom of the cliff." Much contemporary discussion about reducing health
inequalities by increasing access to medical care misses this point. We
should be looking as well to improve social conditions--such as access to
basic education, levels of material deprivation, a healthy workplace
environment, and equality of political participation--that help to
determine the health of societies.

These conditions have unfortunately been virtually ignored within the
academic field of bioethics, and in public discussions about health care
reform. Academic bioethics is quick to focus on exotic new technologies and
the vexing questions they raise for doctors and health administrators, who
must make decisions about patient care and the allocation of scarce medical
resources. And we all worry about the doctor-patient relationship under
managed care, as insurance companies have taken a newly aggressive role in
making medical decisions. But with some significant exceptions neither
academic nor popular discussion has looked "upstream," past the new
technologies, managed care, and the organization of health insurance, to
the social arrangements that determine the health achievement of societies.

We hope to fill this gap by exploring some broader issues about health and
social justice. To avoid vague generalities about justice, we shall advance
a line of argument inspired principally by the theory of "justice as
fairness" put forth by the philosopher John Rawls.1 We find Rawls¹s theory
compelling as an account of justice quite apart from its usefulness as an
approach to the health care issue. But even those who do not share our
ideas about justice may find our argument a helpful first step in thinking
about social justice and public health.

Rawls¹s theory of justice as fairness was not designed to address issues of
health care. He assumed a completely healthy population, and argued that a
just society must assure people equal basic liberties, guarantee that the
right of political participation has roughly equal value for all, provide a
robust form of equal opportunity, and limit inequalities to those that
benefit the least advantaged. When these requirements of justice are met,
Rawls argued, we can have reasonable confidence that others are showing us
the respect that is essential to our sense of self-worth.

Recent empirical literature about the social determinants of health
suggests that the failure to meet Rawlsian criteria for a just society is
closely related to health inequality. The conjecture we propose to explore,
then, is that by establishing equal liberties, robustly equal opportunity,
a fair distribution of resources, and support for our self-respect--the
basics of Rawlsian justice--we would go a long way to eliminating the most
important injustices in health outcomes. To be sure, social justice is
valuable for reasons other than its effects on health. And social reform in
the direction of greater justice would not eliminate the need to think hard
about fair allocation of resources within the health care system. Still,
acting to promote social justice may a key step toward improving our health.

Social Determinants of Health Let¹s take a closer look at some of the
central findings in the recent literature on the social determinants of
health, each of which has implications for an account of justice and health
inequalities.

Cross-national inequalities A country¹s prosperity is related to its
health, as measured, for example, by life expectancy: in richer countries
people tend to live longer. This well-established finding suggests a
natural ordering of societies along some fixed path of economic
development: as a country or region develops economically average
healthimproves.

But the evidence suggests that things are more complicated. Figure 1 shows
the relationship between the wealth of nations, as measured by per capita
gross domestic product (GDPpc), and the health of nations, as measured by
life expectancy. Clearly, GDPpc and life-expectancy are closely associated,
but only up to a point. The relationship levels off when GDPpc reaches
$8,000 to $10,000; beyond this threshold, further economic advance buys
virtually no further gains in life expectancy. This leveling effect is most
apparent among the advanced industrial economies (see Figure 2), which
largely account for the upper tail of the curve in Figure 1.

Closer inspection of these two figures shows some startling discrepancies.
Though Cuba and Iraq are equally poor (each has a GDPpc of about $3,100),
life-expectancy in Cuba exceeds that in Iraq by 17.2 years. The difference
between the GDPpc for Costa Rica and the United States is enormous (about
$21,000), yet Costa Rica¹s life-expectancy exceeds that of the United
States (76.6 to 76.4). In fact, despite being the richest nation on the
globe, the United States performs rather poorly on major health indicators.

Taken together, these observations show that the health of nations may
depend, in part, on factors other than wealth. Culture, social
organization, and government policies also help determine population
health, and variations in these factors may explain many of the differences
in health outcomes among nations.

Relative income One especially important factor in explaining the health of
a society is the distribution of income: the health of a population depends
not just on the size of the economic pie, but on how the pie is shared.
Differences in health outcomes among developed nations cannot be explained
simply by the absolute deprivation associated with low economic
development--lack of access to the basic material conditions necessary for
health such as clean water, adequate nutrition and housing, and general
sanitary living conditions. The degree of relative deprivation within a
society also matters.

Numerous studies have provided support for this relative-income hypothesis,
which states, more precisely, that inequality is strongly associated with
population mortality and life-expectancy across nations. To be sure,
wealthier countries generally have higher average life expectancy. But rich
countries, too, vary in life-expectancy (see the tail of Figure 1), and
that variation dovetails with income distribution. Wealthy countries with
more equal income distributions, such as Sweden and Japan, have higher
life-expectancies than does the United States, despite their having lower
per capita GDP. Likewise, countries with low GDPpc but remarkably high
life-expectancy, such as Costa Rica, tend to have a more equitable
distribution of income.2

We find a similar pattern when we compare states within the United States.
Wealthier states typically have lower mortality rates. But if we control
for differences in state wealth, income inequality accounts for about 25
percent of the between-state variation in age-adjusted mortality rates.
Furthermore, a recent study across US metropolitan areas found that areas
with high income inequality had an excess of death compared to areas with
low inequality. This excess was very large, equivalent in magnitude to all
deaths due to heart disease.3

Most of the evidence for this pattern comes from cross-sectional studies,
which compare different places (countries, states, metropolitan areas) at a
single point in time. But longitudinal studies, which look at a single
place over time, support similar conclusions. Widening income differentials
in both the United States and the United Kingdom have coincided with a
slowing down of improvements in life-expectancy. In many of the poorest
areas of the United Kingdom, the mortality rate for several cohorts of
relatively young people has increased as income inequality widened. In the
United States between 1980 and 1990, states with the highest income
inequality showed slower rates of improvement in average life-expectancy
than did states with more equitable income distributions.4

Individual SES Finally, when we move from comparing whole societies to
comparing their individual members, we find, once more, that inequality is
important. At the individual level, numerous studies have documented what
has come to be known as the socioeconomic gradient: at each step along the
socioeconomic ladder, we see improved health outcomes over the rung below.
This suggests that differences in health outcomes are not confined to the
extremes of rich and poor, but are observed across all levels of
socioeconomic status.5

Moreover, the SES gradient does not appear to be explained by differences
in access to health care. Steep gradients have been observed even among
groups of individuals, such as British civil servants, who all have
adequate access to health care, housing, and transport.6

The slope of the gradient varies substantially across societies. Some
societies show a relatively shallow gradient in mortality rates: being
better off confers a health advantage, but not so large an advantage as
elsewhere. Others, with comparable or even higher levels of economic
development, show much steeper gradients. The slope of the gradient appears
to be fixed by the level of income inequality in a society: the more
unequal a society is in economic terms, the more unequal it is in health
terms. Moreover, middle income groups in a country with high income
inequality typically do worse in terms of health than comparable or even
poorer groups in a society with less income inequality. We find the same
pattern within the United States when we examine state and metropolitan
area variations in inequality and health outcomes.7

Pathways Earlier, we cautioned that correlations between inequality and
health do not necessarily imply causation. Still, there are plausible and
identifiable pathways through which social inequalities appear to produce
health inequalities. In the United States, the states with the most unequal
income distributions invest less in public education, have larger uninsured
populations, and spend less on social safety nets. The facts on educational
spending and educational outcomes are especially striking: controlling for
median income, income inequality explains about 40 percent of the variation
between states in the percentage of children in the fourth grade who are
below the basic reading level. Similarly strong associations are seen for
high school drop-out rates. It is evident from these data that educational
opportunities for children in high-income-inequality states are quite
different from those in states with more egalitarian distributions. These
effects on education have an immediate impact on health, increasing the
likelihood of premature death during childhood and adolescence (as
evidenced by the much higher death rates for infants and children in the
high inequality states). Later in life, they appear in the SES gradient in
health.

When we compare countries, we also find that differential investment in
human capital--in particular, education--is a strong predictor of health.
Indeed, one of the strongest predictors of life-expectancy among developing
countries is adult literacy, particularly the disparity between male and
female adult literacy, which explains much of the variation in health
achievement among these countries after accounting for GDPpc. For example,
among the 125 developing countries with GDPpcs less than $10,000, the
difference between male and female literacy accounts for 40 percent of the
variation in life-expectancy after factoring out the effect of GDPpc. The
fact that gender disparities in access to basic education drives the level
of health achievement further emphasizes the role of broader social
inequalities in patterning health inequalities. Indeed, in the United
States, differences between the states in women¹s status--measured in terms
of their economic autonomy and political participation--are strongly
correlated with higher female mortality rates.

These societal mechanisms--for example, income inequality leading to
educational inequality leading to health inequality--are tightly linked to
the political processes that influence government policy. For example,
income inequality appears to affect health by undermining civil society.
Income inequality erodes social cohesion, as measured by higher levels of
social mistrust and reduced participation in civic organizations. Lack of
social cohesion leads to lower participation in political activity (such as
voting, serving in local government, volunteering for political campaigns).
And lower participation, in turn, undermines the responsiveness of
government institutions in addressing the needs of the worst-off. States
with the highest income inequality, and thus lowest levels of social
capital and political participation, are less likely to invest in human
capital and provide far less generous social safety nets.8

In short, the case for social determinants of health is strong. What are
the implications of this fact for ideas of justice?

Inequalities and Inequities When is a health inequality between two groups
"inequitable"? Margaret Whitehead and Goran Dahlgren have suggested a
useful and influential answer: health inequalities count as inequities when
they are avoidable, unnecessary, and unfair.9

The Whitehead/Dahlgren analysis is deliberately broad. Age, gender, race,
and ethnic differences in health status exist independent of the
socioeconomic differences we have been discussing, and they raise distinct
questions about equity. For example, should we view the lower life
expectancy of men compared to women in developed countries as an inequity?
If it is rooted in biological differences that we do not know how to
overcome, then it is unavoidable (and therefore not an inequity). This is
not an idle controversy: taking average, rather than gender-differentiated,
life expectancy in developed countries as a benchmark will yield different
estimates of the degree of inequity women face in some developing
countries. In any case, the analysis of inequity is only as good as our
understanding of what is avoidable or unnecessary.

The same point applies to judgments about fairness. Is the poorer health
status of groups whose members smoke and drink heavily unfair? We may be
inclined to say it is not unfair, provided that participation in such risky
behaviors is truly voluntary. But if many people in a cultural group or
class behave similarly, then the behavior might acquire the qualities of a
social norm--in which case we might wonder just how voluntary the behavior
is (and therefore how much responsibility we should ascribe to them for
it). Whitehead¹s and Dahlgren¹s terms leave us with an unresolved
complexity of judgments about responsibility, and, as a result, with
disagreements about fairness and avoidability.

The poor in many countries lack access to clean water, sanitation, adequate
shelter, basic education, vaccinations, and prenatal and maternal care. As
a result of some, or all, of these factors, infant mortality rates for the
poor exceed those of the rich. Since social policies could supply the
missing determinants of infant health, these inequalities are avoidable.

Are these inequalities also unfair? Most of us would think they are,
perhaps because we believe that policies that create and sustain poverty
are unjust, and perhaps also because we object to social policies that
compound economic poverty with lack of access to the determinants of
health. The problem of justice in health care becomes more complicated,
however, when we remember one of the basic findings from the literature on
social determinants: we cannot eliminate health inequalities simply by
eliminating poverty. Health inequalities persist even in societies that
provide the poor with access to all standard public health and medical
services, as well as basic income and education health, and they persist as
a gradient of health throughout the social hierarchy, not just between the
very poorest groups and those above them.

What, then, are we to think of the health inequalities that would persist,
even if poverty were eliminated? To eliminate health inequalities, should
we eliminate all socioeconomic inequalities? We might believe that all
socioeconomic inequalities, or at least all inequalities we did not freely
choose, are unjust--but very few embrace such a radical egalitarian view.
Indeed, we may well believe that some degree of socioeconomic inequality is
unavoidable, or even necessary, and therefore not unjust. On issues of this
kind, we should take guidance from a well-articulated account of social
justice--the one put forth by John Rawls.

Justice as Fairness In A Theory of Justice,Rawls sought to show that a
social contract designed to be fair to free and equal people would lead to
equal basic liberties and equal opportunity, and would permit inequalities
only when they work to make the worst-off groups fare as well as possible.
Though Rawls¹s account was devised for the most general questions of social
justice, it also provides a set of principles for the just distribution of
the social determinants of health.

Rawls did not talk about disease or health in his original account. To
simplify the construction of his theory, he assumed that his contractors
were fully functional over a normal life span--no one becomes ill or dies
prematurely. This idealization provides a clue about how to extend this
theory to the real world of illness and premature death. The goal of public
health and medicine is to keep people as close to the idealization of
normal functioning as possible under reasonable resource constraints.
Maintaining normal functioning, in turn, makes a limited but significant
contribution to protecting the range of opportunities open to individuals.
So one might see the distribution of health care as governed by a norm of
fair equality of opportunity.

We can now say more directly why justice, as described by Rawls¹s
principles, is good for our health.

Let us start by considering what a just society would require with regard
to the distribution of the social determinants of health. In such an ideal
society, everyone is guaranteed equal basic liberties, including the right
to participate in politics. In addition, there are safeguards aimed at
assuring for all, whether richer or poorer, the worth or value of those
rights. Since, as we argued above, there is evidence that political
participation is a social determinant of health, the Rawlsian ideal assures
institutional protections that counter the usual effects of socioeconomic
inequalities on participation--and thus on health.

Moreover, according to Rawls, justice requires fair equality of
opportunity. This principle condemns discriminatory barriers and requires
robust measures aimed at mitigating the effects of socioeconomic
inequalities and other contingencies on opportunity. In addition to
equitable public education, such measures would include the provision of
developmentally appropriate day care and early childhood interventions
intended to promote the development of capabilities independently of the
advantages of family background. Such measures match, or go beyond, the
best models of such interventions currently in place, such as European
efforts at day care and early childhood education. We also note that the
strategic importance of education for protecting equal opportunity has
implications for all levels of education, including access to graduate and
professional education.

The equal opportunity principle also requires extensive public health,
medical, and social support services aimed at promoting normal functioning
for all. It even provides a rationale for the social costs of reasonable
accommodation to incurable disabilities, as required by the Americans with
Disabilities Act. Because the equal opportunity principle aims at promoting
normal functioning for all as a way of protecting opportunity for all, it
at once aims at improving population health and the reduction of health
inequalities. Obviously, this focus requires provision of universal access
to comprehensive health care, including public health, primary health care,
and medical and social support services.

To act justly in health policy, we must have knowledge about the causal
pathways through which socioeconomic (and other) inequalities work to
produce differential health outcomes. Suppose we learn, for example, that
workplace organization induces stress and a loss of control, and that these
tend, in turn, to promote health inequalities. We should then think of
modifying those features of work place organization in order to mitigate
their negative effects on health as a public health requirement of the
equal opportunity approach.

Finally, a just society restricts allowable inequalities in income and
wealth to those that benefit the least advantaged. The inequalities allowed
by this principle--in conjunction with the principles assuring equal
opportunity and the value of political participation--are probably more
constrained than those we observe in even the most industrialized
societies. If so, just inequalities would produce a flatter gradient of
health inequality than we currently observe in even the more extensive
welfare systems of Northern Europe.

In short, Rawlsian justice--though not devised for the case of
health--regulates the distribution of the key social determinants of
health, including the social bases of self respect. There is nothing about
the theory that should make us focus narrowly on medical services. Properly
understood, justice as fairness tells us what justice requires in the
distribution of all socially controllable determinants of health.

We still face a theoretical issue of some interest. Even if a just
distribution of the determinants of health flattens health gradients
further than what we observe in the most egalitarian, developed countries,
we must still expect a residue of health inequalities: people who are less
well-off in economic terms will continue to be less healthy. Should we aim
to reduce further those otherwise justifiable economic inequalities because
of the inequalities in health status they create?

Suppose we reduce socioeconomic inequalities, and thereby reduce health
inequalities--but the result is that the health of all is worsened because
productivity is reduced so much that important institutions are undermined.
That is not acceptable. Our commitment to reducing health inequality should
not require steps that threaten to make health worse off for those with
less-than-equal health status. So the theoretical issue reduces to this:
would it ever be reasonable to allow some health inequality in order to
produce some non-health benefits for those with the worst health prospects?

We know that in real life people routinely trade health risks for other
benefits. They do so when they commute longer distances for a better job or
take a ski vacation. Trades of this kind raise questions of fairness. For
example, when is hazard pay a benefit workers gain only because their
opportunities are unfairly restricted? When is it an appropriate exercise
of their autonomy? Some such trades are unfair; others will only be
restricted by paternalists.

Rawls gave priority to the principle of protecting equal basic liberties
because he believed that once people achieve some threshold level of
material well being, they will not trade away the fundamental importance of
liberty for other goods. Making such a trade might deny them the liberty to
pursue their most cherished ideals, including their religious beliefs,
whatever they turn out to be. Can we make the same argument about trading
health for other goods?

There is some plausibility to the claim that rational people should refrain
from trading their health for other goods. Loss of health may preclude us
from pursuing what we most value in life. We do, after all, see people
willing to trade almost anything to regain health once they lose it.

Nevertheless, there is also strong reason to think this priority is not
clear-cut, especially where the trade is between a risk to health and other
goods that people highly value. Refusing to allow any (ex ante) trades of
health risks for other goods, even when the background conditions on choice
are otherwise fair, may seem unjustifiably paternalistic, perhaps in a way
that refusals to allow trades of basic liberties is not.

We propose a pragmatic route around this problem. Fair equality of
opportunity is only approximated even in an ideally just system, because we
can only mitigate, not eliminate, the effects of family and other social
contingencies. For example, only if we were willing to violate widely
respected parental liberties could we intrude into family life and "rescue"
children from parental values that arguably interfere with equal
opportunity. Similarly, though we give a general priority to equal
opportunity over the Difference Principle, we cannot achieve complete
equality in health any more than we can achieve completely equal
opportunity. Justice is always rough around the edges.

Suppose, then, that the decision about trade-offs is made by the
legislature in a democratic society where everyone has a fair chance to
participate. Because those principles require effective political
participation across all socioeconomic groups, we can suppose that groups
most directly affected by any trade-off decision have a voice in the
decision. Since there is a residual health gradient, groups affected by the
trade-off include not only the worst off, but those in the middle as well.
A democratic process that involved deliberation about the trade-off and its
effects might be the best we could do to provide a resolution of the
unanswered theoretical question.

In contrast, where the fair value of political participation is not
adequately assured--and we doubt it is so assured in even our most
democratic societies--we have much less confidence in the fairness of a
democratic decision about how to trade health against other goods. It is
much more likely under actual conditions that those who benefit most from
the inequalities--that is, those who are better off--also wield
disproportionate political power and will influence decisions about
trade-offs to serve their interests. It may still be that the use of a
democratic process in non-ideal conditions is the fairest resolution we can
practically achieve, but it still falls well short of what an ideally just
democratic process involves.

If we were to achieve a just distribution of resources, then, with the
least well-off being as well off as possible, there would still be health
inequalities. But decisions about whether to reduce those inequalities even
more are matters for democratic process. Justice itself does not command
their reduction.

Policy Implications We earlier suggested that the Whitehead/Dahlgren
analysis of health inequities (inequalities that are avoidable and unfair)
is useful. We then suggested that the Rawlsian account of justice as
fairness provides a fuller account of what is fair and unfair in the
distribution of the social determinants of health. The theory provides a
more systematic way to think about which health inequalities are
inequities. And it delivers the conclusion that most health inequalities
that we now observe world wide among socioeconomic and racial or ethnic
groups are "inequities" that should be remedied. Even the countries with
the shallowest health gradients, such as Sweden and England, have viewed
their own health inequalities as unacceptable and initiated policy measures
to mitigate them. Clearly, the broader WHO efforts in this direction are,
probably without exception, also aimed at true inequities.

Before saying more about the kind of reforms outside the health care system
that would improve our health, we want to head off a misconception. We are
not suggesting that we should simply ignore medical services and health
sector reform because other steps will have a bigger long-term health
payoff. Even if we had a highly just distribution of the social
determinants of health and of public health measures, people will still
become ill and need medical services. The fair design of a health system
arguably should give some extra weight to meeting actual medical needs.

To see the importance of meeting medical needs, let¹s distinguish between
"identified victims"--people who are already ill and have known needs--and
"statistical victims," whose lives would be spared illness by robust public
health measures and a fairer distribution of social determinants of health.
We might be tempted to judge these lives impartially, judging statistical
lives saved to be just as valuable or important as identified victims. But
other considerations temper our inclination to such impartial reallocation
from identified to statistical victims, and suggest that we give special
moral weight to the urgent needs of those already ill. Medical providers
may legitimately believe that the good that they can control through their
delivery of medical care has a greater claim on them than the good that
would be brought about by more indirect measures beyond their control. More
generally, many of us will be connected as family members and friends to
the identified victims and will feel that we have obligations to assist
them that supersede the obligations we have to more distant, statistical
victims.

We do not suggest, then, that our society should immediately reallocate
resources away from medicine to schools, for example, in the hope and
expectation that a better-educated population will be healthier. But the
arguments here suggest that some reallocations of resources to improve the
social determinants are justifiable.

Domestic Policy What sorts of social policies should governments pursue to
reduce health inequalities? The menu of options ought to include policies
aimed at equalizing individual life opportunities, such as investment in
basic education, affordable housing, income security, and other forms of
antipoverty policy. Though the connection between these social policies and
health may seem somewhat remote, and they are rarely linked to issues of
health in our public policy discussions, the evidence outlined earlier
suggests that they should be part of the debate. The kinds of policies
suggested by a social determinants perspective encompasses a much broader
range of instruments than would be ordinarily considered for improving the
health of the population.

Consider, then, four examples of social policies that might improve health
by reducing socioeconomic disparities: investment in early childhood
development, nutrition programs, improvements in the quality of the work
environment, reductions in income inequality, and greater political fairness.

1. Early Life Intervention.Growing evidence points to the importance of
early childhood environment in influencing the behavior, learning, and
health of individuals later in the life course. Ensuring equal opportunity
requires interventions as early in life as possible. Several studies have
demonstrated the benefits of early supportive environments for children. In
the Perry High/Scope Project, children in poor economic circumstances were
provided a high quality early childhood development program between the
ages of three and five. Compared to a control group, those in the
intervention group completed more schooling by age 27; were more likely to
be employed, own a home, and be married with children; experienced fewer
criminal problems and teenage pregnancies; and were far less likely to have
mental health problems.

Compensatory education and nutrition in the early years of life seem also
to yield important gains for the most disadvantaged groups. As part of the
War on Poverty, the federal government introduced two small compensatory
education programs: Head Start for preschoolers and Chapter 1 for
elementary school children. Evaluations of these programs indicate that
children who enroll in them learn more than those who do not. So the
program creates more equality of opportunity. Educational achievement,
meanwhile, is a powerful predictor of health in later life, partly because
education provides access to employment and income, and partly because
education has a direct influence on health behavior in adulthood, including
diet, smoking, and physical activity. So the program also leads to more
health equality.

2. Nutrition.A similarly persuasive case can be made for nutritional
supplementation in low income women and children. An analysis of the
National Maternal and Infant Health Survey found that participation of low
income pregnant women in the WIC program (the Special Supplemental
Nutrition Program for Women, Infants, and Children) was associated with
about a 40 percent reduction in the risk of subsequent infant death. A
mother¹s nutritional state affects her infant¹s chance of death not just in
the first year of life, but also throughout the life course. Thus a woman¹s
pre-pregnant weight is one of the strongest predictors of her child¹s birth
weight; in turn, low birth weight has been shown to be linked with
increased risks of coronary heart disease, hypertension, and diabetes in
later life. It follows that investing in policies that reduce early adverse
influences may produce benefits not only in the present, but also for
future generations.

3. Work Environment.We alluded earlier to the finding that the health
status of workers is closely linked to the quality of their work
environment, specifically to the amount of control and autonomy available
to workers on their jobs. Low control work environments--such as monotonous
machine-paced work (e.g., factory assembly lines) or jobs involving little
opportunity for learning and utilization of new skills (e.g., supermarket
cashiers)--tend to be concentrated among low-income occupations. Michael
Marmot and colleagues have shown that social disparities in health partly
arise as a consequence of the way labor markets sort individuals into
positions of unequal authority and control. Exposure to low-control,
high-demand job conditions are not only more common in lower status
occupations, but they also place such workers at increased risks of
hypertension, cardiovascular disease, mental illness, musculoskeletal
disease, sickness absence, and physical disability.

A growing number of case studies from around the world have concluded that
it is possible to improve the level of control in workplaces by several
means: increasing the variety of different tasks in the production process,
encouraging workforce participation in the production process, and allowing
more flexible work arrangements, such as altering the patterns of shift
work to make them less disruptive of workers¹ lives. In some cases it may
even be possible to re-design the workplace and to enhance worker autonomy
without affecting productivity, since sickness absence may diminish as a
consequence of a healthier work-place.

4. Income Redistribution.Many policies suggested by the social determinants
perspective tend to fall under the category of anti-poverty policy.
However, research on the social determinants of health warns us that
anti-poverty policies do not go far enough in reducing unjust health
disparities. Though priority should go to reducing the plight of the
worst-off, the fact is that health inequalities occur as a gradient; the
poor have worse health than the near-poor, but the near-poor fare worse
than the lower middle class, the lower middle class do worse than the upper
middle class, and so on up the economic ladder. Addressing the social
gradient in health requires action above and beyond the elimination of
poverty.

To address comprehensively the problem of health inequalities, governments
must begin to address the issue of economic inequalities directly. Evidence
we sketched earlier indicates that the extent of socioeconomic
disparities--the size of the gap in incomes and assets between the top and
bottom of society--is itself an important determinant of the health
achievement of society, independent of the average standard of living. Most
importantly, economic disparities seem to influence the degree of equality
in measures of political participation, including voting, campaign
donations, contacting elected officials, and other forms of political
activity. The more unequal the distribution of incomes and assets, the more
skewed the patterns of political participation, and consequently, the
greater the degree of political exclusion of disadvantaged groups.

Inequalities in political participation determine the kinds of policies
passed by national and local governments. For example, Kim Hill and
colleagues studied the relationship between the degree of mobilization of
lower-class voters at election time and the generosity of welfare benefits
provided by state governments. Even after adjusting for other factors that
might predict state welfare policy--the degree of public liberalism in the
state, the federal government¹s welfare cost-matching rate for individual
states, the state unemployment rate and median income, and state
taxes--robust relationships were found between the extent of political
participation by lower-class voters and the degree of generosity of state
welfare payments. In other words, who participates matters for political
outcomes, and the resulting policies have an important impact on the
opportunities for the poor to lead a healthy life.

For both the foregoing reasons--that it yields a higher level of health
achievement as well as greater political participation--the reduction of
income inequality ought to be a priority of governments concerned about
addressing social inequalities in health. Although discussion of strategies
is beyond our scope here, a number of levers do exist by which governments
could address the problem of income inequality, spanning from the radical
(a commitment to sustained full employment, collective wage bargaining, and
progressive taxation) to the incremental (expansion of the earned income
tax credit, increased child care credit, and raising the minimum wage).

International Development Our discussion has implications for international
development theory, as well as for economic choices confronted by
industrialized countries. To the extent that income distribution matters
for population health status, it is not obvious that giving strict priority
to economic growth is the optimal strategy for maximizing social welfare.
Raising everyone¹s income will improve the health status of the poor--the
trickle-down approach--but not as much as by paying attention to the
distribution of the social product. Within the developing world, a
comparison of Kerala, a state in India, with highly unequal countries such
as Brazil and South Africa illustrates this point. Despite having only
one-third to a quarter of the income of Brazil or South Africa (and thereby
having a higher prevalence of poverty in the absolute sense), the citizens
of Kerala nonetheless live longer, most likely as a result of the higher
priority that the government of Kerala accords to a fair distribution of
economic gains.

The real issue for developing countries is what kind of economic growth is
salutary. Hence Jean Dreze and Amartya Sen distinguish between two types of
successes in the rapid reduction of mortality, which they term "growth
mediated" and "support-led" processes. The former works mainly through fast
economic growth, exemplified by mortality reductions in such countries as
South Korea or Hong Kong. Their successes depended on the growth process
being wide-based and participatory (for example, full employment policies),
and on the gains from economic growth being utilized to expand social
services in the public sector, particularly health care and education.
Their experiences stand in stark contrast to the example of countries such
as Brazil, which have similarly achieved rapid economic growth, but lagged
behind in health improvements.

In contrast to growth-mediated processes, "support-led" processes--for
example, in China, Costa Rica, or Kerala--operate not through fast economic
growth, but through governments giving high priority to the provision of
social services that reduce mortality and enhance the quality of life.

Policies of either kind can succeed in promoting the health of the
population. In either case, success depends on generating a more fair
distribution of income. Once more, health is the byproduct of justice.

We noted earlier that academic bioethics and popular discussion of health
care reform has generally tended to focus on medicine at the point of
delivery and has inadequately attended to determinants of health "upstream"
from the medical system itself. Empirical findings about the social
determinants of health suggests that this is a serious mistake: upstream is
precisely where we need to look. Put these findings together with a
philosophical theory of justice that might apply to any society, and we get
this striking result. In a just society, health inequalities will be
minimized and population health status will be improved--in short, social
justice is good for our health.

------------------------------------------------------------------------

1 John Rawls, A Theory of Justice, rev. ed. (Cambridge, Mass.: Belknap
Press of Harvard University Press, 1999).

2 See Richard G. Wilkinson,Unhealthy Societies: The Afflictions of
Inequality (London: Routledge, 1996).

3 John W. Lynch et al., "Income Inequality and Mortality in Metropolitan
Areas of the United States," The American Journal of Public Health 88
(1998): 1074-1080.

4 Ichiro Kawachi, Bruce Kennedy, and Richard G. Wilkinson, Income
Inequality and Health: A Reader (New York: New Press, 1999).

5 See Douglas Black et al., Inequalities in Health: The Black Report, The
Health Divide (London: Penguin Group,1988).

6 Michael Marmot et al., "Contribution of Psychosocial Factors to
Socioeconomic Differences in Health," Milbank Quarterly 76 (1998): 403-408.

7 Bruce Kennedy et al., "Income Distribution, Socioeconomic Status, and
Self-rated Health: A US Multi-Level Analysis," British Medical Journal 317
(1998): 917-921.

8 For example, the correlation between social capital, as measured by low
interpersonal trust, and the maximum welfare grant as a percent of state
per capita income, is -.76. See Kawachi et al., "Social Capital, Income
Inequality, and Mortality," American Journal of Public Health 87 (1997):
1491-1498.

9 Goran Dahlgren and Margaret Whitehead, Policies and Strategies to Promote
Social Equality in Health (Stockholm: Institute of Future Studies, 1991).

***************************

Do Inequalities Matter?

Michael Marmot

A response to Justice is Good For Our Health, by Norman Daniels, Bruce
Kennedy, and Ichiro Kawachi.

Do inequalities in health matter? Have we not reached the end of history as
far as health in the rich (OECD) countries of the world is concerned? My
answer to both questions, like that of Daniels and colleagues, is that
inequalities in health do matter and that we have not reached the end of
history. Dramatic as have been the advances in health this century in OECD
countries, there is still a way to go. Taking the simplest summary measure,
life expectancy at birth, the United States is in the bottom half of the
rich countries.

In fact, the answer to these two questions may be linked. One reason why
countries have failed to reach their full health potential is persisting
inequalities in health. It has been known for years that there are pockets
of extreme deprivation and poor life expectancy within rich countries. In
Harlem, for example, life expectancy for young men is lower than for men in
Bangladesh. Furthermore, as Daniels and colleagues emphasize, the problem
in rich countries is not one of poor health for the deprived and good
health for the non-deprived, but of a social gradient in health. In the
Whitehall studies of British Civil Servants--white-collar workers in stable
jobs--there was a step-wise relation between grade of employment and
ill-health: the lower the grade the higher the rate of morbidity and
mortality. These findings are typical of those from national figures.

The slope of the ill-health gradient varies over time within countries and
between countries. The fact that it is not a fixed property of society
suggests that it is potentially changeable.

There are, then, at least three reasons to be concerned about inequalities
in health: pragmatic reasons, ethical reasons, and social reasons (what
inequalities in health may reflect about the wider society). In practice,
the distinction between these may be less than might first appear. First,
the pragmatic issue: Daniels, Kennedy, and Kawachi refer to evidence that
in those countries where health inequalities are greatest, overall health
status of the population is lower. It is difficult to lower the coronary
heart disease mortality of the population if only part of the population is
experiencing improvement.

Second, inequalities in health that are potentially avoidable are unfair.
Margaret Thatcher famously asserted that there is no such thing as society.
The rest of us, who think there is, may feel that social justice is a
reason for desiring a reduction in social inequalities in health.

Third, and this is at the heart of the authors¹ argument, if health is a
reflection of wider social influences, then health inequalities are a
reflection of inequalities in society. One might complain that this is not
a sufficient answer to the question of why we should be concerned with
health inequalities: it merely shifts the question further back.

Why, then, should we be concerned with inequalities in society? One could
imagine an argument that went as follows: Americans think that economic
inequalities are a good thing because they reflect economic freedoms that
are essential for wealth creation; they think that social safety nets are a
bad thing in principle. Therefore, the type of society people want is one
characterized by high inequalities of income and wealth and little spending
on social safety nets. If health inequalities happen to follow from such a
set of social arrangements that is unfortunate but not of central concern.
(Writing from Britain, I am a poor judge of what Americans think, but it is
possible that this reflects a prevalent view among the relatively small
proportion of people who actually vote in elections, if not of the large
proportion, for a democracy, who do not.)

The authors appeal to Rawls¹s theory of justice to argue that such a
society is not just, because it does not establish "equal liberties, robust
equal opportunity, a fair distribution of resources, and support for our
self respect." They argue that a just society would go a long way toward
eliminating the most important injustices in health outcomes. I agree with
their conclusions that therefore priority should be given to early life
intervention, ensuring adequate nutrition to those least able to afford it,
improving work environments, and income redistribution.

I am too much influenced by the writing of Amartya Sen, however, to accept
the appeal to Rawls without a quibble, one that may seem minor given that I
accept the conclusions they seem to have reached on the basis of their
Rawlsian analysis. Sen argues that any ethical social system requires
equality of something. The question is, what? (Or, as Sen might put it, in
which space is inequality to be measured?) Equality of economic freedoms is
one such space; equality of basic liberties, as in Rawls, is another. How
to choose between these different notions of equality? One way is with
regard to their consequences, such as health. Some philosophers coin rude
words like consequentialist to describe those who are concerned about
consequences. But Sen, critical of Rawls because of what he considered an
insufficient concern with outcomes, suggested a different evaluative
framework for assessing inequality--one that took account of its impact on
our capability and freedom to lead the lives that we want to lead.

I suspect that there is the basis for important philosophical disagreement
here that perhaps need not detain us at the moment. Even if Daniels and
colleagues do not explicitly share Sen¹s concern with the consequences of
Rawlsian justice, they are nevertheless deeply concerned with health
inequalities. Their argument is a strong one. Concern with social
inequality follows from Rawlsian analysis. That this leads Daniels,
Kennedy, and Kawachi to be concerned with the social determinants of health
and to make recommendations that would lead to the reduction of health
inequalities is all to the good.

------------------------------------------------------------------------

Pockets of Poverty

Marcia Angell

A response to Justice is Good For Our Health, by Norman Daniels, Bruce
Kennedy, and Ichiro Kawachi.

Daniels, Kennedy, and Kawachi have written an important and deeply humane
essay. They take as their starting point the well-known observation that
wealthier populations are on average healthier and that the more evenly
wealth is distributed within a population, the better the average health.
Thus, life expectancy in rich countries is higher than in poor countries,
and it is higher in countries like Sweden, where wealth is spread
relatively evenly, than it is in similarly rich countries, such as the
United States, where there are vast disparities. The authors conclude,
reasonably enough, that there are social determinants of health, and that
these determinants vary across and within populations. Less reasonably,
they conclude that there is something about inequality itself, quite apart
from poverty, that is a risk to health. On this basis, they propose that
lessening income disparities within populations would improve the overall
health of a population. Their philosophical underpinning is Rawls¹s theory
of justice, but that is hardly necessary for the case that the reforms they
suggest would make the United States a healthier, wiser, and more decent
nation.

There is no question that socioeconomic status and health are tightly
linked, and the effect of one on the other can be huge. For example, in a
study of the use of aspirin to prevent heart attacks in male physicians,
the modest benefit of aspirin was swamped by the benefits apparently
conferred by the high socioeconomic status of these men. The overall rate
of fatal heart attacks in the physicians was only 12 percent of what would
be expected in men from the general population. (To be sure, we don¹t know
whether doctors might do better than other privileged groups, such as
lawyers, because of their medical education, but even if that were true,
medical education would still have to be counted as a social determinant.)
Very few medicines or interventions can offer such a benefit.

Indeed, the fact that social advantage correlates so closely and powerfully
with health can make it extremely difficult to interpret the results of
clinical research. Studies of the effect of passive smoking on childhood
asthma, for example, are impossible to interpret unless they attempt to
control for socioeconomic status. Without such control, it is impossible to
know whether the increased prevalence of asthma in the children of smokers
is really because of passive smoking or because smokers are more likely to
be poor and poverty itself is associated with asthma for other reasons.
Similarly, studies of the effect of lead exposure on intelligence are
confounded by socioeconomic status. The children of well-educated parents
are more likely to do well on IQ tests and are also less likely to be
exposed to lead. It is hard to know, then, what causes low IQ scores: lead
or lack of parental education.

Yet, despite the undoubted importance of socioeconomic status to health, no
one knows which aspect of social standing matters--wealth or education or
occupation or some other condition--much less how it operates. We are
dealing here with a black box--the most mysterious and powerful of all
determinants of health. Differences in medical care seem to account for
only a small part of the effect, as pointed out by Daniels, Kennedy, and
Kawachi. The lion¹s share of the effect is caused by other factors, mostly
unknown. Since it is inconceivable that money in the bank or a sheepskin on
the wall could directly affect health, they must be markers for the real
factors that matter.

What might those factors be? Most good studies of the subject--and there
are lamentably few--try to control for the usual suspects, such as
cigarette smoking and heavy drinking, both of which are more frequent among
people of lower socioeconomic status. Even after controlling for them, the
health disparities across social strata persist, although they are
lessened. The increased frequencies of trauma, substance abuse, and HIV
infection among the disadvantaged cannot explain the differences, either,
since death rates from other causes, such as heart disease and cancer, are
also higher in poor people. One can imagine a host of influences that might
affect health--such as diet, stress, exposure to infectious agents or
toxins--that are related to socioeconomic status, but there is very little
evidence to point to any of them as a major cause of the health disparities
across income groups.

Some people suggest that in analyzing the association between health and
socioeconomic status, we tend to confuse cause and effect. They believe
that privilege does not lead to better health, but rather the
reverse--healthier people tend to become richer and better educated,
because they are more energetic and competitive. A variation of this view
holds that both health and wealth stem from good genes. Whatever the answer
to the question of why socioeconomic status is correlated with health, the
question deserves serious study. Good research in this area will
undoubtedly yield enormous dividends in understanding human biology and
health, and Daniels, Kennedy, and Kawachi are on solid ground in pointing
us in that direction.

They are on less solid ground in their contention that inequality somehow
contributes to poor health directly, above and beyond the effects of
poverty itself. Although there is some evidence of that from international
comparisons, it is by no means consistent. Denmark, for example, has about
the same per capita wealth as the United States with less inequality, but
its life expectancy is lower. Kerala and Costa Rica, which provide the
strongest support for a direct benefit of equality, are such outliers that
it is risky to generalize from them.

Unequal societies, by definition, have pockets of poverty and pockets of
great wealth. If the pockets of poverty contribute disproportionately to
population measures of health--such as average life expectancy--that would
explain the apparent correlation between inequality and poor health. I
believe that is the likely explanation. Inequality just seems to be a
direct contributor to poor health, whereas the real cause is poverty.
Daniels, Kennedy, and Kawachi base their argument for a direct effect of
inequality on the notion of a linear health gradient that operates equally
across all socioeconomic strata, so that the wealthy benefit as much as the
poor lose. But the evidence for that is weak. There have not been
sufficient studies of a broad enough range of income levels to know what
the shape of the curve is. The best information may come from international
comparisons showing that with increasing wealth, the health benefits become
smaller and smaller until a plateau is reached.

One need not invoke some mysterious effect of inequality on health to make
a very strong argument for lessening inequalities that lead to deprivation
at the low end of the scale. Poverty is crippling not only physically but
intellectually and spiritually. It cripples any wealthy society that
tolerates it on a large scale, as does the United States. In addition to
the loss of human potential and the social pathology that grows out of
poverty, the costs include the callousness that inures the rest of society
to its presence, even as many people enjoy extraordinary riches.

The fact that there are also health consequences of poverty, whether they
are exacerbated by inequality or not, is doubly punishing and adds greatly
to the injustice. Daniels, Kennedy, and Kawachi are right about that. F.
Scott Fitzgerald famously pointed out that the rich are different from the
rest of us. But what is less well known is that he observed that no
difference that divides people is so important as that between the well and
the sick. I agree.

------------------------------------------------------------------------
Equal Opportunity

Sudhir Anand and Fabienne Peter

A response to Justice is Good For Our Health, by Norman Daniels, Bruce
Kennedy, and Ichiro Kawachi.

In the field of public health it is common knowledge that the determinants
of people¹s health include many factors other than medical care. In the
same vein, the contemporary literature on social inequalities in health has
stressed the importance of social factors other than access to health care.
It is thus surprising that bioethics has, until recently, focused
exclusively on medical care and neglected the ethical implications of
broader social factors that impact on people¹s health. The good news is
that this is changing, and there is now significant ongoing work on the
topic of health equity. The essay by Daniels, Kennedy, and Kawachi is an
example. We agree with the authors that social inequalities in health raise
important questions about justice, but have some comments on the details of
their argument.

Daniels, Kennedy, and Kawachi define social inequalities in health as
differences in average health between socioeconomic groups. If we are
concerned about social justice, however, this definition does not go far
enough. The empirical literature shows significant health differentials
with respect to other social groups too--for instance, those defined by
gender, race and ethnicity, or geographical location. The problem is
compounded if these factors interact with socioeconomic status and, as a
result, health differentials exist within socioeconomic groups--for
example, black men with low incomes have worse health than white men with
similar incomes.

The authors use the framework developed by Margaret Whitehead and Goran
Dahlgren to link empirical research on social inequalities in health with
philosophical work on justice. According to Whitehead and Dahlgren,
inequalities in health are inequitable (unjust) if they are "avoidable,
unnecessary, and unfair." Yet, fairness surely subsumes what is unavoidable
and what is "necessary." Problems of justice and fairness only arise if a
certain outcome could have been otherwise; and if what is necessary is
interpreted to mean something other than what is unavoidable, then a
judgment on what is necessary must ultimately be made with reference to
justice and fairness. The framework thus reduces to the question: when are
health inequalities unfair or unjust, and why?

The authors base their response on Rawls¹s theory of justice as fairness.
The issue is whether the existing framework of Rawlsian justice can take
care of the problem of social inequalities in health, or whether we need to
rethink what justice requires in addressing health inequalities. The
authors do not settle this issue, and two different views of the
relationship between Rawlsian justice and social inequalities in health are
identifiable in their paper.

The first view is that although Rawls did not have the problem of social
inequalities in health in mind when he formulated justice as fairness,
implementing his principles will go a long way toward reducing such
inequalities. This argument suggests that we do not need to pay special
attention to the problem of health inequalities. By ensuring greater
justice in Rawls¹s original sense, we will--as a side-effect--also solve
the problem of health inequalities. This view relies on the premise--not
made explicit by the authors--that inequalities in health are unjust if,
and only if, they are the result of unjust social arrangements.

The second view is based on the premise that a conception of justice should
explicitly tackle problems of inequalities in health. This view relies on
an extension of the Rawlsian principle of "fair equality of opportunity."
Daniels originally developed this view in his book Just Health Care to deal
with fair access to health care. According to this account, health is
defined as normal functioning and fair equality of opportunity requires the
maintenance of normal functioning. But if broader social factors affect
people¹s health, there is no reason why the extension of the principle of
fair equality of opportunity should apply only to health care and neglect
the other determinants of health.

We thus discern two views as to how Rawlsian justice might apply in dealing
with the problem of social inequalities in health. Their simultaneous
presence in the essay creates a tension, since they each might yield a
different assessment of what should be done and for what reason. Whereas
the extended fair equality of opportunity principle seems to require
efforts to correct health inequalities as such, the first account only
addresses the problem indirectly. Moreover, the first account is contingent
on the empirical relationships that Kawachi, Kennedy, and others have
observed. Had the empirical relationships observed been the reverse--for
example, had higher income inequality been associated with smaller health
inequalities--then implementing Rawlsian justice according to the first
view could actually worsen inequalities in health. On the other hand, the
extended fair equality of opportunity view is not contingent on any such
empirical relationships, and appears to provide a stand-alone case for
redressing social inequalities in health.

The tension becomes even more apparent when the authors consider whether we
should correct those social inequalities in health that remain after we
have implemented Rawlsian justice according to the first view.
Specifically, they ask whether it would ever be "reasonable to allow some
health inequality in order to produce some non-health benefits for those
with the worst health prospects." If trade-offs between health and
non-health goods are admissible, these health inequalities may be
justifiable--provided the non-health gains under just social arrangements
compensate for the health losses. The authors suggest that decisions about
trade-offs and compensation are matters for the democratic process, and not
for justice itself to settle. But what if pervasive inequalities in health
remain--or are even exacerbated--with or without compensation through other
goods? What is needed, it seems to us, is an account that explicitly
evaluates the distribution of health outcomes and recognizes that health
inequalities raise independent problems of social justice. The extended
fair equality of opportunity principle would appear to provide the basis
for such an account. Were the authors to develop this account, though, the
first view of what Rawlsian justice requires would seem to be redundant.

Recent research on social inequalities in health does raise important
questions about justice and, in principle, we have no problem with the use
of a Rawlsian approach. To us, however, the account provided by Daniels,
Kennedy, and Kawachi remains ambiguous about the precise conception of
justice that is invoked to address the problem of social inequalities in
health.

------------------------------------------------------------------------
Policy Options

Ted Marmor

A response to Justice is Good For Our Health, by Norman Daniels, Bruce
Kennedy, and Ichiro Kawachi.

The article by Daniels, Kennedy, and Kawachi promises far more, I am
afraid, than it delivers. This is all the more disappointing since I
applaud the effort to join empirical analysis of health and health care to
normative disputes about what public policies should be enacted and
realistic discussions of what policies can be implemented. My reservations
fall into two categories. First, I am concerned about the authors¹ basic
claim that inequality per se is "bad for our health." Second, I am not
convinced that, even assuming income inequalities cause significant health
inequalities, justice requires more effort to reduce income differences
than to make access to medical care more equal. In that connection, I raise
questions about the general presumption that discovering causal pathways in
social arrangements leads directly to what the authors call "policy
implications."

Before proceeding, however, two prefatory remarks. First, I want to declare
an interest. As the co-editor of Why Some People are Healthy and Others
Not: The Determinants of Population Health,I found the public health
sections of this paper annoyingly self-referential: the main empirical
research they cite is their own. Although I am not an epidemiologist, I
spent five years meeting regularly with a number of Canadian social
scientists who wrote the basic chapters of that book. Nothing in this
article makes a large advance over the understanding that book communicated
nearly six years ago. What¹s more, there is considerable evidence that
Kennedy and Kawachi did not carefully consider its findings. More
generally, the authors¹ whole discussion seems curiously disconnected from
the extensive research that precedes their work. The Lalonde Report of
1975, the most widely distributed public document in Canadian history,
argued that, at the levels of income in societies like Canada¹s,
conventional investments in medical care were unlikely to produce big
improvements in population health. Not a word about that, let alone of the
efforts of scholars like James House of the University of Michigan to chart
the causal pathways by which "social determinants" work their health
effects, appears in their essay. And there is only a throwaway line about
what may be the most illuminating research on these questions--the
Whitehall study that Michael Marmot has pioneered. In that study, the
gradient in health outcomes, given equal exposure to risk, appeared to be
connected to the amount of control over lives that differently situated
workers have. Daniels, Kennedy, and Kawachi say little about the physical
pathways that might account for these differences--even though Marmot, and
others, have explored them.

Second, space considerations prevent my taking up the theoretical issues
raised by the authors¹ treatment of how justice claims and the social
determinants of health are linked. This is a serious topic, done with
considerable care in the article, and worthy of separate analysis. Instead,
I want to focus on a couple of key points.

How Bad Is Inequality? Have the authors made a compelling case for the
claim that "justice is good for our health"? I think not, but one needs to
begin by asking what precisely the authors are asserting. The most cautious
presumption is that inequality appears to affect the health of populations.
That, of course, is consistent with the correlations the authors present.
But there is little in the way of a rigorous defense of this claim. One
difficulty is that the precise meaning they attach to "inequality" is
unclear: sometimes it seems to mean the Gini coefficient (a standard
measure of income dispersion), sometimes (as with the section on the
Rawlsian conception of justice) it seems to be linked to the distribution
of stress and control at work.

But the most important problems, I suspect, are the technical arguments
against using correlations--whether at the national, state, or local
level--to support causal claims. Scholars such as Harold Pollack of
Michigan, Jeff Milyo of Tufts, and Ingrid Ellen of New York University all
contest the idea that, controlling for income, inequality itself
necessarily matters. As Pollack put it recently at an academic meeting,
"cross sectional regressions that use inequality measures such as Gini are
virtually uninterpretable." He goes on to say that "it is frustrating that
uncritical use of these measures is so pervasive in public health analyses
of [United States] and cross-national comparisons." Pollack¹s grounds,
which I find plausible, are straightforward: "Money matters near the bottom
of the distribution and may not matter at all for many outcomes when one
exceeds the median. Controlling for the median income, then, any income
dispersion measure is highly correlated with the percentage of the
population under the poverty line."1 So it is not dispersion itself that
matters for health, but the proportion of the population that is poor.

The connection between inequality and health, then, is far from obvious to
other analysts and this article does little to dispel the skepticism
Pollack¹s remarks exhibit. That does not mean there are no connections
between inequality and health. Rather, the connections are less obvious
than the article suggests, and their implications, for what counts as a
just society or what policies such societies should pursue, are less
compelling than presented.

Policy Implications? I turn now to the topic with which I am most familiar:
drawing policy implications from admittedly controversial empirical
findings. Assume for the moment that the article¹s central contention about
inequality and health is correct. Assume further that we know how to reduce
income inequalities, that the technology of redistribution is available and
implementable. Does that mean we should turn to inequality-reducing
policies as a matter of health policy, subordinating the claims on
resources made by modern medicine?

The authors¹ discussion of this issue is more nuanced than their treatment
of the social determinants of health. Daniels, Kennedy, and Kawachi concede
that identifiable illness commands our attention and utilitarian concerns
about "net benefits" need not always trump our humane allocation of care
for the ill. But two larger issues of public policy analysis remain. First,
the discussion of the purposes of health care policy is terribly truncated.
National health insurance finds justification not simply in efforts to
improve on measures of a population¹s health. We care about equality of
access to medical care because suffering, pain, uncertainty, and the myriad
other features of being ill or injured ought not, it is widely believed,
vary primarily with one¹s ability to pay. The fact that even in systems of
universal health insurance there are pockets of unequal response to illness
does not dispel the egalitarianaspiration--or the social cohesion and sense
of fairness--that such efforts both reflect and symbolize.

Secondly, I am concerned that the authors understate the gains that
fairness in allocating medical care has proven capable of generating, and
that they overstate the likelihood that we can do very much about more
basic determinants of social equality.

Put more generally, there is no reason to treat a theoretical possibility
as a compelling policy option unless both the worth of that aim and its
implementability dominate the alternatives. Nothing in contemporary America
suggests that we are likely to move more quickly in reducing income
inequality than we are to make health care more fairly available. The
expected value of a policy option is, in short, its idealized results times
the likelihood of achieving them. A one-in-ten chance of getting a dollar
is a lot less valuable, looked at this way, than a one-in-two chance of
getting fifty cents; indeed the latter is two-and-a-half times better.

It is worth remembering that when national health insurance was a more
prominent option in the l970s, one of the arguments against it was that
medical care outlays were wasted, that more powerful tools for improving
America¹s health were available. A quarter of a century later, Americans
enjoy longer lifespans, but with a distribution of health care that is
shameful. Those nations with a reasonably fair distribution of income
protection against the costs of illness might well gain by concentrating at
the margin on health-improving policies outside of medical care. The United
States, I contend, should address the do-able but difficult task of making
medical care more fairly distributed before taking on the more utopian task
that Daniels and colleagues suggest.

------------------------------------------------------------------------

1 Delivered at the October 1999 meetings of the American Association of
Programs in Policy Analysis and Management, in Washington, D.C.

------------------------------------------------------------------------
Political Problems

Ezekiel Emanuel

A response to Justice is Good For Our Health, by Norman Daniels, Bruce
Kennedy, and Ichiro Kawachi.

That social factors strongly influence health and life expectancy is well
established news. Indeed, the classic finding in this regard is that life
expectancy in the United States and Britain, for instance, rose
dramatically at the end of the nineteenth century. This is generally
attributed to social factors--better housing, water treatment, working
conditions, and nutrition--because effective medical interventions came
much later (surgery did not become safe until the turn of the century,
penicillin was not manufactured in large quantities until World War II,
chronic dialysis did not become available until the early 1960s,
resuscitation was first reported in 1960, and the intensive care unit did
not become a fixture in hospitals until the late 1960s). Reinforcing this
link between social factors and health outcomes are numerous studies both
in the United States and abroad of the causes for the substantial and
persistent decline in cardiovascular mortality since the 1960s. They show
that about half of the decline is due to "social factors," such as smoking
cessation and changes in diet, while 40 percent or so of the decline is
attributable to direct medical interventions, such as better control of
blood pressure, cardiac surgery, and cardiac care units.

The real issue is what to do about this information. How should the
understanding that social factors have a profound and significant impact on
health outcomes and inequalities affect research and social policies?
Daniels, Kennedy, and Kawachi suggest that we should stop concentrating on
health care and look "upstream." Our attention should be focused on
improving access to basic education, level of material deprivation, a
healthy workplace environment, and equality of political participation.
While Daniels and colleagues (strangely) single out bioethicists for
chastisement, their admonition to stop worrying about exotic new
technologies, the doctor-patient relationship, the performance of managed
care, and even the fair allocation of health care resources seems directed
to everyone--health policy experts, health care administrators,
politicians, and the general public. Their advice: if you care about
improving the health of the country, stop obsessing about "increasing
access to medical care" and campaign for social justice. This is what
Michael Marmot and others once called "the extreme version of the upstream
focus [in which] action to reduce inequalities in health should therefore
focus on the causes of social inequalities." It is strange to hear this
call at the very moment expanding access to medical care has again surfaced
in the national debate since the 1993 Clinton debacle.

Who could disagree that we should focus more attention on social justice?
Even if they had absolutely no impact on health, narrowing income
inequality in the United States, improving the educational system, and
reforming the political process to reduce the influence of money and
enhance popular participation are independently worthy goals that demand
our attention. However, linking them to health outcomes--making improving
health an important reason and motivation to advocate social justice--will
likely be ineffectual. It may well be counter-productive, at least in the
United States.

First, it is highly unlikely that Americans are going to be roused to
support improvements in social justice because such changes will (or, more
accurately, may) lead to improvements in health outcomes. Those of us
dedicated to a more just society find the American public¹s toleration of
gross--and growing--inequalities in income and political power puzzling and
frustrating. Yet this is the reality in which changes will have to be
fashioned. While Americans do not seem interested in lower taxes at the
moment, neither are they clamoring for higher marginal tax rates on the
rich; while they want campaign finance reform, it is hardly a burning issue
that will determine more than a handful of votes. The one issue of social
justice that inflames Americans is education. And this is not because it
will lead to better health outcomes, but to economic advancement; people
worry that the educational system is failing many kids, including their
own, and thereby locking them out of good jobs in the future.

As politically salient as health care is, it hardly seems as if the
American public, at least, is likely to be persuaded to support higher
marginal tax rates, campaign finance reform, or a host of other things
because these changes may narrow health inequalities or even improve their
own health. Somehow "Support higher taxes on the rich, live longer" or "Ban
soft money, improve your health" are unlikely to be persuasive or plausible
to the public.

If we want to reduce health inequalities and to improve health outcomes,
following Daniels and colleagues by focusing "upstream" and getting
bioethicists, health policy experts, and the public discussion to focus on
income inequality is likely to be even more frustrating than focusing
directly on health care has been for the last thirty years. And this is
probably not limited to the United States. Aversion to redistribution and
income equality may be more extreme in the United States than elsewhere,
but throughout the developed world the embrace is of more, rather than
less, socioeconomic inequality. Health care is unlikely to be the horse to
carry social justice measures over the finish line.

Linking health improvement too closely to social justice could actually
backfire. In the United States, health care programs have won broad support
for many years. This is not only true for funding of the National
Institutes of Health, but also for support of Medicare and other health
programs. This is in part because health care is viewed as something that
benefits everyone in society; health programs are not viewed as special
interest programs or as programs for the poor, racial minorities, or other
groups. As Rashi Fein has pointed out, the difference in support of
Medicaid and Medicare is closely linked to one being viewed as a "poor
person¹s" program and the other being a general program that also happens
to benefit the poor. Whether we like it or not, it is precisely because
health care is viewed as key to equal opportunity, without overtly or
intentionally redistributing income, that it garners such strong public
support. This is an essential foundation piece for any chance of forging a
majority to support some version of universal health coverage in the United
States.

Convincing the American public to look "upstream" and making general
redistributive efforts key to improving health is unlikely to further the
former and could well undermine the latter; resistance to redistribution is
likely to be stronger than endorsement of expanded health access. Indeed,
the more Americans are told how much re-distribution Daniels, Kennedy, and
Kawachi contemplate to secure health improvements--income inequalities less
"than those observed in even the most industrialized countries," early
childhood interventions that "go beyond the best models of such
interventions we see in European efforts," etc.--the more dismissive they
are likely to be of gazing "upstream."

So what should be done about these data? One objective is to determine the
effect of narrowing income inequality on the health of the "best off."
While it is strange to worry about the best off, if narrowing income
inequality improves overall health by raising the low end, but somehow
decreases the health of the rich, they are likely to resist. There is no
reason to think this will occur. Indeed, Medicare shows the opposite; it
produced general improvements in health delivery that benefited the
well-off. But demonstrating a "trickle up" for the health of the well-off
would undercut at least this element of opposition to social justice.

Another lesson is to design health interventions that take into account
thedifferential impact on the lower socioeconomic groups. Smoking cessation
targeted at minorities, universal prenatal care that ensures the poor are
covered, and opposing development and coverage of services that only the
rich will have access to all utilize the knowledge gained while sustaining
support for health.

Finally, bioethicists, health policy experts, and others should keep the
focus on universal access and the just allocation of health care resources.
Most Americans are dissatisfied with the current health care system; what
is needed is an alternative a majority can endorse. While this may not have
the same total impact on improving health outcomes as substantial income
redistribution, health care services still account for 25 to 40 percent of
improvements in health outcomes. This is substantial, worth securing for
everyone, and will enhance social justice. And it is part of the current
public discussion.

Admonishing caution in shifting the perspective of bioethicists, health
policy experts, and the public "upstream" should be construed as neither an
argument against greater efforts at social justice nor a dismissal of the
importance of the social determinants of health, but as a warning not to
forsake attention to greater access to health care.

------------------------------------------------------------------------
First Contact

Barbara Starfield

A response to Justice is Good For Our Health, by Norman Daniels, Bruce
Kennedy, and Ichiro Kawachi.

 No one who is concerned with justice as a basic principle would deny the
value of income redistribution, particularly in view of the powerful
relationship between income inequality and ill health. The policy question
turns on issues of feasibility, practicality, and time frame. What might be
complementary strategies while the struggle for income redistribution takes
place?

Most people know that the United States has the most costly health system
in the world, but few realize that the health of the US population is worse
than than the the populations of many other industrialized nations. The
only major health indicator for which the United States is not near the
bottom of the rankings is life expectancy at age 80 or older. In the
general scheme of things, there are many determinants of ill health
(biological, social, psychological, environmental, genetic) and they are
likely to act in ways that differ in different population subgroups. While
more doctors, more hospitals, and more technology rarely, if ever, produce
better health of populations, certain aspects of health systems do have a
positive impact on health. That is, relatively new health policy studies
show that health systems that are better oriented to primary care, with
specialty care serving primary care rather than the other way around,
achieve better health. As a strategy for achieving equity in health
services delivery primary care is generally associated with other
progressive political approaches, such as more equitable tax policies and
better distribution of income. But even then, studies have shown that areas
that are better endowed with primary care physicians and less well endowed
with specialists have better health as measured by a wide variety of health
indicators, regardless of the degree of income inequality. This is the case
both internationally and within all fifty US states. Still, the strength of
the relationship between primary care resources and health varies across
different population subgroups, indicating that other determinants also
play an important role.

Most people in the United States think that having free access to
specialists assures them the best quality of care, thus reflecting an
unmitigated faith in the power of medical and surgical tests and
procedures. Although access to specialists is important for those who need
it, unrestrained access to specialists is potentially dangerous.
Unnecessary technology, which is more often applied by specialists than by
primary care physicians, can be harmful to health. So are many medications.
Estimates of the unanticipated adverse effects of technology and
medications, along with the adverse effects due to errors in their
administration, account for somewhere between the third and fourth leading
cause of death in the United States.

A long-term relationship with a primary care practitioner can help people
decide when specialty care is not really needed, thus reducing the ill
effects of non-indicated interventions. Public realization that the goal to
improve primary care, including appropriate referral to specialists when
indicated, is critical to improving the health of the population. It is no
accident that the elderly have the best health status (relative to other
countries): they are the only segment of the population with continuous
assured financial access to health services since 1965, which made it
possible to build long-term relationships with primary care physicians.

Income redistribution may go a long way to improving health, but there will
also have to be simultaneous attention to changing other social and health
policies. As one pundit said, "For every complex problem, there is a simple
solution, and it is wrong." There is no simple solution to reducing
systematic health inequalities. A policy reorientation that recognizes the
importance of universal access to high quality primary care services backed
up, when indicated, by appropriate specialty care resources, is a critical
part of the strategy. Managed care has derailed a national focus on
building a strong primary care infrastructure by pretending to be organized
around primary care. In its focus on profit-making and cost-cutting,
managed care, in its current incarnation, fails to fulfill any of the
important functions of good primary care.

Unfortunately, a poorly informed public focuses on the ills of managed care
without understanding how managed care got here in the first place. People
want direct access to specialists, in the mistaken belief that this will
improve their health, without any recognition that they risk harm to health
from over-use of potentially dangerous interventions. The evidence is clear
that the best way to achieve better health is to greatly enhance the
contributions of primary care, which focuses on meeting and solving
people¹s health needs, including appropriate referrals when they are
indicated. A more informed public and better public policy will have to be
marshaled to address the many possible approaches to reducing the
systematic disparities in health across population subgroups and the
relatively dismal health status of this nation. Income redistribution is
important, but it is unlikely to happen any time soon. In the meantime,
other strategies, including improving the equity-enhancing aspects of
health systems, are likely to be more practical and feasible.

------------------------------------------------------------------------
A Health Agenda

Emmanuela Gakidou, Julio Frenk, and Christopher Murray

A response to Justice is Good For Our Health, by Norman Daniels, Bruce
Kennedy, and Ichiro Kawachi.

The paper by Daniels, Kennedy, and Kawachi is an important piece from the
perspectives of social epidemiology, ethics, and policy. It raises critical
issues about the determinants of health inequalities and proposes policies
that might contribute to their reduction.

We at the World Health Organization (WHO) give great importance to the
reduction of health inequality. In a new framework for the assessment of
health system performance proposed by WHO, reducing health inequality is
one of four main goals for health systems. The other three goals are
improving health status, enhancing the responsiveness of the health system
to the legitimate expectations of the population, and protecting people in
a fair manner from the financial consequences of caring for health. By
explicitly listing the reduction of health inequality as one of the
intrinsic goals of health systems, WHO illustrates the prominence that this
problem should receive in the health policy agenda.

Daniels, Kennedy, and Kawachi make four policy recommendations for the
reduction of health inequalities--all of which constitute social policies
that do not primarily involve the health sector. They suggest that "to
address comprehensively the problem of health inequalities governments must
begin to address the issue of economic inequalities." We agree that
economic redistribution policies are intrinsically important, independent
of their effect on the reduction of health inequalities, and as such should
constitute good social policies regardless of the degree of health
inequalities present in a society. Similarly, we consider health to be
intrinsically important, independent of its association with other
components of well-being.

When it comes to health, specialized entities, both at the international
level (e.g., the WHO) and the national level (e.g., various ministries of
health), have great potential to influence policies aimed at the reduction
of health inequalities. They also have the capacity to be involved in
inter-sectoral approaches aimed at improving specific determinants of
health. But these entities can only do so much. For example, it may be
possible to convince ministries of finance to raise taxes on tobacco, yet
ministries of health typically have little capacity to influence broad
economic redistribution policies. Therefore, these actors typically
concentrate on within-the-health-sector approaches and on intersectoral
initiatives to improve specific determinants of health.

Before making explicit policy recommendations we need to be better informed
about how health systems of various countries perform in the achievement of
their goals. Once we have a better understanding of the factors influencing
the performance of a health system we will be better equipped to articulate
policies that will lead to the reduction of inequalities.

Health Inequality We define health inequality to be the differences in
health across individuals in a population. We are using the individual as
the unit of analysis and are interested in studying the inequality in the
distribution of health. We propose to use health expectancy as the measure
of health. Health expectancy--the number of years that an individual born
today is expected to live in the equivalent of full health--reflects the
risk of mortality and the risk of non-fatal health outcomes that an
individual faces at each age. It is important that the measure of health
reflects not only a risk of death but also the risk of being in ill-health.

Before we try to measure inequality of health expectancy, though, we first
ask what, if any, components of health expectancy are either not amenable
to change or arise from fully informed choices of individuals to decrease
their health expectancy through the pursuit of risky activities. If there
are differences that could never be remedied by intervention or new
technology, one might argue that we should be uninterested in them. But
what component of the distribution of health expectancy is not amenable to
intervention? That due to genes? That due to chance during birth? In both
cases, the argument that we cannot intervene to change the effects on the
distribution of health expectancy seems specious. There is little evidence
of significant cross-population variation in the contribution of genes. And
with current improvements in technology and future progress, it is likely
that even genes will become amenable to change.

What about volition? How much of the distribution of health expectancy for
a population is due to fully informed choices of individuals who have a
taste for risky behavior? This seems like a very slippery slope. What
choices affecting health are fully informed? Would we exclude the effects
of tobacco on health expectancy because smoking is a choice? Even if we
claim that the choice was informed, should it be excluded? We argue that it
should not be excluded. First, in most cases health risks are not adopted
because of a love of risky behavior but rather for other, less informed,
reasons.1 Second, the true volitional component of the distribution of
health expectancy is likely to be very small and can well be ignored. This
argument is similar to ones used to explain certain measures of income
inequality, where the variation in the distribution of income due to
different trade-offs between leisure and income within the population is
routinely ignored in the measurement of income inequality.

Finally we ask, how can health expectancy be measured? Risk, after all, is
not observed; only outcomes are. But the distribution of health risks can
be reasonably approximated through a variety of techniques. Together, they
allow us to measure the distribution of four key dimensions: child
mortality risk, adult mortality risk, life expectancy and health expectancy
through small area analyses, and non-fatal health outcomes.

1. Child Mortality Risk.We can observe the variation in the proportion of a
mother¹s children who have died, which provides information at a very fine
level of aggregation (namely households) on the distribution of child death
risk. Using simulation, we can evaluate the difference in the distribution
of outcomes from that which would be expected based on a distribution of
equal risk. Data on children ever born and children surviving for women of
different ages are widely available from the Living Standards Measurement
Studies (LSMS), the Demographic and Health Surveys (DHS) and many censuses
and surveys. We have implemented this strategy for measuring child mortality.

2. Adult Mortality Risk.We do not have good data to measure the
distribution of adult mortality. Information on the survivorship of
siblings could in principle be used but it would refer to average mortality
experience over decades and the technical challenges have yet to be solved.
Other strategies need to be developed.

3. Life Expectancy or Health Expectancy for Groups.We can divide the
population into groups that are expected to have similar health
expectancies and measure directly the health expectation for those groups.
Inevitably, this will underestimate the distribution of health expectancy.
The more refined the groupings are, the more we will approximate the true
underlying distribution. Small-area analyses hold out the promise of being
one of the most refined methods for revealing the underlying distribution
of health expectancy in a population. For example, a detailed age-sex-race
group analysis of counties in the United States has revealed a range in
life expectancy across counties of 41.3 years--almost as large as the range
across all countries of the world.

4. Non-Fatal Health Outcomes.Measurement of non-fatal health outcomes on
continuous or polychotomous scales provides more information from which to
estimate the distribution of risk across individuals. Numerous surveys
provide information on self-reported health status using a variety of
instruments. The main problem to date with this information is the
comparability of the responses across different cultures, levels of
educational attainment, and incomes. For example, the rich often report
worse non-fatal health outcomes than the poor. Problems of comparability
must be resolved before such data sets can be used to contribute to
estimation of health expectancy in the population.

For the WHO, the way forward will be to simultaneously pursue the
development of methods and data sets to measure these different dimensions
of the distribution of health expectancy. We recognize that there is a
great need for new methods to integrate these different measurements into
one estimation of the distribution of health expectancy in populations.
Based on the wide array of measures used to summarize the distribution of
income several measures of the distribution of health expectancy can be
developed.

Determinants of Health Inequality The measurement of health inequalities
across countries is a crucial step towards a better understanding of its
determinants. Once the performance of the health system has been assessed
with respect to each goal, including the reduction of health inequalities,
we intend to proceed with analytical work on the determinants of performance.

The study of the determinants of health and its distribution will involve
those socioeconomic variables that are likely to play an important role.
The relationship between socioeconomic status and health is a complex one
and we would like to differentiate among the following four interactions,
each of which is very important and needs to be better understood: 1) how
the average level of socioeconomic status affects the average level of
health in a population; 2) how the average level of socioeconomic status
influences the distribution of health; 3) how the distribution of
socioeconomic status affects the average health; and 4) how the
distribution of socioeconomic status influences the distribution of health
in a population.

The framework presented above will provide us with the opportunity to study
these relationships, since individuals have been used as the unit of
analysis in the measurement of health and health inequality.
Individual-level data provide us with a greater capacity to analyze this
complex relationship than aggregate-level data. This approach has been
criticized by those looking at differences in health status across social
groups for ignoring the important relationship between socioeconomic status
and health. On the contrary, our proposed measurement strategy will use
socioeconomic status and its distribution as potential determinants of
health inequality. When individuals are grouped by a socioeconomic variable
and differences in health status across these groups are reported, the a
priori assumption is made that the variable used to group individuals is
the most important determinant of health inequality. Our approach does not
make any a priori assumptions in the measurement of health inequality but
uses potential determinants, including socioeconomic status, as explanatory
variables. Both approaches, however, recognize that socioeconomic status
and its distribution may be powerful determinants of health inequality.

The reduction of health inequalities is a key goal of health systems. Once
the complex web of determination of these inequalities is better
understood, there will be a pressing need for policies aimed at reducing
them. The health system will play a central role in the formulation and
implementation of these policies, whereby efforts directed specifically at
health care institutions will have to be accompanied by initiatives
involving other sectors.

------------------------------------------------------------------------

1 The cost of being fully informed about the health consequences of
different choices often is prohibitively high. Most individuals are forced
to make choices with incomplete or incorrect information. When the choice
to take on risk and the outcome are separated in time, the rate at which
individuals discount the future can profoundly influence choices about health.

------------------------------------------------------------------------
Lost in Translation

Steffie Woolhandler and David Himmelstein

A response to Justice is Good For Our Health, by Norman Daniels, Bruce
Kennedy, and Ichiro Kawachi.

In the 1920s, a major current of Soviet health planning contemplated the
withering away of curative medicine. Many believed that as socialism
evolved into communism--eliminating inequality, poverty, alienation, and
oppression--prevention, in the broadest sense, would obviate the need for
doctors and hospitals.

This Soviet view was an extreme, caricatured version of a long, more
holistic socialist tradition. Rudolph Virchow (the founder of modern
pathology) and Friedrich Engels analyzed the social determinants of disease
in Silesia and the British working class, respectively. Early in his
career, Salvador Allende (then minister of health in a Popular Front
government) described the social origin of disease and suffering in his
book La Realidad Medico-Social Chilena, and concluded that only broad
structural change in Chilean society could adequately address health problems.

The Marxist tradition has delineated a socialized biology; patterns of
health, disease, even physiology that are shaped in interaction with a
specific social environment. As Richard Levins has pointed out, there is a
"late twentieth century capitalist" pancreas, not in the sense of a
particularly wealthy organ, but rather an organ stressed to the point of
diabetes by a variety of socially determined factors: patterns of diet
dictated by agribusiness; living and school environments antagonistic to
exercise; work situations that constrain meal schedules and physical
activity; and a profit-driven health care system that fails to embrace
prevention.

Recent work on inequalities in health, which forms the empirical foundation
for Daniels, Kennedy, and Kawachi¹s piece, is a statistical restatement and
verification of this tradition: Virchow and Engels¹ prose descriptions are
being translated into the modern scientific language of epidemiology. Such
translation is a great service; it presents alien and suspect ideas in
acceptable academic format. Concern over social and health inequality has
become a legitimate focus for academic work, encouraging altruism among
colleagues and students, and providing ammunition for progressive
reformers. Press attention has followed, and with it, pressure on policy
makers to ameliorate inequality.

But in the translation from socialism to epidemiology something has also
been lost. In analyzing typhus, Virchow found the social seeds of disease,
and prescribed (and participated in) efforts to overthrow a social system
in which "thousands always must die in misery so that a few hundred may
live well." Allende¹s prescription for the ill health due to social
inequality was a united front to uproot capitalism and imperialism in Chile.

Daniels, Kennedy, and Kawachi describe the phenomenology of inequality and
injustice, but leave its origins and perpetrators obscured. Hence their
prescriptions call for a change in policy, but not in power. They would
redistribute wealth, but don¹t renounce the reign of the market or the
inviolability of property rights that are the mother and father of inequality.

Failing to identify the perpetrators of poverty and inequality can also
lead to confusion over policy choices. Daniels, Kennedy, and Kawachi imply
that resources might profitably be shifted from a profligate health care
system into programs to upgrade the standard of living for the poor. In
Canada, the Health Minister published a more explicit statement of a
similar view. Yet when health care was cut, the resources were transferred
to the rich, not to the poor.

The people and institutions (the corporate elite or ruling class) that
benefit from the unequal provision of health care also benefit from the
unequal distribution of wealth, education, and power. When these powerful
groups are weakened--through popular mobilization, inequality shrinks and
health care improves. Thus, the Great Society social programs of the 1960s
substantially shrank inequalities in income, education, wealth, and
housing--and coincided with the passage of Medicare and Medicaid. In
Sweden, recent cuts in health spending have coincided with a more general
attack on the welfare state.

Finally, we would raise three minor quibbles with Daniels, Kennedy, and
Kawachi. First, health care is so expensive in the United States that for
sick people, inadequate insurance often means poverty. Indeed, ill-health
is the leading cause of personal bankruptcy. Second, we are uncomfortable
with their implication that health is the key metric for measuring a
society. Does Japanese longevity make Japan a model society? Third,
analyses focused on inequality should not obscure the horrific absolute
deprivation in our society. In 1995, 11.6 million Americans went hungry,
4.4 million had their gas or electricity turned off, and one million were
evicted from their homes.

These criticisms should not detract from the great service that Daniels,
Kennedy and Kawachi have performed in this paper and elsewhere. They have
been effective leaders in academia in the fight for equality and justice.

------------------------------------------------------------------------
For Daniels, Kennedy and Kawachi on why "Justice is Good For Our Health,"

 Daniels, Kennedy, and Kawachi Respond

Justice is good for our health in two ways. At the "point of delivery,"
justice requires universal access to a system of health care that meets our
health needs fairly under resource constraints. "Upstream" from the health
care system, justice demands fair distributions of liberty, opportunity,
and basic resources. Achieving these fair distributions--correctly
specified, we think, by John Rawls¹s theory of justice--turns out to be a
crucial determinant of public health, so that justice improves overall
population health while reducing health inequalities. Our policies must
keep both pursuits of justice in view. Justice in the delivery of health
care and background justice in the society are both good for our health.

With that brief summary of our view, we turn to the comments by respondents
on our empirical, philosophical, and policy claims.

Empirics The empirical section of our essay presented some basic ideas
about the social determinants of health for a general reader; we did not
offer, as Marmor rightly says, a comprehensive discussion, let alone
historical review, for an expert audience. Our brevity, however, may have
produced some misunderstandings that we would like now to address.

Several respondents suggest that much health inequality can be attributed
to the difference between the very poor and everyone else, rather than to
inequalities in the population as a whole. If their claim is right, then
public policy ought to focus on getting the very poor above an acceptable
threshold, but should worry less about inequalities in the rest of the
population.

We are doubtful about the underlying claim. Recall that one of the most
intriguing findings we reported is the socioeconomic gradient in health
status: though the poorest individuals are at the greatest risk of dying
compared to the richest, the excess risk for mortality does not stop there.
Even individuals with household incomes in the $50,000 to $70,000 range
experience excess risk of death compared to the most affluent group. This
difference in risk of death cannot be due to the former being unable to
afford the basic necessities of life; some other processes are at work.
Moreover, this finding holds across countries and within relatively
middle-income groups. We mentioned, for example, the Whitehall Study of
British civil servants. The participants in this study might all be
classified as white-collar workers with steady employment and access to
universal health care; they were certainly not in poverty. Yet even in this
reasonably well-off cohort, the lowest occupation grades have four times
the mortality risks of the top grade. Health inequalities are not simply a
matter of the poor versus everyone else.

In a similar vein, Ted Marmor and Marcia Angell suggest that the
relationship between income inequality and health in the United States
might be attributable simply to differences in pockets of poverty across
states, so that the states with more or larger pockets of poverty would
also have worse average health outcomes. Several recent studies have
addressed this possibility by including individual income and health
outcomes along with state-level measures of income distribution. These
studies show that even when we control for individual income, thereby
accounting for between-state differences in the number of poor individuals,
individuals living in states with high levels of income inequality have
poorer health outcomes than those living in more egalitarian states,
regardless of their own individual income. To be sure, individuals with the
lowest incomes are worst off in health terms as well. But even those with
middle incomes have worse health outcomes than their counterparts in states
with more equitable income distributions. These studies show that the
observed relationship between income inequality and health is not simply a
function of the underlying association between individual income itself and
health. Of course, being well-off is good for your health. But living in a
more equal society is also good for your health.

Moreover, and finally, income inequality is not a "mysterious" cause of
undesirable health outcomes. It works through identifiable causal pathways,
including unequal access to opportunities such as education, healthy
employment, and health care; reduced social cohesion; distortions in
political participation; and the stress effects of relative lack of
control. These pathways are in and of themselves important contributors to
health and well-being and should serve as relatively uncontroversial levers
for policy intervention.

Philosophy Social justice is an important value, independently of its
impact on health. Ezekiel Emanuel and other respondents emphasize this
point, and we entirely agree. We don¹t just want justice exclusively
because it promotes population health or a fair distribution of health care.

Nevertheless, health is such a basic good that any plausible account of
justice must say something about the distribution of health care in a
society. Amartya Sen¹s theory, as Marmot notes, goes directly to the issue.
According to Sen, the point of justice is to promote our "positive
freedom," our ability to be and do what we choose. Disease and disability
strike at the heart of that ability: they directly diminish our capability
to function well. And because they do, justice directly requires an
appropriate distribution of the social determinants of health.

On John Rawls¹s theory of justice, things are more complicated. Justice,
according to Rawls, requires a fair distribution of basic liberties,
opportunities, and economic resources: health does not figure directly in
the view. Still, two lines of argument connect Rawlsian ideas of justice to
concerns about health care. The first argument rests on the empirical
discovery, described in our article, that achieving a fair distribution of
liberties, opportunity, and economic resources also causes a fair
distribution of population health. As Anand and Peter suggest, there might
have been no such relationship between the principles of justice and a fair
distribution of public heath. The principles might not have addressed the
key social determinants of health at all, or might have resulted in greater
health inequality. The second argument involves an extension of Rawls¹s
theory that does make the reduction of health inequalities a direct
requirement of justice. The idea is that we cannot ensure to people the
reasonable array of opportunities that justice requires without protecting
healthy functioning: serious impairments of health mean serious limits on
opportunities. Because justice requires such a reasonable array, we must
ensure a proper distribution of the social determinants of health, once we
discover what they are.

Anand and Peter find a "tension" between these two arguments: while we
argue here that Rawlsian principles indirectly and fortuitously assure an
equitable distribution of health, the extension of Rawls¹s view proposed by
Daniels directly requires equitable distribution because it includes an
assurance of normal functioning in the guarantee of equal opportunity. So
it may seem "redundant" to appeal to other principles as a way of securing
the equitable distribution of health. In particular, it may seem redundant
to emphasize the importance of fair income distribution for public health.
Opportunity does it all.

We find neither "tension" nor "redundancy," but a happy convergence of two
lines of argument. We do not intend to propose two theories of justice
bearing on health but only one, the extended theory that addresses health
through its affects on opportunity. When we discover what the key social
determinants of health are, we conclude--via the equal opportunity
principle--that justice requires these determinants to be distributed in
whatever ways produce equity in the distribution of health. At the same
time, we observe that the other principles of justice support a similar
conclusion. So the theory produces a convergence, not tension or
redundancy. Had the facts about the social determinants been different--for
example, had greater income equality been detrimental to public health--we
might have had a much more troubling result, inasmuch as different
principles of justice, with independent value, might then have worked at
cross-purposes.

A final point on philosophy: Unlike Marmot, we see a convergence of Sen¹s
and Rawls¹s view, at least once the extension of Rawls¹s theory to health
is made. In his early work, Sen argued that Rawls¹s emphasis on liberties,
opportunity, and economic resources obscured the problems produced by
disease and disability. Consider two people with exactly the same
liberties, opportunities, and economic resources, but one of whom is
disabled: these two would have very different levels of positive
freedom--different "capabilities for functioning," as Sen puts it--but
Rawls¹s theory would, he said, obscure this difference. So Sen urged that
our views about justice should focus directly on our real capabilities for
doing things, and not simply on our liberties, opportunities, and economic
resources.

Once it is understood, however, that assurances of normal functioning are
included in the idea of a reasonable array of opportunities, Rawls has a
way to blunt the force of Sen¹s criticism: he can register the difference
that a disability makes precisely as a difference in opportunity. From
Sen¹s side, his Development as Freedom focuses on the protections of
liberties and opportunities in ways that make him address issues long of
central concern to Rawls. Terminology aside, there is less difference than
meets the eye.

Policy A final set of comments relates to the role of health care in
reducing health inequalities. Marmor and Emanuel take issue with our
emphasis on reducing social disparities, and suggest that the more urgent
public policy problem is lack of access to health care in this country.
There is, of course, no gainsaying that medical care contributes to the
health of individuals and populations. And as Barbara Starfield notes,
primary care is particularly effective and important. It should not be
overlooked in discussions of how society should go about reducing health
inequalities.

Nonetheless, we disagree with Marmor¹s false forced choice analogy--that we
must somehow choose between expanding coverage of health care and devoting
our energies to changing the social distribution of other resources
(income, education, and opportunities for healthy work). These are not
competing objectives, but synergistic goals. Popular support for universal
health care coverage arises (when it does) out of a shared egalitarian
ethos that is itself a product of maintaining a relatively short distance
between the top and bottom of the social hierarchy. Witness, for example,
the birth of the National Health Service, which arose in Britain during an
unusually cohesive, egalitarian environment that followed the Second World
War. Conversely, societies that tolerate a high degree of inequality, such
as the United States, also have enormous difficulty in forging a consensus
about providing such communal benefits as health care. When the social
distance between the "haves" and "have nots" is wide, there is
correspondingly little motivation for those who are already covered by
health insurance to care about the plight of the uninsured.

A broader social movement seeking a more egalitarian distribution of
resources may well be a pre-condition for conducting a meaningful national
debate about universal health care (and for addressing the issues of power
that Woolhandler and Himmelstein raise). It is probably no accident that
the failed reform efforts of the Clinton administration appealed to
middle-class self-interest and to the self-interest of large employers
worried about costs, with no appeal to the moral considerations about
equality and fairness that lie at the heart of universal coverage. To
concentrate our efforts on expanding health care coverage just because it
seems more "doable" is therefore to confuse the prescription with the cause
of the underlying illness.

Marmor¹s excellent book Why Some People are Healthy and Others Not rendered
a valuable service by reminding us that access to health care is not the
major determinant of health inequalities. Again, the experience of the
National Health Service in Britain has taught us that provision of
universal health care does not by itself eliminate or reduce health
disparities (in fact, they have widened). The Whitehall Studies similarly
indicate that among individuals who have access to the same health care
service, there can be three- to four-fold differences in the risk of
premature mortality according to one¹s access to other resources (the
amount of control on the job, prestige, income and wealth, and so on). In
other words, a society interested in reducing health inequalities is
unlikely to achieve it by focusing on the provision of health care alone.
Policies to improve population health must concern themselves as much with
the sources of health (and correcting inequalities in their distribution),
as with the instrumental means of curing illnesses (i.e., the provision of
health care).

We would agree with the view that adding poor health to the list of
outcomes associated with income maldistribution is unlikely to spark a
revolution, or to make Americans care more about trends in income
distribution. The point we wish to reiterate is that good health (which
most people do care about) depends to a large extent on factors that lie
outside the health care sector, and that a society wishing to reduce health
inequalities needs to engage willingly in intersectoral efforts--early
childhood investment, narrowing the income gap, ensuring healthy
workplaces, and other similar policies mentioned in our essay.

Health is too important to be left to the doctors alone.

CONTRIBUTORS
Sudhir Anand is professor of economics at Oxford University and adjunct
professor at the Harvard School of Public Health.

Marcia Angell is a physician and editor-in-chief of the New England Journal
of Medicine.

Norman Daniels is Goldthwaite professor of philosophy at Tufts. He is
author of Justice and Justification: Reflective Equilibrium in Theory and
Practice.

Ezekiel Emanuel is chair of the Department of Clinical Bioethics at the
Warren G. Magnuson Clinical Center of the National Institutes of Health and
a breast oncologist.

Emmanuela Gakidou, Julio Frenk, and Christopher Murray work on the World
Health Organization¹s global programme on evidence for health policy.

Ichiro Kawachi is director of the Harvard Center for Society and Health. He
is co-editor of the Reader on Income Inequality and Health.

Bruce Kennedy is assistant professor at the Harvard School of Public Health.

Ted Marmor is professor of public policy at Yale and author of The Politics
of Medicare.

Michael Marmot is professor of epidemiology and public health at University
College London.

Barbara Starfield is University Distinguished professor at The Johns
Hopkins Medical Institutions.

Steffie Woolhandler and David Himmelstein practice and teach internal
medicine in Cambridge, Mass. They are vocal proponents of national health
insurance.

Visit our Web Site for information about our Seniors Participatory and
Community Quality of Life Projects!  Free Reports Also.

  http://www.utoronto.ca/qol      http://www.utoronto.ca/seniors

  ********************************************************************
  Long have I looked for the truth about the life of people together.
  That life is crisscrossed, tangled, and difficult to understand.
  I have worked hard to understand it and when I had done so
  I told the truth as I found it.

  - Bertolt Brecht
  ********************************************************************

Dennis Raphael, Ph.D.
Associate Professor and Associate Director,
Masters of Health Science Program in Health Promotion
Department of Public Health Sciences
Graduate Department of Community Health
University of Toronto
McMurrich Building, Room 101
Toronto, Ontario, CANADA M5S 1A8
voice:    (416) 978-7567
fax: (416) 978-2087
e-mail:   [log in to unmask]