10224.pdf
Media
- extracted text
-
j
COMMISSION ON MACROECONOMICS AND HEALTH
A WHO commission examining the interrelations among investments in health, economic growth and poverty reduction
CMH Working Paper Series
Paper No. WG1 : 3
Health, Inequality, and Economic
Development
Author:
Angus Deaton
Date: May 2001 (revised)
Health, inequality, and economic development
Angus Deaton
JEL No. 112
ABSTRACT
I explore the connection between income inequality and health in both poor and rich countries. I
discuss a range of mechanisms, including nonlinear income effects, credit restrictions, nutritional
traps, public goods provision, and relative deprivation. I review the evidence on the effects of
income inequality on the rate of decline of mortality over time, on geographical pattens of
mortality, and on individual-level mortality. Much of the literature needs to be treated
skeptically, if only because of the low quality of much of the data on income inequality.
Although there are many puzzles that remain, I conclude that there is no direct link from income
inequality to ill-health; individuals are no more likely to die if they live in more unequal places.
The raw correlations that are sometimes found are likely the result of factors other than income
inequality, some of which are intimately linked to broader notions of inequality and unfairness.
That income inequality itself is not a health risk does not deny the importance for health of other
inequalities, nor of the social environment. Whether income redistribution can improve
population health does not depend on a direct effect of income inequality and remains an open
question.
COMMUNITY HEALTH CELL
Library and Information Centre
No. 367, Srinivasa Nilaya, Jakkasandra,
I Main, I Block, Koramangala, Bangalore - 560 034.
THIS BOOK MUST BE RETURNED BY
THE DATE LAST STAMPED
Angus Deaton
Center for Health and Wellbeing and
Research Program in Development Studies
347 Wallace Hall
Princeton University
Princeton, NJ 08544.
Tel: 609-258-5967
Fax: 609-258-5974
deaton@princeton.edu
1. Introduction
Any introduction to the topic of health and economic development is likely to begin with the
celebrated Preston curve, Preston (1975), which shows the cross-country relationship between
life-expectancy and income per head. Among the poorest countries, increases in average income
are strongly associated with increases in life expectancy but, as income per head rises, the
relationship flattens out, and is weaker or even absent among the richest countries. As Preston
noted, if such a nonlinear relationship holds within countries, countries with a more equal
distribution of income will have a higher average life-expectancy. The health of the rich is not
much affected by their income, so that transfers of income from rich to poor will improve the
average health of the nation. In recent work, this relationship between income inequality and
health has moved from being a supporting player in the story to center stage. Not only does
income equality promote health because income does more for the health of the poor, but it also
serves as a marker for other desirable features of society. According to this account, equal
societies have more social cohesion, more solidarity, and less stress, they offer their citizens
more social support and more social capital, and they satisfy humans’ evolved preference for
fairness. Equal societies are healthier, an argument that is the main topic of Wilkinson (1992,
1996, 2000), as well as the collection of papers edited by Kawachi, Kennedy, and Wilkinson
(1999).
This paper explores the theoretical and empirical basis for a connection between inequality
and health. I shall be concerned with poor as well as rich countries, and with the links between
health and income inequality at different levels of economic development. The proposition that
income inequality is a health risk was first proposed for the wealthy countries on the flatter upper
I
part of the Preston curve, where chronic diseases have replaced infectious diseases as the main
cause of mortality. But as we shall see, many of the arguments that income inequality is a health
risk are as plausible for poor as for rich countries and in some cases, more so.
I devote a substantial fraction of the paper to theory, albeit illustrated with at least some of
the evidence. With a few exceptions, the literature does not specify the mechanisms through
which income inequality is supposed to affect health. In consequence, there is little guidance on
exactly what evidence we should be examining, or whether the propositions are refutable at all.
Section 2 lays out some of the possible stories, starting with the simple case in which health is
affected by income, and there is no direct effect of inequality. This is sometimes referred to us
the “absolute income hypothesis,” to emphasize that it is income that matters for health, not
income relative to other peoples’ incomes, nor income inequality. A name that would be at least
as good is the “poverty” hypothesis, that ill-health is a consequence of low income, in the sense
that more income improves health by more among those with low incomes than among those
with high incomes. It is important to start by establishing the full range of implications for this
simple case, and in particular what role is played by income inequality. I also discuss what
happens when we make health depend, not on absolute income, but relative income, and what the
relative income hypothesis implies for the relationship between health and inequality.
Other theoretical accounts concern the possible effects of income on investments in health
and education, the two-way link between nutrition and earnings at low levels of income, and the
possible negative effects of inequality on the ability of the political process to deliver public
goods. I also consider arguments that our evolutionary history predisposes us towards fairness,
and sickens us when live in unequal environments. Such an account can be made consistent with
2
a story in which relative deprivation is a cause of ill health, and in which inequality is important
across groups, but not within them. I also briefly consider the link between inequality and crime.
Finally, 1 consider the important case in which income inequality is in part a consequence of illhealth, so that policies that reduce the likelihood of sickness, shorten its duration, or ameliorate
its effects on earnings, can also narrow income inequalities. Some income inequality is a con
sequence of the fact that earnings cannot be completely insured against ill-health, so that better
health insurance is likely to help reduce inequalities in income.
Section 3 turns to the evidence, most of which comes from developed economies. I review
cross-country studies on adult mortality for rich countries and on child mortality for both rich and
poor countries. A major question is whether the quality of the international data on income
distribution is of sufficiently high quality to support the inferences that are being made. Better
data, or at least more consistent data, are available within countries, and some of the most
interesting evidence on inequality and health comes from studies looking across areas within
developed countries, such as Britain, Canada, and the United States, and in a few poor countries.
I also review the evidence from studies that link individual mortality and morbidity to the
ambient level of income inequality.
My tentative conclusion is that there is no direct link from income inequality to ill-health, in
the sense that individuals are no more likely to be sick or to die if they live in more unequal
places. The raw correlations that exist in (some of the) data are most likely the result of factors
other than income inequality, some of which are intimately linked to broader notions of
inequality or unfairness. The fact that income inequality itself is not a health risk does not mean
that inequalities more generally are not important, let alone that the social environment in which
3
people live is irrelevant for their health. Indeed, I shall argue precisely the opposite. But we must
narrow and focus our search if we are to make the leap from correlation to policy.
In an attempt to address a relatively broad audience, I have kept the discussion as non tech
nical as possible, replying on citations (and an Appendix) to provide formal demonstrations
whenever possible, hi the few occasions where there are equations in the main text, I have
attempted to provide ample verbal explanation. The literature on income inequality and health is
already extensive, and 1 have tried to tell a coherent story rather than covering everything or
producing an annotated bibliography of the literature
2. Theoretical accounts of income inequality and health
2.1 Individual, group, and national health
2.1.1 Health, income and poverty
Most people find it intuitively plausible that the effects of income on health, if they exist at all,
should be greater among poor people than among the rich. Although there is a great deal of
evidence that the effect is not confined only to the poor, the intuition is supported by the evid
ence. Figures 1 and 2 were calculated using data from the National Longitudinal Mortality
Survey in the US. This is a national follow-up study of about 1.3 million people (with about
three-quarters of a million people in the public release data) who were interviewed in Current
Population Surveys or in a census related sample around 1980, and whose deaths over a follow
up period of 3,288 days were ascertained by matching to the National Death Index, see Rogot et
al (1992) for details. The survey collects each person’s family income (within one of seven
ranges) so that it is possible to link the probability of death during the follow-up to family
4
income and other variables, the most important of which are sex and age. For adults between the
ages of 18 and 85, the log-odds of mortality is approximately linear in age, so a convenient way
to summarize the data is to estimate a logit model for the probability of death in which age is
entered linearly, together with a series of dummy variables for the age categories. The results are
shown in Figure 1, separately for white men and white women aged between 15 and 85 at the
time of the interview. This graph shows that, for both men and women, the log-odds of mortality
is approximately linear in the logarithm of family income. As a consequence, when we use the
results to plot the age-adjusted probability of death against income in Figure 2, we get the non
linear curves as shown. That the curve for women appears less curved than the curve for men is
an insignificant consequence of the choice of scale.
The curves in Figure 2 illustrate the point first made by Preston (1975), though Preston’s data
were on countries, not people. The effect of income on reducing the probability of death at the
bottom of the income distribution—and the bottom circle is at about the 5th percentile of family
income—is much greater than its effect at the top of the distribution. As a result, if income is
redistributed from the rich, whose health is not much affected, to the poor, whose health is more
responsive to income, average health will improve. Other things being equal, including average
income, nations (or other groups) with a more equal distribution of income will have better
average group health. The same is true internationally; anything that raises the GDP of poor
countries relative to that of rich countries will improve average health across the world. Within
poor countries, infant and child mortality is likely to be particularly sensitive to changes in
income near the bottom of the distribution so that, once again, redistribution towards the poor
will reduce child mortality even without raising average incomes.
5
It is often useful to think of this story in terms of poverty. If a country with a high average
income has a great deal of income inequality, then there are a relatively large number of people
with low income whose health is poor. Although there is no poverty line in Figure 2, below
which income matters, and above which it does not, it is at the bottom of the income distribution
that the relationship between income and health really matters. And if a rich country has a lot of
poor people, it will have low average health relative to its per capita income.
An important application of these ideas is provided by the history of mortality in Britain and
the US in the 20th Century. Wilkinson (1989) looks at mortality differences by social class in
Britain from 1921 to 1981, and argues that mortality fell most rapidly at times when income
differentials were narrowed, particularly at times when incomes of the poor rose more rapidly
than those of the rich, such as during World War II. Sen (1999, Figures 2.2 and 2.3) shows how
life expectancy in England and Wales from 1901 to 1960 grew most rapidly in the decades
1911-21 (by 6.5 years) and 1940-51 (by 6.8 years), and more slowly at other times, 4 years in
1901-11, 2.4 years in 1921-31, 1.4 years in 1931—41, and 2.8 years in 1951-60. Sen shows that
the decadal rate of growth of GDP per capita is strongly negatively correlated with decadal
increases in life expectancy, and like Wilkinson, focuses on the degree of sharing during the two
wars, as well as on the direct nutritional and health interventions that took place during and
immediately after the second war. Both wars brought well-paying employment opportunities to
many people in Britain for the first-time, including many women. Hammond (1950) discusses
how wartime food policy in the 1940s brought fresh milk and vitamins to working people, even
to the extent that their nutritional status actually improved during the hostilities. Reductions in
income inequality during the wars, if they existed, marked an improvement in the conditions of
6
the working people, among whom better incomes and better nutrition would have had the largest
effects on mortality.
As I shall argue in Section 3.5, it is not possible to link recent increases in income inequality
in the US and Britain to mortality changes. It should be noted, however, that recent increases in
income inequality, though large enough by postwar standards, are probably not large relative to
earlier compressions, particularly those associated with the World Wars, see Lindert (2000) and
Goldin and Katz (2001).
The inequality inducing effect of nonlinearity in the relationship between individual income
and individual health has come to be known in the literature as a “statistical artefact,” Gravelie et
al (1998), in order to distinguish it from mechanisms in which income inequality has a direct
effect on individual health. The term is unfortunate to the extent that it suggests there is no real
link between income inequality and health, and seems to carry the implication that redistributive
policy cannot improve average population health. By contrast, the effect of nonlinearity is so
plausible that it would be surprising if income redistribution did not improve average health, and
if there were no link between income inequality and average health across population aggregates.
2.1.2 Poverty, health and the effects of inequality at different income levels
The absolute income (or poverty) hypothesis tells us a good deal about how we can expect
average income and income inequality to affect population health at different levels of
development. At the most obvious, the effects of per capita income mirror those of individual
income, and become less important the richer is the country. Eventually, we would expect
income inequality to lose its effect too, but it is not enough that average income be high enough,
7
we also need everyone’s income to be high enough. The bottom tail of the income distribution
has to be pulled up beyond the point at which income has much effect on health. Before that,
there will still be poverty even in rich countries, so that income inequality will still matter as well
as average income. In consequence, the absolute income or poverty hypothesis implies that,
among the poorest countries, average income is what matters for population health, and income
inequality is relatively less important. Among rich countries, average income is less important,
and income inequality relatively more important. Eventually, neither will matter much for
population health but, under plausible assumptions, the effect of income inequality relative to
that of average income continues to grow as countries become richer. These implications of the
absolute income hypothesis are important because it is often claimed that the differential effects
of average income and income inequality on health are as described, and that the observation
helps establish the case for a direct effect of inequality on health. While the finding it is certainly
consistent with such an effect, it is also consistent with a simple model in which income has a
larger effect on health among the poor.
The rest of this subsection is devoted to demonstrating the claim in the last paragraph in the
context of a simple illustrative model. A more realistic case is developed in the Appendix. I
assume that health is a quadratic function of income. Suppose that individual i lives in country s,
and that her health his is a quadratic function of her family income
so that if, for convenience,
I write everything relative to worldwide means, signified by overbars, we have
his - h = a + p (yis -y) - y (yls -y )2.
(1)
(Note that equation (1) could be rewritten as a linear relationship between health, income and
income squared, though the parameters would have to be redefined.) Income promotes health, but
8
by less for the rich, so that both p and y are positive. Equation (1) is assumed to hold true
everywhere, for all individuals, whether or not they live in poor or rich countries, and wherever
they are in the epidemiological transition. To find out what happens to the country means, we
have to average equation (1) across all i in each country 5. When the last term on the right-hand
side is averaged, it gives a term in the squared deviation of average income from the global
average, plus the variance within each country. Looking across countries, (1) then becomes
h s -h
(2)
where variables subscripted by s are population means, and v5' is the variance of income in
country s, which, in this context, is the natural inequality measure. Equations (1) and (2) demon
strate the essential point that, provided the individual equation is nonlinear, which in this case
means that the parameter y is not zero, the inequality of income matters in the aggregate relationship (2), even though it plays no part in the individual relationship (1).
These equations also allow us to be precise about the links between health, average income,
and income inequality, and to see how those links change with income. The effect of average
income on average health is the derivative of (2) with respect to ys, which is p - 2y(y5 -y)
which, since y is positive, gets smaller as average income increases. The effect of inequality on
population health is (minus) the parameter y (inequality makes average health worse) which is
constant. As a result, as a country becomes richer and average income rises, the effect on
population health of income inequality becomes more important relative to the effect on
population health of average income. Even when income is large enough for the effect of average
income to be zero, the effect of inequality remains constant.
9
2.1.3 Separating the aggregation effect from the direct effect
One use of the curves in Figure 2 is to calculate how income inequality can be expected to affect
mortality across aggregates of people, such as cities, states, or whole countries. This allows a
direct link to be drawn from the degree of curvature of the income-mortality relationship in the
individual data to the role of income inequality in explaining group mortality. Such calculations
have been done by Wolfson et al (1998), also using the NLMS data, who then go on to show that
the effects predicted by the nonlinearity are insufficient to explain the actual relationship between
income inequality and mortality across the US states. But it turns out that it is not strictly neces
sary to go back to the individual data, and that it is often possible to calculate the effects of the
aggregation from the grouped data themselves. This is important, because it gives us a way of
telling whether the effects of inequality at the group level can be explained through the nonlinear
effects of income alone, or whether we have evidence of a direct effect of inequality at the indi
vidual level.
The result is due to Miller (2000), who points out that the shape of the curve in the individual
data is carried through to the aggregate data, albeit with additional terms in inequality, so that any
additional effects of inequality due to a direct effect on individual health can be observed using
only the aggregate data. Again, the quadratic case illustrates the point. Suppose that for indivi-
dual i in state s, equation (1) holds as before, with the addition of a direct negative effect of
inequality on health. We can write this as
h.
his -h = a + P(TZi.-?) -Y(y/5-T)2
(3)
The presence of vs in the individual equation represents a pure effect of inequality on individual
health, whose size depends on the parameter 0. When we average (3) over all individuals in state
10
s, we again get equation (2), but with the additional variance term carried through,
h s - h = »+ PO'j-y) -y(y4.-y)2
(Y+9)v5.
(4)
In equation (4), the variance term appears from the aggregation, through the parameter y, but also
through the “direct” effect of inequality on health, the parameter 0. But because the quadratic
term appears in the state level relationship (4) just as it does in the individual relationships (1) or
(3), we can estimate y from the aggregate data by looking at the quadratic term in (4). Given y,
the parameter 0 can be recovered by subtraction, so that the aggregate data are potentially
informative about whether the effect of inequality is an aggregation effect, or something else.
2.1.4 Relative income, absolute income, and inequality
Individual health is affected by many things other than individual income, and it is possible that
the relationship between health and income itself is spurious, with income standing proxy for
some other variable. An obvious candidate for such a variable is education; if conditional on
education, income has no effect on health, redistributive taxes that narrow the distribution of
income will have no effect on health. In the US, the relationship between income and mortality is
not much affected by controlling for education, Elo and Preston (1998), Deaton and Paxson
(2001a), but there are few other studies or data sets that would allow us to come to general
conclusions about the separate roles of income and education.
Even if income is the crucial variable, it is possible that health is determined, not by absolute
income, but by income relative to some aspiration level, or relative to the incomes of others.
Easterlin (1975) long ago found evidence that happiness is independent of income in the longrun, and health may follow the same pattern. That health depends on income relative to average
11
incomes of one or more reference groups is the relative income hypothesis. This could happen in
a number of different ways. One case is where relative income determines access to material
goods, for example when the people who live in a town are the market for local land for housing,
with the richest getting the hilltop plots with fine views, and the poorest getting the plots
downwind of the smokestacks, see Gaarder (2000) who argues that this is one mechanism that
sets who gets exposed to air pollution. The local housing case is an example where it is not
money itself that is important, but rank, here determined by money. More generally, rank is
important in determining how much control people have over their lives, particularly at work
and, as demonstrated in the Whitehall studies, the degree of such control is an important
contributor to health.
The relative income theory is consistent with an effect of income inequality, although it does
not imply it. To see how things might work out, suppose that, as before, individuals are labeled i.
but that we use 5 to index the relevant comparison or reference group. An individual’s health
might then be above or below the population mean depending on whether her income is above or
below the average income in the reference group. Hence, with [3 positive, we might write
h.IS
h + P (yis Z-)-
(5)
For the group as a whole, group average health hs is just average health h, which is the same for
all groups, so that neither group income nor group income inequality has any effect on group
health. This annihiliation of the effect of income on health as we move from the individual to the
group will always happen if the reference group is within the unit over which we are aggregating.
For example, if reference groups are geographically local, average income will be unrelated to
average health at the region, province, or state level. Such an account offers one explanation for
12
why income might be related to health within groups, but not between them.
The relative income story is also consistent with a role for income inequality at the group
level, and the mechanism is the same as for the absolute income hypothesis, the nonlinearity of
the relationship between income, now relative income, and health. Again, an obvious example is
provided by the quadratic version of equation (5), in which individual health depends on income
relative to group income and the square of income relative to group income:
h.IS
6 + POvj;) -Y0\-y5)2
(6)
When we average health over individuals within each group, the first term vanishes, as before,
while the second term becomes the variance of income,
h s = h - y v,
(7)
This provides us with one account of what several researchers have observed, that within states
or countries, individual health is related to individual income, while between them, average
health is independent of average income but is negatively related to income inequality. In Sub
section 2.5 below, I present another formulation that has similar properties.
2.1.5 The impossibility of identifying reference groups
An immediate problem with the implementation of any model of relative income is the identi
fication of the relevant reference group. In a few cases, such as Whitehall, the reference group (or
at least what is likely people’s most important reference group) is a ready-made part of the
design. More usually, reference groups are not clearly defined, and indeed people will often have
multiple such groups, comparing themselves to their neighbors, to their co-workers, to those they
meet in social and religious organizations, and to those they see on television or read about in
13
newspapers. One way of dealing with this is to recognize that reference group incomes cannot be
observed, and to work out the effects of the omission on the relationship between the two things
that we can observe, health and income. As shown in Deaton (2001), this procedure brings
inequality back into the story even when it has no direct role.
Figure 3 illustrates for the simple case where there are two groups, labeled “economists” (on
the left) and “doctors” on the right. Income is measured on the horizontal axis, health on the
vertical axis. The two ellipses show where economists and doctors are located in terms of their
health and incomes. One can think of individuals in each group as scattered within the two
elliptical areas. Doctors have higher incomes than economists, and within each group, individual
health depends on individual income relative to other members of the group. The two parallel
steep lines show the relationships between income and health for each of the two groups.
Although doctors have higher incomes than economists, their individual health is no better on
average because their absolute income does not matter, only their income relative to other
members of their group. Suppose that an epidemiologist analyzes the data on economists’ and
doctors’ health, but without knowing which is which. When the data are pooled, the relationship
between health and income is the flatter, broken line. By mixing the two groups, omitting the
relevant information on group, the relationship is “attenuated” or flattened out.
Inequality comes into this story because the degree of flattening depends on ratio of withingroup income inequality to between-group income inequality. This is easy to see in the Figure. If
doctors and economists are moved further apart, by moving the two ellipses horizontally away
from each other, the broken line will become flatter , i.e. more attenuated. If the within group
inequality is increased, holding between group difference fixed, so that the ellipses are stretched
14
out along their individual income to health lines, the broken line will become steeper, increasing
the gradient between income and health. The steepness of the gradient depends positively on the
ratio of within to between group inequality. For example, in Whitehall, if there is only one
reference group, that of British civil servants, the gradient of health with respect to income (or
rank) is likely to be steeper than in a study containing a mixture of indistinguishable reference
groups. More generally, if health depends on relative, not absolute income, and there is an
increase in income inequality that increases inequality within groups more than it increases it
between groups, the slope of the gradient will increase.
In Deaton (2001), I show that if health depends on relative income as in equation (5), so that
h.IS = h + p o;j
yS )7 + £.IS
(8)
is a random term that ensures that there is a scatter around the line, then the expectation
where
of health conditional on individual income takes the form:
E(his\y^
’ yis
h + J30—
2
where
2
(9)
+ ^b
2
and Qb are the within- and between-group variances of income.
2.2 Inequality, education, health, and economic growth
The literature on economic development has recently re-examined the links between inequality
and growth. In the past, economists have often thought that income inequality would be good for
growth. One mechanism works through saving. If only the rich save, and if saving finances
investment for growth, redistribution towards the rich will be growth enhancing. More generally,
the disincentive effects of redistributive taxes are themselves likely to inhibit initiative, entre15
preneurship, and risk-taking, and thus to lower output and discourage growth. The recent
literature has explored a number of mechanisms that operate in the other direction, and several of
these are relevant to the argument here, or can be adapted to be relevant. I draw in particular on
the useful survey articles by Benabou (1996) and Aghion et al (1999); see also Bertoia (2000),
Bardhan, Bowles, and Gintis (2000) and Galor and Moav (2000).
In an economy where everyone can borrow and lend without restriction, each person’s
investment should not be restricted by each person’s own resources. If someone has an invest
ment opportunity that can be expected to earn a good rate of return, it should be undertaken; the
money needed to fund it will be earned by the investment, with profit left over. If the person with
the idea, or with the ability to carry out the project lacks the funds to do so, he or she can borrow
money, paying it back as the project pays off. If we think of investments as construction projects.
building a factory or a bridge, the financial standing of the investor will indeed often not be
relevant. But if we are thinking of investments in education (human capital) or health (health
capital), matters may be different, and the lack of access to borrowing on reasonable terms may
prevent many poor people from making profitable investments in themselves or in their children.
The inability to borrow is likely to afflict the poor more than the rich; for one thing, the poor do
not have other assets that can be used as collateral. In consequence, redistribution of income—or
of assets, such as land—from rich to poor may increase levels of education and health. Education
promotes health—Elo and Preston (1996) estimate that, around the world, a year of extra
education reduces mortality rates by about 8 percent—so that redistribution will improve average
health, both directly and indirectly.
Even when loans are available in principle, many parents may be unwilling to take the risk of
16
borrowing substantial sums of money for their children’s education. Although the investments
may be profitable on average, not all will be, and not all children will be either willing or able to
help their parents repay the loans taken out in their behalf Some poor parents will simply not be
willing to take the risk. Once again, redistribution may increase investment in health and in
schooling.
It is also possible that even some better-off people are unwilling to borrow much to educate
their children, or to go deeply into debt for expensive medical care. If so, redistribution of
income or of assets from rich to poor may reduce health and educational investment among the
rich almost as much as it increases them among the poor. Even so, redistribution may promote
growth if the returns to investment are higher among the poor than among the rich. If poor people
rarely finish primary school, and if rich people usually finish college, and if the rate of return,
either in future earnings or in subsequent health, is higher to an additional year of elementary
school, (Schultz, 1991, Strauss and Thomas, 1998) the redistribution from rich to poor will
increase the rate of return to education as a whole, increasing both growth and average health. A
similar story may hold for girls’ education, which may have a higher or lower rate of return in
earnings than boys’ education, but certainly has a higher rate of return in terms of health.
Figure 4 shows relevant data on education for rural India, see Filmer and Pritchett (1999) for
related evidence for a wide range of countries. The 52nd Round of the National Sample Survey
collected data in 1995-96 on education enrolment and status of respondents. The graph shows
the percentage of boys and girls aged from 7 to 12 (inclusive) that are currently enrolled in school
as a function of the logarithm of total household expenditure per head, a measure of overall
living standards in the household in which the child resides. The graphs are calculated using non17
parametric regression using 20,307 boys and 17,321 girls aged 7 to 12. Boys are more likely to be
in school than girls, both are more likely to be in school when they live in better-off households,
and the effects of additional resources are larger for girls than for boys. Although the slopes of
the curve vary somewhat with per capita expenditure, they are clearly flatter to the right. In
consequence, redistribution from richer to poorer households will increase the total percentage of
children in school, and will increase girls’ enrolment more than boys’. Because education is an
input into both health and production, and because the rate of return to education is higher at
lower than at higher levels of education, and higher for girls than for boys, the positive effects of
redistribution will be further enhanced
Figure 5 shows some contrasting data on child vaccinations from the same Indian survey. As
was the case for school enrolment, the fractions of children under 5 who have all three of the
most important vaccinations is lower for poor than for rich households. However, unlike
education, the slope is almost as steep among better-off than among poorer households, so that
this graph indicated no major effect of redistributing income on raising child vaccination rates.
Nor is there any great difference between vaccination rates between boys and girls.
These examples are only suggestive. The curves in Figures 4 and 5 make no attempt to
control for other factors
such as parental education, school quality or health service provision
that are likely to be positively correlated with both household income and the outcome, so that
the effects of income are almost certainly overstated. Nevertheless, the graphs illustrate that
income redistribution may (or may not) increase health and educational investments in children.
18
2.3 Nutritional wages, destitution, and land-holdings
The nutritional wage model provides an elegant and rigorous account of how inequality affects
both health and earnings while explicitly recognizing that health and earnings are simultaneously
determined. The ideas go back to Leibenstein (1957), with the fundamental work by Mirrlees
(1975) and Stiglitz (1976). An excellent and still useful survey of the field is Bliss and Stem
(1978). That nutritional wage models can account for persistent poverty and destitution in poor
countries is eloquently argued in Dasgupta and Ray (1986), (1987) and Dasgupta (1993); an
excellent textbook summary is provided by Ray (1998, Chapter 13).
An important insight of the nutritional wage story is that poor workers, at risk of under
nourishment, may not be able to earn enough to buy the food that would sustain the work
required. If this nutritional trap is to be avoided, two conditions must be satisfied. The first is
physiological; the worker must consume enough calories to be able to do undertake whatever
work he or she does for wages, as well as to sustain basic metabolism and resting bodily
functions. The second condition comes from the market; the wages from earnings must be
sufficient to buy the calories consumed. But these two conditions are not necessarily mutually
compatible, especially when wage rates are low. At such wages, there may be no combination of
work and calories that satisfies both the physiological and market conditions. In this situation,
there is a critical wage rate that just enables the worker to work and survive. For any wage rate
below it, long-term survival is not possible by working in the labor market.
The existence of this critical wage seriously disrupts the usual analysis of supply and demand
in the labor market. Employers demand labor in the usual way and are prepared to employ more
labor the lower is the market wage. Provided the wage rate is high enough, above the critical
19
point, higher wages elicit more labor from workers in the usual way. However, when wages are
lower than the critical level, there is no labor supply, because workers can no longer obtain
enough calories to do the work. As a result, if the demand for labor is less than its supply at the
critical wage, the resulting unemployment cannot be eliminated by decreasing the wage rate to
bring supply and demand into balance. The surplus workers are unemployed, and worse than that,
must live on whatever they can beg or come by without working. They are chronically under
nourished, and cannot undercut those in work by offering to work for less because, at any lower
wage, they will not be well-enough nourished to be useful to potential employers.
What are the solutions to the destitution produced by such a labor market? One possibility is
to find a way of providing nutrition outside of the labor market, so that even at a low wage rate,
the worker will have enough food to work. This will happen if basic nutritional needs can be met
without having to work for wages, for example by growing food on the worker’s own plot of
land. Seen this way, the problem of destitution is a problem of landlessness, and can be resolved
by redistribution of land so as to give to every family a small plot that is sufficient to guarantee
their basic needs, and to enable them to participate gainfully in the labor market. Destitution is
therefore a problem of maldistribution and its solution is a more equitable distribution of land.
The nutritional explanation of destitution and chronic malnutrition in India by Dasgupta and
Ray is echoed by Fogel’s (1994) account of the economic and health history of Europe. He
argues that, even at the end of the eighteenth century in England and France, food production was
so low that, given its likely distribution over people, perhaps a fifth of the population was
capable of no more than a few hours of light work each day. These people were chronically
malnourished, short in stature, and in life expectancy. Only the increase in agricultural product20
ivity in the nineteenth century permitted an escape from this nutritional trap, and the beginning of
the transition to better health, lower mortality, and lengthening life-expectancy.
The story of nutrition and wages has much to commend it. Unlike most other accounts, it
directly confronts the two-way causality between health and earnings. It also provides a general
equilibrium explanation of unemployment and poor health that has an obvious relevance to poor
countries now as well as to the historical record in now rich countries. It also pinpoints land
reform as a policy prescription. And for our immediate purpose, it identifies the inequality of
land holding as a cause of malnourishment and poor health.
Yet the theory has many critics who doubt its descriptive realism and fault its implications,
see in particular Binswanger and Rosenzweig (1984), Rosenzweig (1991, 720-28) and Strauss
and Thomas (1998). Workers who are trapped by their low nutrition and inability to work would
devote all their energies to finding food, and would have no energy for consuming anything other
than food, for saving, or even for procreation, Gersovitz (1983). Wage rates appear to be flexible
downward, which the theory says is impossible. Even in the poorest economies, food is typically
too cheap relative to the wage rate to make the trap plausible. For example, Subramanian and
Deaton (1994) calculate that in rural Maharashtra in 1983, 2,000 calories (in the form of standard
coarse cereals) could be purchased for less than 5 percent of the day wage, a finding that is
consistent with the observation that poor agricultural workers in India typically eat their fill of
cheap calories at the end of the work day, see also Swamy (1997). With food so cheaply obtain
able, nutritional wage traps seem too easy to escape.
More broadly, the model does not draw a clear distinction between nutrition, which comes
from the food that can be bought for money, and nutritional status, which depends on disease as
21
well as on nutritional inputs. A plentiful supply of food will not nourish someone whose drinking
water and food are contaminated, and chronic malnutrition typically needs to be addressed
through public health measures as well as by increasing the supply of food. This criticism
applies, not only to the nutritional wage theory as an account of destitution in poor countries, but
also to McKeown (1976) and Fogel’s (1994) arguments about the historical importance of nutri
tion, see particularly Preston (1996).
2.4 Inequality, politics, and public goods
It is often argued that inequality may make it more difficult for people to agree on the provision
of public goods, such as health, water supply, waste disposal, education and police. Such
mechanisms have long been recognized in the literature on political economy, and a simple
formal account has recently been provided by Alesina, Baqir, and Easterly (1999). The story is
one of a local community whose members want to provide a public good, and who must decide
how much to spend. A useful way to think of this is to imagine the members of the community
evenly spaced out in a circular town, with a clinic to be built in the center. The radius of the
town—or the size of the community—is a metaphor for inequality; in equal communities all live
close to the center, while in unequal ones, the members are on average much further from the
middle. The value of the clinic to each person diminishes with their distance from the center, so
that the people at the edge of town value it less than those who live in the middle. The average
(or median) value of the clinic is therefore higher the more compact is the community. If the size
of the clinic is decided by voting, more specifically by the wishes of the median voter, a larger
clinic will be built in the more compact town, because the median value of the clinic to its users
22
is higher. More generally, when people’s preferences are heterogeneous, goods held in common
are less valued on average, and fewer of them will be provided.
Although they note the potential application to income inequality, Alesina, Baqir, and East
erly think of their model as applying to racial divisions in the United States, and to ethnic
fractionalization more widely, a variable that often appears to have negative consequences in
cross-country growth studies. In the context of the cities and counties of the US examined by
Alesina et al, ethnic fractionalization is closely related to the fraction of the population that is
black, which is positively related to total spending, but negatively related to the shares in
spending of “productive” public goods, such as health, roads, and education. For health, the total
effect offsets the share effect, so that the absolute amount of health spending is positively
associated with fractionalization. Income inequality is included in the models, although the
results are not presented in the paper. However, I understand from private communication with
one of the authors that the effects vary from model to model and that there are no robust negative
effects of inequality on either total spending or its distribution. It should also be noted that
Putnam and his collaborators (1983), in their study of social capital in Italy, also see equality as
an important element of the civic community.
While I know of no work in developing countries that links the provision of public health
resources to income inequality, Szreter (1988) has provided a fascinating account of politics and
sanitation in Britain in the middle of the 19th century, see also Easterlin (1999). The urbanization
of population associated with the industrial revolution led to a sharp reduction in public health,
with mortality higher in cities than in the countryside, and a decline in overall life expectancy.
Urban populations often had no access to clean water, and no facilities for disposal of human and
23
other wastes, which were allowed to accumulate as a perpetual hazard to health. Crowding aided
the transmission of infectious diseases, some of which can only be sustained in populations
above a critical size. Pollution from smoke and other factory discharges contaminated the atmos
phere and the environment. Yet many cities, in Britain and in Europe, were slow to address these
problems, even when the necessary policies were well understood. Although the germ theory of
disease was generally accepted only after 1870, earlier explanations, such as the “miasma”
theory, also emphasized the importance of cleaning up the environment. And while money is
always a factor limiting public construction, these were periods of relatively rapid economic
growth. Indeed, the coexistence of rapid economic growth and mortality increase (as well as a
decrease in stature) during this period is regarded as something of a puzzle by economic
historians, see Haines and Kintner (1999), Schofeld and Reher (1991) and Fogel (1996), who are
typically so confident of the link between health and incomes that they often use measures of the
former—such as stature—as reliable indicators of the latter, Steckel (1994).
Szreter (1988) argues that the key to understanding the mortality transition in England lies is
local politics. Although the industrializing cities were in fact well supplied with fresh water, it
was used for commercial purposes, not supplied to homes. And the new entrepreneurial class,
rich though they were, saw no point in spending each others’ money for public sanitation which
had no obvious commercial benefit. It was only after political reform, and particularly the limited
political emancipation of working men, that new political coalitions could develop that made
sanitation and public health a priority.
This is a story of 19th Century England, not of the world today, and it is about politics, not
about income inequality. Yet central to the plot are the Reform Acts and their (limited) extension
24
of democracy which may or may not have (directly) reduced income inequality, but certainly
reduced political inequality. This increase in equality was not only valuable in its own right, it
subsequently helped reduce other inequalities, in health, and perhaps eventually in incomes, in
accord with the general thesis of Sen (1999). In a more contemporary context, it is hard not to see
political action, or at least the lack of it, as one of the reasons behind the low level of provision of
schools and clinics in Indian villages. Chattopadhyay and Duflo’s (2001) work in India shows
how the mandated representation of women as leaders of village councils (in a randomly selected
third of all Indian villages) has led to small but perceptible gains in public goods important to
women and children, particularly water, fuel, and roads (whose construction provides employ-
ment opportunities for women) though, perhaps surprisingly, male leaders invest more in
education.
2.5 Evolution, equality, and relative deprivation.
2.5.1 Evolution, stress, and inequality
It might easily be supposed that hierarchic, unequal societies are an inevitable part of the human
condition. Yet for the vast majority of our evolutionary history, humans lived in hunter-gatherer
groups that were not only not hierarchic, but aggressively egalitarian, Erdal and Whiten (1991).
As has long been argued, perhaps first by McKeown (1976, 1979), human health is maximized
when we live under the conditions under which we evolved, pursuing regular exercise (walking
10 to 15 miles a day, as foragers and hunters did), and eating low fat, low-salt, low-meat, sugar-
free, high fibre, largely vegetarian diets. By the same token, given that hierarchies and social
inequalities were unknown for most of our history, modem inequalities are likely to be a hazard
25
to our health. This argument is forcefully and eloquently put by Wilkinson (2000).
That foraging groups were egalitarian appears to be widely agreed. Such arrangements could
perhaps have come from lack of a technology for storing food. When a kill has been made, and
the meat is too much to be consumed at once, sharing and subsequent reciprocity are the only
mechanisms that can turn meat today into meat tomorrow. Members of groups that used such
mechanisms would therefore have a survival advantage over members of those who did not, so
that a preference for sharing, fairness and reciprocation may be evolved attributes. With the
invention of settled agriculture, with the associated ability to fill and hold food in granaries, as
well as to build herds of animals, egalitarian and reciprocal sharing was less efficient, and could
give way to hierarchies within which rich and powerful individuals could dominate others.
Although such systems and their industrial successors are vastly more productive than is
foraging, the benefits come at the price of a nagging and health-compromising outrage over the
loss of equality. And while humans will perhaps evolve to suit this new environment, we have
only given up foraging for a very short time, only 50,000 or so years of our perhaps million-year
history. Adaptation to the new environment has its benefits, in terms of production, longevity,
health, and population size, but it has a lingering cost that prevents our health from reaching its
full potential.
Wilkinson and others have begun to weave together a plausible story of the processes that
support such an account. Psychosocial stress is the main pathway through which inequality
affects health. Wilkinson draws a contrast between societies in which relationships “are
structured by low-stress affiliative strategies which foster social solidarity” on the one hand, and
societies characterized by “much more stressful strategies of dominance, conflict and submission.
26
Which social strategy predominates is mainly determined by how equal or unequal a society is.”
(Wilkinson, 2000, p. 4). Equality is seen as a precondition for the existence of stress-reducing
networks of friendships, while inequality and relative deprivation are seen as compromising
individual dignity, and promoting shame and violence. At the same time the biological mechan
isms through which chronic stress compromises health are beginning to be understood; excellent
surveys can be found in Sapolsky (1998), Brunner and Marmot (1999) and Wilkinson (2000,
Chapter 2.)
2.5.2 Income inequality and relative deprivation
The story outlined above is persuasive in many respects. That the social environment in which
we live helps determine our health is surely right, and the effects of psychosocial stress are now
well-documented in both human (particularly Whitehall civil servant) and animal populations.
Yet it is less clear why income inequality is the only, or even the prime villain in the piece. I have
made an attempt to explain why it might be so in Deaton (2001). In this account, I treat income
relative to other members of a reference group as the key variable, and hypothesize that stress on
each individual is proportional to the total amount of income accruing to community members
with higher incomes, expressed in units of group mean income. Figure 6 illustrates. It shows the
cumulative distribution, F(x), of income x for the reference group, and calculates the mortality
risk, on the vertical axis, for a person with income x, on the horizontal axis. This person is hurt
by all the incomes of the people above him, to the right of x. Consider one such person, with
income y. The burden that the people around y place onx is in proportion to their distance above
x, or the distance y -x, so that for the group in the strip from y to y + dy, the contribution to the
27
mortality risk of someone with income x is the shaded area, which is the product of the number
of people, dF(y), and the income difference y-x. The same sort of effect operates for all
incomes above x, so that the total risk is the total area above the curve from point A up to the
highest income.
This area, divided by the mean of group income, was proposed by Yitzhaki (1979) as a
measure of relative deprivation, a term frequently used by Wilkinson in his discussions of social
stress. Although the meaning here is more specialized (and precise) than that intended by
Wilkinson, the formalization appears to be well-chosen, in that it generates many of the effects
that he proposes. If relative deprivation is drawn as a function of income, we get curves like
those in Figure 7, drawn for a selection of states in the U.S. The theory can be tested by matching
these curves to the relationship between mortality and income for each state.
The measure of relative deprivation defined above and illustrated in Figures 6 and 7 has a
number of interesting properties. First, relative deprivation, and thus mortality risk, is lower for
people with higher income. Second, the rate at which relative deprivation declines with income is
lower at higher incomes. This convex shape is then consistent with the relationship between
mortality risk and income that we find in data such as the NLMS, and which is illustrated in
Figure 2. Third, the amount of relative deprivation at any given level of income depends on the
amount of income inequality in the group. Indeed, the amount of relative deprivation for someone with mean income is itself an inequality index, the relative mean deviation, sometimes called
the Pietra or Robin Hood index. More unequal “states” in Figure 7, like the District of Columbia
and California, have higher relative deprivation curves than Maine or Florida, where income is
more equally distributed. Fourth, if we average relative deprivation over all the members of the
28
group, so as to get group mortality risk, we get another measure of inequality, the gini coefficient.
This conforms almost exactly to Wilkinson’s (1998) statement that “income inequality summarizes the health burden of individual relative deprivation.”
In summary, the relative deprivation theory of mortality risk has three important implications:
(i) within groups, mortality risk is a convex and declining function of income; (ii) conditional on
an individual’s relative income, inequality matters for individual health; and (iii) for groups,
mortality risk is independent of group income, but is directly related to the gini coefficient. The
results of testing this theory are postponed to Section 3, where they can be presented in the
context of other, related work on mortality differences across the US states.
2.6 Inequality and crime
Crime is often treated as a public health issue. This is clearly appropriate in the case of homicide
or other crimes against persons, but there is also a great deal of violence and stress associated
with other crimes, such as theft or drug dealing. Much of the work on crime and inequality is
unabashedly empirical, but there is some theory to provide guidance. The seminal paper on the
economics ofcrime is Becker (1968), who proposed that criminal activity should be seen as the
outcome of an optimal choice in which the expected benefits of crime, money from theft, or the
satisfaction of murdering an enemy, are weighed against the expected costs, including the costs
of legitimate activity foregone, as well as the costs of punishment weighted by the probability of
apprehension. Criminals are no different from everyone else so that we would all be criminals
given the right incentives. Becker’s theory was used by Ehrlich (1973) to examine crime rates
across the states of the US, who then needed an operational measure of the benefits of crime
29
relative to the opportunity cost in terms of legitimate activity in the labor market. Ehrlich argued
that the benefit of theft is likely to be related to average (median) community income or wealth.
Because only those with low incomes, and thus low opportunity cost of legitimate activity, were
likely to find the benefits of crime greater than the costs, Ehrlich included in his regression the
fraction of the population whose incomes were below half median state income, calling this a
measure of income inequality. (It would more usually be thought of as a poverty measure.)
In general, there is no reason to suppose that only the poor will commit crimes. Instead, we
might model the net benefit of a contemplated crime for person A against person B as propor
tional to the income difference between them. If so, the incentive for person A to commit a crime
is proportional to the sum of all such income differences above him, which takes us back to the
measure of relative deprivation proposed in the previous subsection. Within any community, the
average of such incentives over all people is simply the gini coefficient of income, a result that
might have been approached directly by noting that the gini coefficient is the ratio to the mean of
(half the) the average of all income differences between people. If the rate of apprehension and
mean income is the same in all communities, it would therefore be plausible that higher inequal
ity communities would show higher crime rates.
The probability of apprehension will generally depend on the level of crime, which may
obscure any simple relationship with income inequality. Indeed, as Wittenberg (2000) shows, the
interaction of crime and crime prevention can generate multiple equilibria, including outcomes
where there is a great deal of inequality, but very little crime. When the potential victims are rich
enough to make attractive targets, they may also be rich enough to afford extensive security
measures. Whites in apartheid South Africa were well protected against the crime that might
30
have been expected given the extraordinary levels of income inequality between them and the
Black majority population. At the other end of the spectrum, murder is common among hunter
gatherer bands precisely because, in these egalitarian structures there is no central control, no
hierarchic structure, and no police.
One might also challenge the basic premise of Becker’s analysis, that criminals are no
different from the rest of us, but simply face different costs and benefits. That everyone will turn
to crime given sufficient incentives may be doubted, as in Sen’s vehement denial that inequality
leads to crime, “crime needs some assertion and confidence and you can’t do it when you are
down on your knees. It has been seen that even during famines, which are periods of massive
inequality, there is no increase in crime. I have seen the Bengal famine, and even there, people
would die outside sweet shops, but not a glass would be broken,” Business Standard (2001).
Nevertheless, as we shall see in Section 3, the data often show a quite robust correlation
between homicide and inequality.
2.7 Income inequality as a consequence of ill-health
It is clear that not only does economic status influence health status, but also that health status
affects economic status. Unfortunately, and with the notable exception of the nutritional wage
model, the literature has tended to concentrate on one or other direction of causality, without
paying enough attention to the interaction between the two, something that deserves serious study
in its own right. At low levels of income and nourishment, the effects of disease and food (mean
ing income) should not be thought about separately. Inadequate nourishment compromises the
immune system and makes it less able to resist infection; ensuring adequate income is likely to
31
be important even for a strategy that focuses mainly on public health. At the same time, although
income may be sufficient to guarantee adequate nutrition, nutrition will not improve nutritional
status if the body is unable to absorb the food because of chronic diarrhea or intestinal infection.
The provision of public health measures, clean water and waste disposal, is necessary even for a
strategy that focuses mainly on growth and income. Even more obviously, and in rich countries
as well as poor, the ability to work is compromised by ill-health. People with low incomes may
be more likely to contract a disease, they are less able to spend money to mitigate its consequ
ences, and they may find it difficult to comply with complex and time-intensive medical regimes
or to seek medical attention in the first place.
When interactions are important, the distribution of income will depend on the distribution of
health. Any measure that reduces the spread of health conditions across the population will
narrow the distribution of income. In particular, anything that helps people recover more rapidly
from an illness will reduce the persistence of ill-health, which reduces the long-term variance of
health across the population. Better insurance arrangements, or better and more widespread
clinics are obvious candidates to reduce such persistence, and so will not only improve popula
tion health (if they work at all), but also improve the distribution of income. Clean water, whose
lack affects the poor more than the rich, will improve the incomes of the poor relative to the rich,
and reduce income inequality. Malaria eradication campaigns and vaccination drives will have
the same effect, not only improving population health, but also narrowing the distribution of
income. This would be true even in a Lewis world in which there is an unlimited supply of labor
at the subsistence wage. Better health may not improve the wage rate, and cannot do so under the
Lewis assumption, but it can enable more people to work at that wage.
32
HE ^>OO
10 Z24
pol
> s,
Income differences between countries are also affected by their differences in population
health. In consequence, the speed at which new health technology is transmitted from the
industrialized to developing countries narrows their relative population healths as well as their
relative incomes, see Sachs (2001). Current debates about international pricing of drugs and
patent protection are important in this context. Whether the faster transmission of drugs and
vaccines will improve the distribution of income within poor countries depends on the prevalence
of disease within the income distribution. For AIDS, prevalence in poor countries is (still) higher
among the rich, while for malaria or tuberculosis, prevalence is higher among the poor. In the
past, health innovations have often widened health inequalities when first introduced; examples
are better sanitation a century or more ago, Preston and Haines (1991), tobacco and lung cancer
more recently, and child health improvements in contemporary Brazil, Victora et al (2000).
Faster transmission of best-practice healthcare may therefore widen the income distribution
within the receiving countries, at least in the first instance.
3. Empirical evidence on inequality and health
3.1 Measuring income inequality
Income inequality is not an easy thing to measure so that, before looking at the evidence on the
effects of inequality, it is worth starting with a discussion of what we know about inequality
itself. There are both conceptual and practical issues. Comprehensive treatments of the theory of
inequality measurement were developed in the 1970s by Atkinson (1970) and Sen (1973), the
latter updated in Foster and Sen (1999). Although there are a number of axioms on the nature of
inequality that are broadly accepted, these will not always be sufficient to permit us to make
33
unambiguous inequality rankings between any two distributions of income. Instead, the axioms
induce a “partial ordering” whereby we can sometimes rank one distribution as more unequal
while, in other cases, we can judge the inequality of distributions only by choosing a specific
inequality measure, with different measures giving different results. In particular, different
inequality measures give a different emphasis to different parts of the distribution. For example,
the gini coefficient is more sensitive to inequality (or to measurement error) at the top of the
income distribution, whereas measures that work with the logarithms of income, such as the
Theil measures, or the variance of logarithms, are quite sensitive to inequality at the bottom.
Although neither the conceptual issues nor the choice of inequality indicator are probably the
most important issues in the health literature, it should be noted that many of the indexes that are
used in the public health literature do not satisfy even the generally accepted axioms. For
example, the Robin Hood index (more usually known as the relative mean deviation), Kennedy
et al (1996), is unaffected by transfers between individuals on the same side of the mean. If a
transfer program were to transfer incomes from those just below the median to those near the
bottom, the Robin Hood index would not change, even though there would have been a real
reduction in inequality (and very likely a decrease in mortality risk too.) Perhaps most of the
public health work uses as its inequality measure the share of income accruing to the bottom x
(often 50) percent of the population. Once again, transfers within the bottom x percent, or within
the top 1 -x percent, will leave the measure unaffected, even though such transfers are capable
of having a substantial effect on income inequality more broadly.
There has also been a good deal of discussion about the appropriate definition of household
income, and in particular the treatment of household size. A standard procedure in the economics
34
J
literature is to “equivalize” household income by dividing income by some measure of household
size, either household size itself, or the number of equivalent adults, for example the number of
adults plus half the number of children, or the square root of the total number of people in the
household. Such per equivalent measures attempt to capture the resources available to each
person in the household, and recognize that, at the same level of income, members of a larger
household are worse off than members of a smaller household. (Even so, it should be noted that
there is some evidence, Elo and Preston, 1996, that, conditional on family income, larger family
size may not increase mortality.) When income inequality is calculated, it is also important that
equivalized income be assigned to individuals, and that inequality be calculated over persons, not
households. These apparently technical details can sometimes have serious effects on the
measurement of income inequality, and their treatment in the public health literature has often
been cavalier.
Conceptual problems are dwarfed by measurement problems. Income itself is hard to
measure, and the difficulties multiply when we try to measure income inequality. Measurement
error in income, even if it has little effect on the measurement of mean income, will inflate the
measured variance and measured income inequality. The measurement of income is sensitive to
survey design, particularly to the choice of the reference periods for income (longer reference
periods give lower measured inequality), and to exactly how the income question is asked. The
degree of disaggregation of income categories is important, as is whether incomes are reported as
a number, or in a set of predefined ranges such as less than $10,000, $10,000 to $20,000, and so
on, up to some open ended top category, such as more than $50,000. The choice of cutoff points
for the ranges is important, particularly the top band which effectively limits the highest income
35
that can be reported. Some surveys permit people to report negative incomes (losses from busi
ness activities) and some do not. Some surveys collect data on income, and some on consump
tion; the latter is almost always less unequally distributed than the formal. The response rate from
surveys varies over space and time, and richer households are typically less likely to agree to
participate, in many cases because they live in communities where the enumerators cannot reach
them, see Groves and Couper (1998). In rich countries, and in some not so rich countries,
response rates have been falling over time, perhaps in response to the increasing competition
from market researchers. There are also differences across countries in the degree to which
people are prepared to cooperate with government surveys. Response rates in the US are typically
much higher in the US than in Western Europe.
Because different countries use different survey instruments, and have different survey
protocols, useful cross-country comparisons of income inequality require detailed knowledge of
the specific surveys. Similar issues sometimes arise with comparisons over time within one
country, even where the statistical service is of the highest quality. Specifically, the US Census
Bureau, in the summer of 2000, decided that the large increase in household income inequality
between 1992 and 1993, which it has previously presented as real, was in some unknowable part
due to changes in survey methodology, particularly changes in the highest level of income per
mitted in the questionnaires, as well as the introduction of computer-aided interview technology,
Jones and Weinberg (2000). In consequence, the US no longer has a consistent continuous time
series of household income inequality. That the US is worse (as opposed to more transparent)
than other countries seems unlikely.
International data on income inequality come from a number of standard sources. Perhaps the
36
best is the Luxembourg Income Study (LIS) which contains information on the distributions of
disposable income for 25 (wealthy) countries over a period of 20 years, although not all countries
have data for all years, see Gottschalk and Smeeding (2000, pp. 273—4). The LIS permits access
to the micro data from broadly comparable income data for the covered countries. Note that the
underlying surveys do not use the same questionnaire, so that the comparability is not perfect, nor
are response rates the same for all countries. Nevertheless, the data are well-documented and
have been widely analyzed, so that their properties are well understood. Some authors have taken
income inequality data for industrialized countries from other, non-LIS sources, such as Sawyer
(1976); these are now superceded by the LIS.
Matters are a good deal more difficult for income distribution data from the large numbers of
developing countries that are not covered by the LIS. For many years, popular sources of income
distribution data were Jain (1975) and Paukert (1973), which are essentially compendia of
inequality estimates then available in the World Bank and International Labor Office, respect
ively. In more recent years, research on international patterns of income inequality has been
transformed by the availability on the World Bank website of the inequality data assembled by
Deininger and Squire (DS) (1996). These data, which have seen widespread use, contain more
than 2,600 observations on gini coefficients (and many quintile shares) for more than 100
developed and developing countries for dates between 1947 and 1994. To be included in the DS
data set, estimates have to come from an identifiable source, be national in coverage, and be
based on either consumption or income. (Which comes from which is identified.) A subset of the
observations are labeled “high-quality” and these have been widely (and mostly uncritically) used
in a large number of papers, including papers on income inequality and health. Much of the high37
quality data comes from industrialized countries, so that many researchers interested in develop
ment have used at least some of the “non-recommended” data.
While DS’s data and documentation are a great improvement over what was previously avail
able, they do not support the uncritical use that has been made of them, as shown in an important
study by Atkinson and Brandolini (1999). Atkinson and Brandolini focus their attention on the
subset of the DS data for the OECD countries, for which there are good, well-documented
surveys (including the LIS) which can be used for comparison. DS’s “high-quality” estimates do
not do well in this comparison, either across countries, or in some cases over time within
countries. For example, DS shows Sweden as one of the more unequal countries in the OECD,
with more income inequality than the UK, whereas in the LIS (as reported in Gottschalk and
Smeeding, 2000), Sweden has the lowest income inequality and the UK the highest apart from
the US. In some cases, such as Germany, the DS time series of inequality is quite different from
that computed directly from the surveys. Although the DS data may be more reliable for poor
countries than for rich, it is unlikely, especially since the poor countries contain a much larger
fraction of the data that are not endorsed by DS themselves.
We are currently in the position of not having any consistently reliable set of data on income
inequality outside the countries covered by the LIS. This is in spite of the existence of the 50 or
so surveys that have been collected under the aegis of the World Bank’s Living Standard
Measurement Survey (LSMS) which was set up in 1980 with the original purpose of generating
comparable data on income distribution for a wide range of countries. While the LSMS surveys
are broadly comparable, the questionnaires are not identical across surveys, and some have
differed a great deal. Nevertheless, a research program within the World Bank could be set up to
38
use the LSMS surveys, together with the other unit-record data sets available, to generate a series
of inequality measures for which the quality guarantee could be based on a detailed analysis of
both survey protocols and the individual data.
Research on inequality and health has also used data on measures of inequality for areas
within countries. In principle, such measures are less problematic, if only because they are
usually calculated from national surveys using a uniform survey instrument so that, even if there
are errors, the patterns across areas may not be much affected. One problem is sample size,
especially for small areas. Inequality measures are usually less precisely estimated than means so
that, for example, the US Bureau of the Census publishes estimates of mean income by state
using the Current Population Survey, which has a sample size of around 50,000 households each
year. However, it publishes inequality measures by state only for three year moving averages.
Several developing countries also have regular, national household surveys that are large enough
to support considerable disaggregation; India, Indonesia, and Pakistan are examples. The Indian
National Sample Survey (NSS) collects detailed consumption data from more than 120,000
households every five years or so, and these surveys are designed to be representative for more
than 70 regions of the country. Yet even this survey does not support the measurement of income
inequality for districts, the level at which the Indian census publishes much of its data on child
mortality.
Most household surveys have a two-stage stratified design in which, at the first stage, primary
sampling units (PSUs) are randomly selected with probability proportional to population size.
These PSUs are typically small geographical units, such as villages or census tracts. Within each
PSU, the same number of households are selected so that, over the two stages together, each
39
household in the population has an equal chance of being selected into the survey. In some
studies, investigators have calculated measures of local income inequality based on the house
holds in each PSU. This procedure is obviously dangerous when there are only a few households
in each PSU and, even when this is not the case, respondents within PSUs are sometimes not
randomly selected, so that the relationship between the sample estimate and its population
counterpart cannot be assessed. Note too that PSUs are selected for statistical convenience, not
analytical meaning, and frequently do not correspond to any sensible definition of a community.
3.2 Cross-country studies of income inequality and health
Cross-country studies have played an important part in the literature on income inequality and
health. Preston’s (1975) seminal analysis looked at international patterns of GDP and life
expectancy, and it was on the basis of his findings that Preston suggested that there should be a
negative relationship between income inequality and health. Rodgers (1979) and Flegg (1982)
were early studies that followed Preston’s lead, explicitly looking for (and finding) effects of
income inequality on mortality. Rodgers used the Paukert (1973) data for 56 (unnamed) countries
and, controlling for income and other variables, found hazardous effects of inequality on life
expectancy at birth, life-expectancy at age 5, and on the rate of infant mortality, with the last only
significant in the developed countries in the sample. Flegg (1982) looked only at child mortality,
and found significant effects of income inequality on child mortality in developing countries
using the Jain (1975) data. These authors, like Preston, thought of the nonlinearity in the
relationship between income and health as the basis for their results, and did not propose any
direct effect of income inequality at the microeconomic level. Such a direct effect was found in
40
by Waldmann (1992) who used UN and World Bank sources, supplemented by income
inequality data from Jain (1975), to investigate infant mortality on a cross section of up to 57
developing and developed countries. As expected, he found that, conditional on mean income,
the share of income going to the poorest 20 percent of the population decreased infant mortality,
and more surprisingly, that the share of income going to the top 5 percent increased infant
mortality. This is a direct effect of inequality; the infant mortality rates among the poor increase
when the rich get richer, even when their own incomes do not suffer.
Perhaps the single most cited finding in the literature is Wilkinson’s (1992, 1994, 1996)
demonstration of a relationship between income inequality and life expectancy across a number
of industrialized countries, not only in levels but, more impressively, in changes over time.
Countries, such as France and Greece, that narrowed their income distributions by reducing
relative poverty, increased their life-expectancies, while those, such as the UK and Ireland,
whose income distributions widened, fell behind, Wilkinson (1996, Figure 5.4). Wilkinson
interprets these results as showing that, as countries become wealthier and move through the
epidemiological transition, the leading cause of differences in mortality moves from material
deprivation to social disadvantage. Material deprivation provokes poverty and infectious disease,
while social disadvantage provokes stress and chronic disease.
Later research has cast considerable doubt on the robustness and reliability of many of these
findings. As expected, one of the main difficulties lies in the unreliability of the data on income
inequality. For example, using the Deininger and Squire data, Gravelie, Wildman and Sutton
(2000) fail to replicate Rodgers (1979) results for developed and developing countries, while
Mellor and Milyo (2001) use a sample of 47 developing and developed countries in 1990 and
41
find that the positive correlation between the gini coefficient and infant mortality vanishes once
secondary school enrolment is controlled for, while the negative correlation between income
inequality and life expectancy is eliminated by controlling for income per head. Mellor and
Milyo also fail to replicate Wilkinson’s results for developed economies. Although the DS data
have their own problems, the original results are clearly not robust. The same is true of
Waldmann’s findings; Baumbusch (1995) replicated Waldmann’s analysis using data of the same
vintage, but found that income accruing to the top 5 percent reduced infant mortality once the
data were updated from the 1993 edition of the World Bank’s World Development Report.
The single most important and careful study of the LIS countries is by Judge, Mulligan and
Benzeval (1997), who emphasize the poor quality of the data in previous work, and use the LIS
data in their own examination of life expectancy and infant mortality in Australia, Belgium,
Canada, Finland, France, Germany, Ireland, Italy, Luxembourg, The Netherlands, Norway,
Sweden, Switzerland, the U.K., and the U.S. In these data, which are the best international data
currently available, the correlation between the gini coefficient and life expectancy is -0.17,
insignificantly different from zero, and neither the gini nor other measures of income inequality
are close to significance in any of the regressions explaining life expectancy. The situation is
somewhat different for infant mortality rates, where there is a significant positive (i.e. harmful)
effect of the ratio of the 90th to the 10th percentile. This measure of inequality exerts a significant
effect in several of the regressions, though it becomes insignificant when controls are added for
the negative effects on mortality of female labor force participation. In these data, the raw
correlation between infant mortality and inequality is driven largely by the US, which is very
unequal and has relatively high infant mortality.
42
If these results are not entirely definitive, it is because the LIS data, although better than any
other, are neither fully comparable nor fully accurate. The debate between Wilkinson (1998) and
Judge, Mulligan and Benzeval (1998) has focused on differences in response rates across the LIS
surveys, and their possible effect on the results. It is also possible that, as with the difference
between infant mortality and life-expectancy, there will be links between specific causes at
specific ages and the plausibly associated measures of inequality, see for example Mclsaac and
Wilkinson (1997). Yet, it is surely time to agree that there is currently no evidence that income
inequality drives life expectancy and general adult mortality within the industrialized countries. It
remains to be seen whether this means there is no relationship, or whether there is a relationship
that is being obscured by still inadequate data. Judgement on that depends a good deal on
whether there exists a relationship between income inequality and health in other contexts, on
which more below.
That, conditional on income, there should be a cross-country relationship between infant
mortality and income inequality, at least in poor countries, is both theoretically plausible, and
rather better supported by the (admittedly inadequate) data that are available. The plausibility
comes from recent work from the World Bank which, following the methodology pioneered by
Filmer and Pritchett (1999), has used Demographic and Health Surveys around the world to
construct a synthetic measure of wealth, which is then used to explain infant mortality rates, see
Gwatkin (2000) for an overview. The measure of wealth is an index based on the ownership of
various durable goods, and while the measure is undoubtedly correlated both with actual wealth
and income, we have no way of calibrating the transformation, and thus of using the results to
relate income to child health. Nevertheless, the results show very strong gradients in child health,
43
with infant mortality rates heavily concentrated at the bottom of the distribution. Wagstaff (2000)
uses nine (mostly) LSMS surveys from developing countries to calculate child mortality rates by
quintile of equivalent consumption, and shows that child and infant mortality rates typically
decline most rapidly between the bottom and second quintile. Whether these results imply that
infant mortality rates are convex in income depends on the degree of convexity of the relation
ship between income and the asset index, and on the density function of equivalent consumption.
but that infant mortality is concentrated at the bottom of the income distribution seems likely.
Yet there is also some evidence on the other side. In particular, Murthi, Guio, and Dreze (1995)
find very little effect of poverty on child mortality across districts in India once they control for
other factors, most importantly female literacy and urbanization.
To the extent that the DS data are accepted, there is a good deal of empirical evidence from
developing countries linking infant and child mortality to the DS measures of income inequality
conditional on the level of GDP per head and a range of other variables, for example in Pritchett
and Summers (1996), Filmer and Pritchett (1999) and Hales et al (1999). Whether this evidence
extends to adult mortality and life expectancy is difficult to know, not only because of the data
difficulties with income inequality, but because of the quality of the data on adult mortality. Few
poor countries have complete registration systems for deaths, so that good evidence on adult
mortality (or life expectancy at age 5, for example) is hard to come by for most developing
countries. In a few cases, such as India, there are sample registration surveys, and some data on
adult deaths can be gleaned from the Demographic and Health Surveys. But for most countries,
data on life expectancy are extrapolated from the data on infant mortality rates, and contain little
additional information. An exception to this generalization comes from the countries of Eastern
44
Europe and the former Soviet Union, where life expectancy has been falling as income inequality
has increased (see Marmot and Bobak, 2000, Fig. 3, which shows a 12 countries correlation
coefficient of-0.63.) As is widely recognized, the Eastern European experience is difficult to
interpret because so much else has been going on, so that it is hard to isolate the effect of income
inequality.
Finally, there are a number of cross-country studies that link other health outcomes to income
inequality. Steckel (1995) finds a relationship between human stature (a measure of cumulative
nutritional status) and income inequality on a sample of developed and developing countries
using the income distribution from Jain (1975). Over (1999) looks across cities in the developing
world and finds that the US Census Bureau’s estimates of HIV infection rates are positively
related to the DS measures of countrywide income inequality. He interprets his findings in terms
of upper income men demanding the services supplied by lower income women, and these results
perhaps come closest to providing substance to Farmer’s (1999) contention that disease occurs
along the “fault lines” in the income distribution. Gaarder (2001) argues that income inequality is
likely to worsen the health consequences of pollution because the poor have lower baseline
health and are therefore more susceptible. She includes the gini coefficient in a meta-analysis of
previous estimated effects of particulate concentration on mortality at various sites around the
world and finds significant positive effects. Fajnzylber, Lederman, and Loayza (2000) find a
significant relationship between DS gini coefficients and both homicide and robbery rates for a
group of 45 (for homicides) and 34 (for robberies) developed and developing countries. A good
deal of this is driven by Latin American countries, where both crime and inequality are very high.
Using data on 17 countries in the Americas, and gini coefficients from DS, Mujica et al (2000)
45
confirm the positive correlation (0.55) between homicide and income inequality, but find a
negative correlation (-0.78) between suicide rates and income inequality.
3.3 Within country area studies of income inequality and health
As skepticism has grown about the international relationship between income inequality and
health, attention has switched to studies within countries, particularly of mortality and income
inequality across the states of the US. Two studies, by Kaplan et al (1996), and by Kennedy,
Kawachi, and Prothrow-Stith (1996a, b), both published in the British Medical Journal, and
inspired by Wilkinson’s (1992) cross-country work, found a relationship across the states
between various measures of income inequality and age-adjusted all cause mortality, as well as a
number of other measures, including infant mortality rates, deaths from cancer, coronary heart
disease, homicide, as well as disability, low birth weight, and crime. Kawachi and Kennedy
(1997) established that the results were robust to the choice of inequality indicator, while Lynch,
Kaplan and Pamuk (1998) extended the results to 282 Metropolitan Statistical Areas (MSAs) in
1990, finding that the loss of life from income inequality “is comparable to the combined loss of
life from lung cancer, diabetes, motor vehicle crashes, HIV infection, suicide, and homicide in
1995.” Kawachi, Kennedy and Prothrow-Stith (1997) argue that income inequality works by
reducing social capital, in particular the degree of trust between people, a (very poor) state-level
measure of which is constructed from the General Social Survey. Such an account is very much
in the spirit of stories of psychosocial stress within unequal social structures. In support of this
explanation, Wolfson et al (1999) estimate the degree of nonlinearity in the income to mortality
curve using the NLMS (as in Figure 2) and show that the effects of income inequality on state
46
mortality rates are too large to be explained by the nonlinearity argument alone so that there must
be some direct effect of income inequality on individual mortality. The implications of these
results for economic policy have not gone unnoticed, see for example Kaplan and Lynch’s (2001)
editorial in the American Journal ofPublic Health entitled “Is economic policy health policy?”
These within nation results do not suffer from the same data problems as do the international
comparisons. Income inequality is usually measured from incomes collected in the census, which
is administered in the same form to all households in all states. Nor is there any question about
the existence of the correlation. Figure 8, taken from Deaton (2001) shows a typical scatter plot
between the log odds of age-adjusted mortality (the log of the ratio of the fraction dying to the
fraction not dying) on the vertical axis, and the variance of the logarithm of household income
per equivalent, with equivalents defined as 1 for adults and 0.5 for children aged 18 and less. The
District of Columbia is included as the 51st state and, although it is an outlier in the sense of
having higher income inequality and higher mortality than any state, it lies along the regression
line defined by the other observations.
Nevertheless, there are serious questions about whether the correlation between income
inequality and mortality is robust through time, and whether it comes from the effects of income
inequality or some other factor that is correlated with it. Mellor and Milyo (2001) use data tor the
48 continental states from five census years, 1950, 1960, 1970, 1980, and 1990, and reproduce
the strong hazardous effect of the gini coefficient on all cause mortality when only year dummies,
the age composition of the state, and median income are included as controls. The inclusion of
controls for the average level of education in each state eliminates the significance of the gini
coefficient and, once the authors include controls for the fractions of people in each state who are
47
urbanized and who are black, the gini coefficient attracts a negative sign, though one that is not
significantly different from zero. Similar reversals are found for the fraction of births that are
low-birthweight while, over the five decades, there is no relationship across states between
deaths from cardiovascular disease, from malignant neoplasms, or from liver disease. Indeed, for
the first two, income inequality has a negative and significant relationship with deaths once
controls are entered for income, education, race, and urbanization. Only for homicides and, to a
lesser extent, infant mortality and deaths from accidents, is the gini coefficient a risk factor
conditional on the other controls. Mellor and Milyo also subject the hypothesis to a much more
stringent test, looking at the relationship between 10 and 20 year changes in mortality and the
corresponding changes in income inequality. This is perhaps too severe a test because it places a
great deal of weight on the timing of the link between income inequality and mortality. Even so,
it is worth noting that, with one exception, none of the income inequality to mortality relation
ships survives the test. The exception is homicide, where the relationship with income inequality
is well-determined and holds over time as well as in the cross-section.
Related robustness issues are reported in my own work, Deaton (2001). A particular concern
is the pooling of data across racial or ethnic groups with different incomes and different mortality
rates. In the US, blacks have higher mortality rates than whites, and lower incomes, so that states
with a high fraction black tend to have higher mortality rates as well as higher income inequality.
As can be seen from Figure 8, such states tend to be predominately in the South where many
other special factors are likely to operate. If data are pooled for 1980 and 1990, the log odds of
age-adjusted mortality responds to the gini coefficient with a coefficient of 1.7 for males, and 1.1
for females. In the same regression, the mean of the logarithm of equivalized income reduces
48
mortality for men, but barely significantly, and not at all for women. To illustrate the size of the
effect of inequality, the 1990 gini coefficients for Louisiana and New Hampshire were 0.47 and
0.40 respectively, which would account for a 12 percent difference in mortality rates, more than
half of the difference shown in Figure 8.
If we now confine the calculations to white mortality alone, so that we no longer have the
mechanical effects described above, the coefficient on the gini drops to 1.1 for men and to 0.6 for
women, about a third lower than for all-race mortality. Nevertheless, these effects remain
significantly different from zero, and still show that inequality is a health hazard for the white
population. If we recalculate the gini coefficients so as to measure only inequality among whites,
the effects are further reduced, to 0.6 for men, and 0.4 for women, and only the former is
(marginally) significantly different from zero. This result means that the effect of inequality on
whites comes, not from the inequality of white incomes, but from the inequality between whites
and blacks, raising the suspicion that the effect has more to do with race than with income
inequality. Such a suspicion is borne out by controlling for the fraction of the population that is
black in each state. It turns out that a high fraction black raises mortality rates among both males
and females (note that these are whites) and that conditional on race, income inequality has no
effect on mortality. At this stage, it is unclear why the fraction black should exert such a strong
effect on white mortality (black mortality is also higher in states where there are relatively many
blacks), though it might be argued that it is itself some sort of marker for the inequality that
characterizes race relations in the US. Even so, the effect is not one that works through income
inequality; once the fraction black is included in the regression, the gini coefficient has no effect.
There is an obvious concern here that I have simply replaced one invalid variable, income
49
inequality, with another, racial composition, and that both stand proxy for something else. This is
particularly the case with the state data, where there are at most 51 observations (or 102 observa
tions if we pool data from 1980 and 1990), and where it would be easy to confound racial
composition (or income inequality) with geographical factors, especially given the peculiar role
of the South. Nevertheless, Deaton and Lubotsky (2001) show that the results carry through to
the 287 MSAs that can be consistently identified between 1980 and 1990. These data can be used
to replicate the findings of Lynch, Kaplan, and Pamuk (1998), and to show that, once again, the
inclusion of racial composition eliminates (and sometimes even reverses) the effect of income
inequality. And because there are so many more MSAs than states, it is possible to work within
regions, and to show that whether we look at cities in the South, or cities in other regions, and
conditioning on city average income, white mortality is higher in cities where the fraction of
blacks is higher.
Once the fraction black is controlled for, the cross-state and cross-city mortality results help
elucidate another puzzle, which is why there is such a strong relationship between income and
mortality in the individual data, and so little at the state or city data. Controlling for the fraction
black, the state or city mean logarithm of equivalized income has a significant negative effect on
mortality rates, particularly for men. Although the effects are not as large as in the individual
data, the results suggest that the differences might well be eliminated by controlling for a fuller
range of other factors.
Controlling for racial composition also makes the results consistent with the findings of Ross
et al (2000) who find that, in contrast to the US, there is no relation between income inequality
and mortality for the 10 provinces and 53 metropolitan areas of Canada, where race is not the
50
i
■
salient issue that it is in the US. Yet that there should be no relationship across the states between
mortality and income inequality, either in the US or Canada, is surprising in light of the argu
ments about nonlinearity. For Britain, there appears to be no area study on income inequality and
health, though Ben-Shlomo, White and Marmot (1996) find that mortality in the 8,464 wards of
England is affected not only by an index of deprivation based on household characteristics, but
also by the within-area dispersion of the deprivation index. Again, this is what is to be expected
if the deprivation measure is more closely linked to mortality among high deprivation people.
Chiang (1999) looks at mortality rates in the 21 counties and cities of Taiwan in 1976, 1985, and
1995 using household survey data to calculate measures of income and income inequality. He
finds strong protective effects of income in 1976 and 1985, and little effect of income inequality,
but finds that the situation is reversed in 1995, at which date income inequality is a hazard, and
income has no effect. Chiang interprets his findings as support for Wilkinson’s idea that income
is important at low levels of income, and income inequality at high levels of income which, as we
have seen, is also consistent with a nonlinear effect of income, and no direct influence of
inequality. Regidor et al (1997) find no relationship between (a nonstandard) measure of income
inequality and the prevalence of long-term disability across the 17 regions of Spain.
It is widely believed that there is a link between income inequality and crime (including
homicide) in the U.S. I have already noted Mellor and Milyo’s (2001) finding that homicide was
the only negative health outcome that was robustly linked to income inequality in their tests, and
such findings have consistently appeared in the literature since Ehrlich (1973), see Hsieh and
Pugh (1992) for an oft-quoted review and meta-analysis.
There appear to be few relevant studies from developing countries, even where it would be
51
possible to do so, for example in the work on fertility and child mortality in India by Murthi,
Guio, and Dreze (1995) and Dreze and Murthi (2001). However, Dreze and Kheera (2000) find
that homicide rates across India are unrelated to measures of consumption inequality, but are
positively associated with the fraction of “missing” women. Although the authors do not make
the point, a link between homicide among men and the shortage of women invites a socio-bio-
logical explanation in terms of mating behavior. Pena, Wall and Person (2000) find that infant
mortality risks are higher among the poor, and higher still when the poor live in relatively
wealthy neighborhoods, which is consistent with a negative role for inequality.
3.4 Studies of income inequality and health using individual data
Studies using individual level data face different data problems from either the national or the
area studies. Yet they have the advantage of being able to look for a direct effect of income
inequality without having to handle the effects of inequality that work through aggregation. But
there are compensating difficulties. Because mortality is a rare event, large sample sizes are
required to give enough deaths to reliably estimate mortality rates. At the same time, those few
health related surveys that follow people from interview to death are typically very poorly
endowed with economic information, including incomes. Nevertheless, there are several surveys
in the US that have been used to look at the determinants of mortality at the individual level. The
National Longitudinal Mortality Study starts from data collected in the Current Population
Survey, mostly around 1980, and then uses the National Death Index to check whether members
i
of responding households are dead by each follow up date. Currently, around 1.3 million people
have been tracked for up to a decade. In principle, the CPS provides excellent and detailed
52
HfE -)OO
1UJ24
p0'
Ic
z
A/
i
economic information, but many of the rounds used for the NLMS were not the March surveys,
when income data are collected, and so contain only rudimentary information on household
incomes.
There are two other US health surveys with later merges of death certificate data, the
National Health Interview Survey (NHIS), which interviews around 50,000 households every
year, and the National Health and Nutritional Examination Survey (NHANES) the first round of
which surveyed more than 14,000 people between 1971 and 1975, for whom information on
deaths has been merged up to 1987. A final source of mortality data is the Panel Study of Income
Dynamics (PSID), which has followed around 5,000 households (and their children and splitoffs) since 1968. Because this is a panel survey, returning regularly to each household, deaths are
reported by surviving family members. All four of these surveys have been used to look at the
relationship between mortality, income, and income mortality.
Sweden also has a data set that works in same way as the NLMS in the US, though with even
more comprehensive data. Since 1975, Statistics Sweden has interviewed around 7,000 indivi-
duals each year in its Survey of Living Conditions, and these people have been linked, not only to
the national death statistics, but to income information from the national income tax statistics,
see Gerdtham and Johannesson (2001).
A second line of work has used, not mortality, but self-reported measures of health status.
The questions are included in a large number of surveys, if only because they are asked easily and
quickly. They ask respondents to rate their health on a five point scale from “poor,” to “fair”, to
“good” to “very good” and “excellent.” Many investigators convert this to a binary indicator of
poor health, corresponding to the “poor” and “fair” categories; such an indicator has been
53
validated as a powerful predictor of subsequent morbidity and mortality, even conditional on a
physician’s examination, Idler and Benyamini (1997). These questions are included in the NHIS,
in the Behavioral and Risk Factor Surveillance Study (BRFSS) and, from 1995 onwards, in the
Current Population Survey.
The interpretation of the individual studies, and of the extent to which they support a link
from inequality to health depends a great deal on who is doing the interpreting. Nevertheless,
there is general agreement that the results from these studies are weaker and more ambiguous
than the area studies. For example, Lochner et al (2001), using the NHIS and merged mortality
data, find only a small effect of state income inequality on mortality (relative risk of living in the
top five most unequal states compared with the 10 most equal states of 1.12). This effect, which
is estimated with controls for family income and the state level poverty rate, is only statistically
significant for near-poor whites. Fiscella and Franks (1997) find no effect of PSU level inequality
on the probability of dying in the NHANES follow-up, but there are real questions about whether
their measure of inequality—the share of income accruing to the bottom 50 percent of the
population—can be adequately measured from the PSU data in the NHANES itself, within which
respondents are not randomly drawn. Daly et al (1998) find no effect of state level income
inequality on individual five year mortality rates using data from the PSID. As I shall explain in
more detail below, state level income inequality also has no detectable effect on individual
mortality in the NLMS.
Studies by Soobader and LeClere (1999) who use the NHIS with merged census county level
inequality data, Kennedy et al (1998) using the BRFSS merged with state level inequality data,
Fiscella and Franks (2000) using the NHIS, and Mellor and Milyo (1999) using the CPS with
54
inequality data from the CPS at various geographical levels, all find some effects of income
inequality on self-reported morbidity. But the estimated effects are typically modest, and Mellor
and Milyo show that their effects are removed once controls are introduced for income and its
square, as well as for fixed state effects. (Note again that this last is a severe test; Mellor and
Milyo only have three years of CPS data, so they are effectively demanding a link between
changes in morbidity and changes in income distribution between 1995 and 1997). But LeClere
and Soobader (2001) also demonstrate considerable fragility in the results, showing that the
effects seem to work only for whites aged 18-44 in high inequality counties, and middle-aged
whites in very high inequality counties. There are no effects for other whites, nor for non
Hispanic blacks.
I have already noted the paper by Wolfson et al (1999) showing that the degree of curvature
in the NLMS (Figure 2) is insufficient to explain the large effects of income inequality on
mortality at the state level. In Deaton (2001), I address more directly the role of state income
inequality on mortality in the NLMS. The NLMS distinguishes seven income groups, so that at
the first stage of the analysis, I use a logit model to estimate the log odds of dying during the 10
year follow up as a linear function of age including dummy variables for each of the seven
income groups. These logits are estimated for white males and females separately, using data
only for those aged 18 to 75 at the time of first interview. (The log odds of mortality is approx
imately linear in age over this range.) In order to conduct a state level analysis, each of these
models is fitted to data for a single state, thus allowing inequality—or any other state level
effect—an unrestricted effect on the relationship between mortality and income. The first-stage
produces numbers for each state like the points shown as circles in Figure 1 so that, at the second
55
stage, it is possible to examine whether these points are higher in states where income inequality
is higher. Note that this two-stage procedure is as general as a single stage model in which
individual mortality is linked to state level data on income inequality.
My original concern was to test the model of relative deprivation presented in Section 2.5.2
above. This was done by comparing the effects of each income group in each state with the
predicted values from computing a relative deprivation curve as illustrated in Figure 7. Within
states, the relative deprivation story does well, outperforming a simple model in which income
itself accounts for the differences across income groups. However, the relative deprivation model
accounts for essentially none of the variation in mortality across states which, given the theory.
means that the gini coefficient does not predict interstate mortality differences in the NLMS data.
This finding is supported (even without controls for income) using the 1.3 million observations
in the full NLMS. Table 13 in Rogot et al (1992) shows no correlation across the states between
age adjusted mortality and income inequality, a finding that is in direct contradiction with Figure
9 and with the findings listed at the outset of Section 3.4 above. This contradiction is resolved (at
least in part) by the demonstration that, in these individual level data as in the aggregate state
level data, the fraction black in each state is a powerful predictor of white mortality.
Once again, there is no direct effect of income inequality. Because the racial composition of
states is such a strong predictor of mortality in the aggregate state-level and MSA data, as well as
in the individual data from the NLMS, it would be interesting to discover whether the Lochner et
al (2001) findings on mortality in the NHIS follow-up can also be attributed to racial composi
tion; given their partition of states into inequality groups, it seems likely.
Taking income and mortality together, the Swedish data used by Gerdtham and Johanesson
56
(2001) are probably of higher quality than anything currently available in the US. Gerdtham and
Johanesson use the 284 municipalities of Sweden as their communities, and examine individual
mortality for 41,006 individuals aged between 20 and 84 who were interviewed between 1980
and 1986 and whose mortality was followed-up until the end of 1996. Mortality is assessed
relative to individual income, community income, and community income inequality, with the
latter two measured from the survey data itself, a procedure which is subject to the reservations
raised above. As in all similar studies, individual income is strongly protective, even allowing for
education and a host of other variables, including initial health status, but neither inequality nor
mean community income appeared to have any effect. The last result is evidence against the
relative income hypothesis so that, once again, we are led back to the original model in which
health is an increasing nonlinear function of absolute income.
3.5 Inequality and mortality decline in the US and Britain
A final source of evidence comes from examining whether the increase in income inequality in
the 1980s in both Britain and the U.S. can be linked to mortality. Wilkinson (1996) argues that
for Britain, mortality rates for infants and for young adults fell less rapidly after 1985 than would
have been the case had income inequality remained constant. Figure 5.10 of Wilkinson (1996,
page 97) plots a time series of mortality, not only of infants, but also of children and young
adults, and shows that the sum of age-adjusted mortality rates fell less rapidly after 1985 than it
did in the decade from 1975 to 1985. These findings, together with the corresponding evidence
for the U.S. have recently been examined in Deaton and Paxson (2001b). Their results are as
follows.
57
There were large increases in income inequality in both Britain and the U.S. in the 1970s and
1980s. In both countries, inequality in family and household income increased from the early or
mid-1970s until around 1990, with (arguably) little increase but certainly no decline since. By the
early to mid 1980s, inequality had risen to new postwar highs and continued to increase, at least
until 1990. As pointed out by Wilkinson, the rate of decline of infant mortality was particularly
rapid in the decade from 1975 to 1985, and less rapid thereafter. The same is true in the US,
though the period of rapid decline starts somewhat earlier, in the late 1960s, and finishes earlier,
around 1980. In both countries, the rate of decline of mortality rates among young adults has
slowed steadily, and by 1985 mortality rates are either flat or actually rising. For infants and
young adults taken together, the rate of mortality decline has therefore been a good deal slower in
recent years than in the period before the increase in income inequality.
Even so, it is unlikely that income inequality has much to do with these mortality trends.
First, the episodes of rapid decline in the infant mortality rate are episodes, not trends. Prior to
the periods of rapid decline in each country, progress was slower, with a rate of decline compar
able to that after the end of the episode. Yet income inequality was not high prior to the onset of
the rapid decline, so that in neither country since 1950 has there been a consistent relationship
between income inequality and the rate of decline of infant mortality. Second, we know a good
deal about the causes of the decline in infant mortality, much of which can be attributed to
declines in perinatal mortality through new techniques for preventing the deaths of low-birth
weight babies. These techniques diffused more quickly in the US than in the UK, so that the
rapid decline in mortality started first in the US and its possibilities were more rapidly exhausted
there. There seems no reason other than coincidence to link the timing of this exhaustion to the
58
rise in income inequality. Third, among young adults, much of the increase in mortality is
attributable to HIV/AIDS, for which the rise in income inequality in the mid-1980s is not the
cause. Finally, if we look at mortality rates of adults aged 45 and above, there is a period of
unusually rapid mortality decline (particularly although not exclusively for men) that began
around 1970 (again a little earlier in the US), and continues to the present. So if income inequ
ality is hazardous for the young, it is protective for their elders! Once again, a more convincing
explanation lies in the increased use, first in the US and later in Britain, of life-saving technolo
gies for dealing with cardiovascular disease, angioplasty, coronary artery bypass grafts, and the
use of clotbusting drugs and even aspirin.
4. Summary and conclusions
The stories about income inequality affecting health are stronger than the evidence. Judging by
the explosion of interest and of citations, there is a strong appeal to the idea that, before the
epidemiological transition, income determines mortality while, after it, income inequality
determines mortality. That in poor countries, income protects against poor sanitation, unhealthy
working and living environments, poor nutrition, and a plethora of infectious diseases. That in
rich countries, where these evils are but distant memories, income inequality is an indicator of
the quality of social arrangements, of stress, and of mortality. Yet, as we have seen, even if it is
true that, at higher income levels, income inequality becomes more important as a cause of death,
there is no need to assume that the relationship between income and mortality changes with
economic development. If it is poverty, not inequality, that drives mortality, so that income has a
much bigger effect on health at low than high incomes, average income will eventually cease to
59
be associated with poor health, while the effects of inequality will endure for much longer
because, even in rich economies, there are some who are not so rich. Income inequality will
continue to affect mortality until everyone ceases to be poor, which happens long after average
income has risen out of the range of poverty.
But it is not true that income inequality is a major determinant of population health. There is
no robust correlation between life-expectancy and income inequality among the rich countries,
and the correlation across the states and cities of the US is almost certainly the result of some
thing that is correlated with income inequality, but that is not income inequality itself. The rapid
increases in income inequality in the 1980s have not been associated with any slowdown in the
rate of mortality decline. Studies of individual mortality and income inequality show no link,
except for one survey where the estimated effects are small and are confined to one population
group. Infant and child mortality in developing countries is primary a consequence of poverty so
that, conditional on average income, income inequality is important only because, given average
income, inequality is effectively a measure of poverty. But it is low incomes that are important,
not inequality, and there is no evidence that making the rich richer, however undesirable that may
be on other grounds, has any effect on the health of the poor or their children, provided that their
own incomes are maintained. The only exception to these generalizations is perhaps the case of
homicide, where income inequality itself appears to play a genuine role.
These conclusions are not different from those of earlier commentators, particularly Judge
(1995), Judge, Mulligan and Benzeval (1997) and Doorslaer and Wagstaff (1999). Yet they must
not be misinterpreted. They do not imply that the social environment is not important for indi
vidual health, let alone that individual health is determined by individual characteristics and the
60
provision of personal medical care. We know from Whitehall and from other studies that
positions in hierarchies matter, perhaps through an ability to control one’s life, but in any case
through some mechanism that works through relationships with other human beings. My own
empirical results have drawn attention to another social factor, the effects of racial composition
on mortality, something that remains to be fully investigated. And I have emphasized several
other cases where reductions in deprivation in one dimension, whether it be land ownership,
democratic rights, women’s agency, or income, will bring benefits not only in and of themselves,
but also to the relief from other deprivations, in this case particularly the deprivation of ill-health.
This is of course Sen’s (1999) theme in Development and Freedom, that relief from any one of a
number of interlinked deprivations, each of which is an important unfreedom in its own right,
helps promote relief from the others. This is quite different from a story in which income
inequality is the principal actor and main villain.
My conclusions carry a number of implications for the direction of future research. The most
obvious is that attention should be directed away from further attention to income inequality per
se. I have already emphasized the puzzling role of racial composition on mortality in the US,
about which nothing is understood. More generally, the urgent need is to refocus research to
investigate the role that income plays in promoting health. We need to know much more than we
do about whether the effects of income come from income itself, or from correlates, such as
education, wealth, control, or rank. We need to know why income is so important in the
individual level studies, and so apparently unimportant at the aggregate level. If income is indeed
directly protective, we need to know whether the effect is really nonlinear and by how much,
because it is this, and not any direct effect of income inequality on health, that determines
61
whether and by how much income redistribution can improve population health.
List of works cited:
Aghion, Philippe, Eve Caroli and Cecilia Garcia-Penalosa, 1999, “Inequality and economic
growth: the perspective of the new growth theories,” Journal of Economic Literature. 37(4),
1615-60.
Alesina, Alberto, Reza Baqir, and William Easterly, 1999, “Public goods and ethnic divisions,”
Quarterly Journal ofEconomics. 114, 1243-84.
Atkinson, Anthony B., 1970, “On the measurement of inequality,” Journal ofEconomic Theory.
2, 244-63.
Atkinson, Anthony B., and A. Brandolini, 1999, “Promise and pitfalls in the use of ‘secondary’
data sets: income inequality in OECD countries,” Nuffield College, Oxford, processed.
Available at http://www.nuff.ox.ac.uk/users/atkinson/ (forthcoming, Journal ofEconomic
Literature.)
Bardhan, Pranab, Samuel Bowles, and Herbert Gintis, 2000, “Wealth inequality, wealth
constraints, and economic performance,” in in Anthony B. Atkinson and Francois
Bourguignon, eds., Handbook of Income Distribution. Vol. 1, Amsterdam, North-Holland, pp
541-603.
Baumbusch, Andrew P., 1995, “Income inequality and infant mortality,” Princeton University,
Junior independent work, processed.
Becker, Gary S., 1968, “Crime and punishment: an economic approach,” Journal ofPolitical
Economy. 78, 526-36.
Benabou, Roland, 1996, “Inequality and growth,” in Ben Bernanke and Julio Rotemberg, eds.,
NBER Macroeconomics Annual, Cambridge, Ma. MIT Press.
Ben-Shlomo, Yoav, Ian R. White, and Michael Marmot, 1996, “Does the variation in socio
economic characteristics of an area affect mortality,” British Medical Journal. 312, 1013-14.
Bertoia, Guiseppe, 2000, “Macroeconomics of distribution and growth,” in in Anthony B.
Atkinson and Francois Bourguignon, eds., Handbook ofIncome Distribution. Vol. 1,
Amsterdam, North-Holland, pp 477-540.
Bidani, Benu, and Martin Ravallion, 1997, “Decomposing social indicators using distributional
data,” Journal ofEconometrics. 77(1), 125-39.
Binswanger, Hans and Mark Rosenzweig, 1984, Contractual arrangements, employment, and
wages in rural South Asia. New Haven, Yale University Press.
Bliss, Christopher J., and Nicholas H. Stern, 1978, “Productivity, wages, and nutrition,” Journal
of Development Economics. 5, 331-98.
Brunner, Eric and Michael Marmot, 1999, “Social organization, stress, and health,” in Michael
Marmot and Richard G. Wilkinson, eds., Social determinants ofhealth. Oxford. Oxford
University Press.
Business Standard (New Delhi), 2001, “Poverty decline figures need a relook, says Sen,”
February 28.
62
Chattopadhyay, Raghabendra, and Esther Duflo, 2001, “Women’s leadership and policy
decisions: evidence from a nationwide randomized experiment in India,” LI.M. Calcutta and
Department of Economics, MIT, processed. (March.)
Chiang, Tung-liang, 1999, “Economic transition and changing relation between income
inequality and mortality in Taiwan: regression analysis,” British Medical Journal, 319,
1162-65.
Daly, Mary C., Greg J. Duncan, George A. Kaplan, and John W. Lynch, 1998, “Macro to micro
links in the relation between income inequality and mortality,” The Millbank Quarterly, 76,
315-39.
Dasgupta, Partha, 1993, An inquiry into 'well-being and destitution, Oxford. Clarendon Press.
Dasgupta, Partha and Debraj Ray, 1986, “Inequality as a determinant of malnutrition and
unemployment: theory,” Economic Journal, 96, 1011-34.
Dasgupta, Partha and Debraj Ray, 1987, “Inequality as a determinant of malnutrition and
unemployment: policy,” Economic Journal, 97, 177-88.
Deaton, Angus, 2001, “Relative deprivation, inequality, and mortality,” Princeton University,
Research Program in Development Studies and Center for Health and Wellbeing, Working
paper, Princeton NJ. (NBER Working Paper No. 8099)
Deaton, Angus and Christina Paxson, 2001a, “Mortality, education, income, and inequality
among American cohorts,” in David Wise, editor, Themes in the Economics of aging,
Chicago University Press for NBER. (NBER Working Paper No. 7140)
Deaton, Angus and Christina Paxson, 2001b, “Mortality, income, and income inequality among
British and American cohorts,” Center for Health and Wellbeing, Princeton University,
processed.
Deaton, Angus and Darren Lubotsky, 2001, “Mortality, inequality and race in American cities
and states,” Princeton University, Center for Health and Wellbeing, in preparation.
Deininger, Klaus and L.yn Squire, 1996, “A new data set measuring income inequality,” World
Bank Economic Review, 10, 565-91.
Dreze, Jean, and Reetika Khera, 2000, “Crime, gender, and society in India: insights from
homicide data,” Population and Development Review, 26, 335-52.
Dreze, Jean and Mamta Murthi, 2001, “Fertility, education, and development: evidence from
India,” Population and Development Review, 27, 33—63.
Easterlin, Richard A., 1974, “Does economic growth improve the human lot? Some empirical
evidence,” in P. A. David and W. R. Levin, eds., Nations and households in economic
growth, Stanford, CA, Stamford University Press, pp 98-125.
Easterlin, Richard A., 1996, Growth triumphant', the twenty-first century in historical
perspective, Ann Arbor. University of Michigan Press.
Ehrlich, Isaac, 1973, “Participation in illegitimate activities: a theoretical and empirical
investigation,” Journal of Political Economy, 81, 521-65.
Elo, Irma and Samuel H. Preston, 1996, “Educational differentials in mortality: United States,
1979-85,” Social Science and Medicine, 42, 47-57.
Erdal, David, and Andrew Whiten, 1996, “Egalitarianism and Machiavellian intelligence in
human evolution,” in Paul Mellars and Kathleen Gibson, eds., Modelling the early human
mind, McDonald Institute Monographs, pp 139-50.
63
1
Fajnzylber, Pablo, Daniel Lederman and Norman Loayza, 2000, “Crime and victimization: an
economic perspective,” Economic^ 1, 219-78.
Farmer, Paul, 1999, Infections and inequalities: the modern plagues, Berkeley. University of
California Press.
Filmer, Deon, and Lant Pritchett, 1999, “The impact of public spending on health: does money
matter?” Social Science and Medicine, 49, 1309-23.
Filmer, Dean and Lant Pritchett, 1999, “The effect of household wealth on educational
attainment: evidence from 35 countries,” Population and Development Review, 25, 85-120.
Fiscella, Kevin, and Peter Franks, 1997, “Poverty or income inequality as a predictor of
mortality: longitudinal cohort study,” British Medical Journal, 314, 1724-28.
Fiscella, Kevin, and Peter Franks, 2000, “Individual income, income inequality, health and
mortality: what are the relationships?” Health Services Research, 35, 307-18.
Flegg, A. T., 1982, “Inequality of income, illiteracy and medical care as determinants of infant
mortality in developing countries,” Population Studies, 36, 441-58.
Fogel, Rovert W., 1994, “Economic growth, population theory and physiology: the bearing of
long-term processes on the making of economic policy,” American Economic Review, 84,
369-95.
Fogel, Robert W., 1997, “New findings on secular trends in nutrition and mortality: some
implications for population theory,” in Mark Rosenzweig and Oded Stark, eds., Handbook of
Population and Family Economics, Vol 1A, Amsterdam. Elsevier.
Foster, James and Amartya K. Sen, 1997, “On economic inequality after a quarter of a century,”
annex to Sen, Amartya K., On economic inequality, Oxford, Clarendon Press.
Gaarder, Marie M., 2001, “Can population characteristics account for the variation in health
impacts of air pollution? A meta analysis of PM|0-mortality studies,” University College,
London, processed.
Galor, Oded and Omer Moav, 2000, “From physical to human capital accumulation: inequality in
the process of development,” Departments of Economics, Brown University and MIT,
processed.
Gerdtham, Ulf-G. And Magnus Johannesson, 2001, “Absolute income, relative income, income
inequality and mortality,” Stockholm School of Economics, processed. (January).
Gersovitz, Mark, 1983, “Savings and nutrition at low incomes,” Journal of Political Economy,
91,841-55.
Goldin, Caludia and Lawrence F. Katz, 1999, “The returns to skill in the United States across the
20th Century,” Cambridge, MA. NBER Working Paper 7126.
Gottshcalk, Peter and Timothy Smeeding, 2000, “Empirical evidence on income inequality in
industrial countries,” in Anthony B. Atkinson and Francois Bourguignon, eds., Handbook of
Income Distribution, Vol. 1, Amsterdam, North-Holland, pp 261-307.
Gravelle, Hugh, 1998, “How much of the relation between population mortality and unequal
distribution of income is a statistical artefact?” British Medical Journal, 316, 382-85.
Gravelle, Hugh, John Wildman, and Matthew Sutton, 2000, “Income, income inequality and
health: what can we learn from the aggregate data,” Centre for Health Economics, University
of York, York, UK, processed (May.)
Groves, Robert M. and Mick P. Couper, 1998, Nonresponse in household interview surveys. New
64
York, Wiley.
Gwatkin, Davidson R., 2000, “Health inequalities and the health of the poor: what do we know?
What can we do?” Bulletin of the World Health Organization, 78, 3-18.
Haines, Michael R., and Hallie J. Kintner, 2000, “The mortality transition in Germany,
1860-1935,” Historical Methods, 33, 83-104.
Hales, Simon, Philippa Howden-Chapman, Clare Salmond, Alistair Woodward, and Johan
Mackenbach, 1999, “National infant mortality rates in relation to gross national product and
distribution of income,” The Lancet, 354, 2047.
Hammond, Richard J., 1951, Food, History of the Second World War, United Kingdom Civil
Series, London, H. M. Stationery Office.
Hsieh, C. C. and M. D. Pugh, 1993, “Poverty, inequality, and violent crime: a meta-analysis of
recent aggregate data studies,” Criminal Justice Review, 18, 182-202.
Idler, Ellen L., and Yael Benyamini, 1997, “Self-rated health and mortality: a review of twenty
seven community studies,” Journal of Health and Social Behavior, 38, 21-37.
Jain, Shail B., 1975, Size distribution of income: a compilation of data, Washington, DC. The
World Bank.
Jones, Arthur F., and Daniel H. Weinberg, 2000, “The changing shape of the nation’s income
distribution,” P60-204, Washington, DC, Bureau of the Census. Available at
http://www.census.gov/ftp/pub/hhes/www/p60204.html
Judge, Ken, 1995, “Income distribution and life expectancy: a critical appraisal,” British Medical
Journal, 311, 1282-85.
Judge, Ken, Jo-Ann Mulligan and Michaela Benzeval, 1997, “Income inequality and population
health,” Social Science and Medicine, 46, 567-79.
Judge, Ken, Jo-Ann Mulligan and Michaela Benzeval, 1998, “Reply to Richard Wilkinson,”
Social Science and Medicine, 47,983-85.
Kaplan, George A., Elsie R. Pamuk, John W. Lynch, Richard. D. Cohen, and Jennifer L. Balfour,
1996, “Inequality in income and mortality in the United States: analysis of mortality and
potential pathways,” British Medical Journal, 312, 999—1003.
Kaplan, George A., and John W. Lynch, 2001, “Editorial: Is economic policy health policy?”
American Journal ofPublic Health, 91, 351.
Kawachi, Ichiro and Bruce P. Kennedy, 1997, “The relationship of income inequality to
mortality: does the choice of indicator matter?” Social Science and Medicine, 45, 1121-27.
Kawachi, Ichiro, Bruce P. Kennedy, and Deborah Prothrow-Stith, 1997, “Social capital, income
inequality, and mortality,” American Journal ofPublic Health, 87, 1491-98.
Kawachi, Ichiro, Bruce P. Kennedy, and Richard G. Wilkinson, 1999, The society and
population health reader: Volume 1: Income inequality and health. New York, New Press.
Kennedy, Bruce P., Ichiro Kawachi, and Deborah Prothrow-Stith, 1996, “Income distribution and
mortality: cross-sectional ecological study of the Robin Hood index in the United States,”
Britich Medical Journal, 312, 1004-07.
Kennedy, Bruce, P., Ichiro Kawachi, and Deborah Prothrow-Smith, 1996, “Important
correction,” British Medical Journal, 312, 1194.
Kennedy, Bruce P., Ichiro Kawachi, Roberta Glass, and Deborah Prothrow-Stith, 1998, “Income
distribution, socioeconomic status and self-rated health in the United States,” British Medical
65
Journal, 317, 917-21.
LeClere, Felicia B. and Mah-Jabeen Soobader, 2000, ‘The effect of income inequality on the
health of selected US demographic groups,” American Journal ofPublic Health, 90,
1892-97.
Leibenstein, Harvey, 1957, Economic backwardness and economic growth. New York, Wiley.
Lindert, Peter H., 2000, “Three centuries of inequality in Britain and America,” in Anthony B.
Atkinson and Francois Bourguignon, eds., Handbook of Income Distribution, Vol. 1,
Amsterdam, North-Holland, pp 167-216.
Lynch, John W., George A. Kaplan, and Elsie R. Pamuk, 1998, “Income inequality and mortality
in metropolitan areas of the United States,” American Journal ofPublic Health, 88, 1074-80.
Lochner, Kim, Elsie R. Pamuk, Diane Makuc, Bruce P. Kennedy and Ichiro Kawachi, 2001,
“State-level income inequality and individual mortality risk: a prospective multilevel study,”
American Journal ofPublic Health, 91(3), 385-391.
Marmot, Michael and Martin Bobak, 2000, “International comparators and poverty and health in
Europe,” British Medical Journal, 321, 1124—28.
Mclsaac, Sandra J., and Richard G. Wilkinson, 1997, “Income distribution and cause-specific
mortality,” European Journal ofPublic Health, 7, 45-53.
McKeown, Thomas, 1976, The modern rise ofpopulation, New York. Academic Press.
McKeown, Thomas, 1979, The role of medicine: dream, mirage, or nemesis, Princeton, NJ.,
Princeton University Press.
Mellor, Jennifer M., and Jeffrey Milyo, 1999, “Income inequality and individual health: evidence
from the Current Population Survey, Robert Wood Johnson Health Policy Scholars Working
Paper #8, Boston University School of Management, Boston, MA.
Mellor, Jennifer M., and Jeffrey Milyo, 2001, “Re-examining the evidence of an ecological
association between income inequality and health,” Journal of Health Politics, Policy, and
Law, forthcoming.
Miller, Douglas, 2000, “Income inequality and mortality in the US: Aggregate data and micro
relationships,” Department of Economics, Princeton University, Ph.D. thesis, Chapter 1,
processed.
Mirrlees, James, 1975, “A pure theory of underdeveloped economies,” in Lloyd Reynolds, ed.,
Agriculture in Development Theory, New Haven, CT. Yale University Press.
Mujica, Oscar J., Carlos Castillo-Salgado, Jorge Bacallao, Enrique Loyola-Elizondo, Maria
Christina Schneider, Manuel Vidaurre, and George A. O. Alleyne, 2000, “Income inequality
and suicide risk in the Americas,” Pan-American Health Organization, processed.
Murthi, Mamta, Anne-Catherine Guio, and Jean Dreze, 1995, “Mortality, fertility, and gender
bias in India: a district level analysis,” Population and Development Review, 21, 745-82.
Over, Mead, 1998, “The effects of societal variables on urban rates of HIV infection in
developing countries: an exploratory analysis,’’Chapter 2 in Confronting AIDS: evidence from
the developing world. The European Commission. Online at
http://europa.eu.int/comm/development/aids/limelette/index.htm
Pampel, Fred C., and Vijayan K. Pillai, 1986, “Patterns and determinants of infant mortality in
developed nations 1950-75,” Demography, 23, 525-41.
Paukert, F., 1973, “Income distribution at different levels of development: a survey of evidence,”
66
International Labor Review, 108, 97-125.
Pena, Rodolfo, Stig Wall, and Lars-Ake Persson, 2000, “The effect of poverty, social inequity,
and maternal education on infant mortality in Nicaragua, 1988-1993,” American Journal of
Public Health, 90, 64-69.
Preston, Samuel H., 1975, “The changing relation between mortality and level of economic
development,” Population Studies, 29, 231-48.
Preston, Samuel H., 1996, “American longevity: past, present, and future,” Syracuse University,
Maxwell School, Center for Policy Research, Policy Brief 7/1996.
Preston, Samuel H., and Michael R. Haines, 1991, Fatal years: child mortality in late nineteenth
century America, Princeton, N.J. Princeton University Press.
Pritchett, Lant and Lawrence H. Summers, 1996, “Wealthier is healthier,” Journal of Human
Resources, 31, 841-68.
Putnam, Robert D., with Robert Leonard! and Raffaella Nanetti, 1993, Making democracy work:
civic traditions in Modern Italy, Princeton, NJ. Princeton University Press.
Ray, Debraj, 1998, Development Economics, Princeton, NJ, Princeton University Press.
Regidor, Enrique, Pedro Navarro, Vicente Dominguez and Carmen Rodriguez, 1997,
“Inequalities in income and long-term disability in Spain: analysis of recent hypotheses using
cross-sectional study based on individual data,” British Medical Journal, 319, 898-902.
Rodgers, G. B., 1979, “Income and inequality as determinants of mortality: an international
cross-section analysis,” Population Studies, 33, 343-51.
Rogot, E., P. D. Sorlie, N. J. Johnson, and C. Schmitt, eds, 1992, A mortality study of 1.3 million
persons by demographic, social, and economic factors: 1979-1985 Follow-up, Bethesda,
Md. NIH.
Ross, Nancy A., Michael C. Wolfson, James R. Dunn, Jean-Marie Berthelot, George A. Kaplan,
and John W. Lynch, 2000, “Relationship between income inequality and mortality in Canada
and in the United States: cross sectional assessment using census data and vital statistics,”
British Medical Journal, 319, 989-902.
Rosenzweig, Mark, 1988, “Labor markets in low income countries,” in Hollis Chenery and T. N.
Srinivasan, eds., Handbook ofDevelopment Economics, Vol L, Amsterdam, North-Holland,
pp 713-62.
Sachs, Jeffrey D., 2001, “Tropical underdevelopment,” Cambridge, MA. NBER Working Paper
8119, processed. (February.)
Saplosky, Robert M., 1998, Why don I zebras get ulcers? Un updated guide to stress, stressrelated diseases, and coping. New York. Freeman.
Sawyer, Malcolm, 1976, “Income distribution in OECD countries,” OECD Economic Outlook,
Occasional Studies, Paris, OECD.
Schofield, Roger and David Reher, 1991, “The decline of mortality in Europe,” in Roger
Schofield, David Reher, and Alain Bideau, The decline of mortality in Europe, Oxford,
Clarendon Press, pp 1-17.
Schultz, T. Paul, 1988, “Education investments and returns,” in Hollis Chenery and T. N.
Srinivasan, eds.. Handbook of Development Economics, Vol. 1, Amsterdam. North-Holland,
pp. 543-630.
Sen, Amartya K., 1973, On economic inequality, Oxford, Clarendon Press.
67
Sen, Amartya K., 1999, Development as Freedom, New York. Knopf.
Soobader, Mah-Jabeen, and Felicia LeClere, 1999, “Aggregation and measurement of income
inequality: effects on morbidity,” Social Science and Medicine, 48, 733-44.
Steckel, Richard, 1995, “Stature and standard of living,” Journal of Economic Literature, 33,
1903^40.
Stiglitz, Joseph E., 1976, “The efficiency wage hypothesis, surplus labor, and the distribution of
income in L.D.C.s,” Oxford Economic Papers, 28, 185-207.
Strauss, John, and Duncan Thomas, 1998, “Health, nutrition, and economic development,”
Journal ofEconomic Literature, 36, 766-817.
Subramanian, Shankar, and Angus Deaton, 1996, “The demand for food and calories,” Journal of
Political Economy, 104, 133-62.
Swamy, Anand V., 1997, “A simple test of the nutrition-based efficiency wage model,” Journal
of Development Economics, 53, 85-98.
Szreter, Simon, 1988, “The importance of social intervention in Britain’s mortality decline c.
1850-1914: a reinterpretation of the role of public health,” Social History ofMedicine, 1,
1-37.
Victora, Cesar G., J. Patrick Vaughan, Fernando C. Barros, Anamaria C. Silva and Elaine
Tomasi, 2000, “Explaining trends in inequities: evidence from Brazilian child health
studies,” The Lancet, 356, 1093-98.
Wagstaff, Adam and Eddy van Doorslaer, 2000, “Income inequality and health: what does the
literature tell us?” Annual Review of Public Health, 21, 543—67.
Wagstaff, Adam, 2000, “Socioeconomic inequalities in child mortality: comparisons across nine
developing countries,” Bulletin of the World Health Organization, 78, 19-29.
Waldmann, Robert J., 1992, “Income distribution and infant mortality,” Quarterly Journal of
Economics, 107, 1283-1302.
Wennemo, Irene, 1993, “Infant mortality, public policy and inequality—a comparison of 18
industrialized countries, 1950-1985,” Sociology ofHealth and Illness, 15, 429-46.
Wilkinson, Richard G., 1989, “Class mortality differentials, income distribution and trends in
poverty 1921-1981,” Journal of Social Policy, 18, 307-35.
Wilkinson, Richard G., 1992, “Income distribution and life-expectancy, ” British Medical
Journal, 304, 165-68.
Wilkinson, Richard G., 1994, “The epidemiological transition: from material scarcity to social
disadvantage?” Daedalus, 123, 61-77.
Wilkinson, Richard G., 1996, Unhealthy societies: the affliction of inequality, London.
Routledge.
Wilkinson, Richard G., 1997, “Commentary: income inequality summarizes the health burden of
individual relative deprivation,” British Medical Journal, 314, 1727-28.
Wilkinson, Richard G., 1998, “Letter to the Editor,” Social Science and Medicine, 47, 411-12.
Wilkinson, Richard G., 2000, Mind the gap: hierarchies, health, and human evolution, London,
Weidenfeld and Nicolson.
Wittenberg, Martin, 2000, “Predatory equilibria: systematic theft and its effects on output,
inequality and long-run growth,” Johannesburg. University of the Witwatersrand, Department
of Economics, processed. (July).
68
i
Wolfson, Michael C., George A. Kaplan, John W. Lynch, Nancy A. Ross, and Eric Backlund,
1999, “Relationship between income inequality and mortality: empirical demonstration,”
British Medical Journal, 319, 953-55.
Yitzhaki, Shlomo, 1979, “Relative deprivation and the gini coefficient,” Quarterly Journal of
Economics, 93, 321-4.
Appendix: Income, income inequality and mortality in economic development
In Section 2.1.1,1 used a simple quadratic version of the absolute income hypothesis to illustrate
how income and income inequality were likely to have different effects on population health at
different levels of average income. The quadratic model is useful, but not very plausible, if only
because increases in income will eventually reduce health. A more realistic model links
individual health to a latent variable, itself a function of income, and assumes that death takes
place when this latent variable falls below some critical value. Under suitable assumptions, such
a formulation yields explicit predictions for the probability of death for individuals, as well as for
communities, and the resulting equations provide a better way of linking population health to
population income and income inequality.
As before, I write hjs for the health of individual z in population s, and assume that health is
linear in the logarithm of individual income,
h.IS
a + pirry., + £..
(10)
where the random term e.A, is assumed to be normally distributed with mean zero and variance
o.22. The individual dies when hjs falls below the critical level c. The probability that this
happens, or the probability of death, is written pjs which is
Pis
p(death\yjs) = O
69
o
(11)
/
where 0 is the distribution function of the standard normal. Suppose that, within each country s,
the logarithm of income is normally distributed with mean
and variance
We can then
rewrite (10) as
h.IS
« + Ph + P (lny/4.
+ e»-
(12)
so that ps, the fraction of people dying in the state, or the probability of death conditional on the
state mean of log income can be written
Ps
c-q-Ph
p(death | pv) =
(13)
Equation (13) links population mortality rates to the population mean of log income, p^and the
variance of log income, v,5*
If we differentiate, first with respect to jj,y, and then with respect to v6_, we obtain
I
dPs
P<p
c a - Pp s
O\/l + P2 V 7c?
s
(14)
/
and
dPs
c - q-Pns.
<P
oil
01/1
(15)
where (p is the pdf of the standard normal. As mean log income p s, goes to infinity, the pdf goes
to zero, so that the derivative in (14) goes to zero when income is sufficiently large. The same is
true for the derivative with respect to income inequality in (15) because (p(x)x goes to zero asx
goes to infinity. As countries become sufficiently rich, neither population mean income nor
income inequality have any effect on population mortality rates. However, if we take the ratio of
(15) to (16), the pdf functions cancel, and the ratio grows linearly and indefinitely with mean log
70
)r
income. Hence, as claimed, the effect of income inequality relative to the effect of income is
larger among richer countries.
71
-6-1
actuals
o
■a -6.5 -
MALES
i
predicted
I ■
i -7-
0 /
E
■5
f -7.5-
FEMALES
s
T
8
~io
9
logarithm of family income, 1980 $
Tl"
Figure 1: Age adjusted log odds of mortality and the logarithm of family
income, National Longitudinal Mortality Study
.002 J
•35
3
.0015-1
CD
CD
O)
CD
MALES
£
CD
0>
TJ
.001 q
'o
2 .0005-1
FEMALES
oq
20,000
-----40,000
60,000
Family income in 1980 $
Figure 2: Age adjusted probability of death and family income,
National Longitudinal Mortality Study
Figure 3: Health and income in two reference groups: the effects of
within- and between-group inequality on the gradient
100 H
o
o
€cn
c
22 804
■■
<
-
co
o
co
CM
2
60 H
"o
CD
2
so
<5
404
9.5
10.5
10
logarithm of household total expenditure per head
11
Figure 4: Percentages of children aged 7 to 12 who are enrolled in
school, rural India, National Sample Survey, 1995-96
4
70”
Percentage of children aged 0 to 4 with all three
of BCG, DPT, and OPV immunizations
c
£
75
o
roO)
c
a>
£
Q>
Q.
40 H
30 v
11
10
9.5
logarithm of per capita household expenditure
Figure 5: Percentages of children under 5 who are vaccinated, rural
India, National Sample Survey, 1995-96.
n
B
1
C
—
1
v —J
dF(y)
2
1
g-
A
c
g
0
x
y y + dy
income
X
COMPUTING RELATIVE DEPRIVATION FROM THE INCOME DISTRIBUTION
Figure 6: The definition of relative deprivation from the cumulative
distribution of income
Figure 7: Relative deprivation curves for selected US states, Current
Population Survey 1991.
-4.4-J
DC
(U
MS
-4.6 -I
NV
cu
o
E
■e
•a
•
DE
LA
NY
w
TJ
"O
O
O)
O
AL
-4.8 J
vto
NH*
ME
TX
>5
SD
• HI
•5^
0.4
j
I
j
0.6
0.8
1
1?2
Variance of log household income per equivalent
Figure 8: Inequality and the log odds of mortality, vital statistics and
census data, 1990
Health (In)Equity Concerns
India Health Profile
Health Equity in India
& Regional Disparities
Gender Inequities
& Employment Inequities
Ravi Duggal
Class Inequities
Coordinator
Centre for Enquiry into Health and Allied Themes
www,cehat-0rq
Inequities Across Social Groups
Other Inequities
India Health Profile...
India’s Global Share In:
Facts and Figures
& Doctors (all systems): 1.5 mi. (680 per doc)
& Doctors (allopathy): 600,000 (1700 per doc)
& Nurses: 650,000 (1570 per nurse)
Hospital Beds: 1,000,000 (1020 per bed)
Public Expenditure: $5.80 bi. (1% of GDP)
& Private Expenditure: $29.07 bi. (5% of GDP)
& Health Outcomes:
MMR" 408, Life Expectancy-65
%
17
17
23
20
30
68
14
Population
Total deaths
Child deaths
Maternal deaths
TB cases
Leprosy cases
Persons with HIV
Subsidy in Healthcare
Public v/s Private Hospital Care
.311
»35
I 30
_i
\ r ”‘i
Richnst 70%
I 25
= 20
x>
£ 15
o
£ 10
cc 5
0
________________ 12A
_______ T
10'1
rats
00%-80%
<
Middlu 70%
□
Poorest
20%
2nd
3rd
Income Quintiles
4th
Richest
20%
HOV.-WA.
ZEZM8
Poorest 20%
0
500
1500
1OOO
2000
2500
3000
3500
4000
>*vr !<>■>,I‘<>|iuUG<iii
[g
1 r>>.
|
1
Regional Disparities
EXPENDITURE & OUTCOMES
Rural - Urban divide
Developed - Underdeveloped states
«!■/ BIMAROU or EAG states
-
& Issues and Concerns
Access to health services
Health infrastructure and personnel
Public health resources and expenditures
- Healthcare outcomes
Utilisation Patterns [Percentages)
Rural Areas _ Urban Areas
Rural-Urban Disparities - India
Type of Care
& RURAL (per 1000 population)
- Hospital Beds = 0.2
- Doctors = 0.6
- Public Expenditures =
Rs.80,000
- Out of pocket = Rs.750.000
&
URBAN (per 1000 population)
Hospital Beds = 3.0
- Doctors = 3.4
Public Expenditures =
Rs.560,000
- Outof PocketRs. 1.150,000
Hospital care
Outpatient care
Instit'nl births
Child illnesses
RTI treatment
IMR - 44/1000 LB
U5MR = 87/1000 LB
Births Attended - 73.3%
Full lmmunz = 61%
Median ANCs-4.2
IMR = 74/1000 LB
U5MR = 133/1000 LB
Births Attended = 33.5%
Full lmmunz.=37%
Median ANCs=2.5
Private Public Private Public
55/37 45/63 57/40 43/60
81/74 19/26 80/73 20/27
49
51
55
77
45*
33
55
66
80
45
34*
20
15
28
72
85
17
40
60
Contraception
83
42’
58 i 55
45
ANC
_____________________________
Child Immunz.
Note For hospital and OPD care I*1 figure is for year 1996 and 2nd 1986
Includes Home remedies
’aslitra 2000-01 Health Expenditures (Rs. Million)
Type of Expd. Rural
Urban
Total
Medical care
259.55
7457.24
7716.79
Public Health
4514.34
1947.33
6461.67
Fam. Planning 677.57
61.70
739.27
58.68
167.77
195.59
Other FW
136.91
672.34
840.11
Capital
84.41
305.04
389.45
lai Disparity in Access to MCH Services
80-1
t
60 Y !
□ No Antenatal Care
MCH
■ No TT injection
□ No Iron Folk acid
a Non Insl, Delhsry
■ Births not attended by health
professional
□ Current Contraceptive User
Rural
2
tegional Disparity in Child Mortality
Regional Disparity in Access to Child Health
Services
;!| □ N" miiniiii>>.-■ I ii-ii I
j
0IMR per 1000 live
births
I
It
Iiiiiiimiiisiii ■■■ii
children
B I J-5 MR per 1000
live births
J Q Vila in in ''
NNMRper 1(1(10
Rural
live births
Rural
Rcgii
Regional Disparity in Medical Care
60
Asthma point
prevalence/100,000
■ Tuberculosis point
prevalence/100,000
Jaundice in last 12
months/100,000
Malaria in last 3
months/100,000
nal Differences in Selected Child
Morbidity
RegH
ifferences in Nutrition status of
women and childrenl_.
below 18.5 kg.' m2
B “iiwomen wiihanacmia
"i>c hildren undcrwcighl
•2SD
“»c hildren si unled-2SD
■ “»c hildren wasl cd • 2S D
Rural
3
Gender Inequities
Strong patriarchal society
'& Son preference - sex selection, infanticide
& Sex ratio disparities
& Access to services
& Health outcomes
& Nutritional status
■',i m 85i.i
851 to 950
■151 io 1100
1‘Mta not a /arable
Inequities in Medical Care - Rural
Gender Inequities in Medical Care - Urban
60 -/ I
40
4
hi
30
20
10
0
□ Any Ailment per l<100
r
M ale
0 Any Ailment per
1000 -15 days
■ Hospitalisation rate
-15 days
S Hospitalisation rate
per HKIO/year
□ Untreated Ailments
I
per 1000/ycar
□ UntrcatcdAilments
%
■/
%
□ Borrowings/Asset
Fein ale
nder Inequity in Child Mortality
□ Borrowings/Asset
sales for Hosp %
F e male
Gendt
sales for Hosp. %
Terences in selected Morbidity
O J junJite in h«t 12
month* / 1(10,000
O Mui ar i j in hxt 3
Female
munih' loo.ooo
Female
4
35-
fferences in Nutritional Status of
Children
Gender Differences in selected child
-1
Morbidity
20-
10-
Q '.childrc*
r±
under w eight -2SD
r e innle
childrcn»linlcd-
0
2SD
■ hrrn
O ’• children wjjled •
\
ISO
O ’’childrenunjcmic
Employment Inequities
Class and Social Group Inequities
& Organised sector -15% of population
- Has social security
- Has capacity to buy healthcare form market
- Gets free healthcare
Social Groups - Caste, OBC, Dalits, Adivasis
& Unorganised sector - 85% population
- No social security
- Subsistence or below poverty line existence
- Has to buy healthcare from the market
Issues and concerns
- Sharp inequities in access, healthcare utilisation
patterns and health outcomes
Nutritional status and morbidity differences
Class^
9,,<
80-^
Economic class and purchasing power
Consumption expenditure class
Standard of living index
^-GtaSsand Social Group Inequities in
Access to child health services
Social Group Inequities in Access
to MCH services
7,'i " ‘
2
—
60
u
£ 40-''
; 3»-
V 30--'
4 20-
y 50
L
1
Low
Medium
High
I0-
um
Medium
High
Dalits Adivasis
S t u n d a rd o f Li v in g
I □ Noiinmunisulionorchildren
OBC
Other
S <> c i a I Gru u p«
S Fulllmmunisationofchildrcn
Miamin Asupplcmcnt forchildren
5
lass and Social Group Inequities in Medical
Class ©KTSocial Group Inequities in Access
to medical Care- Urban
II
a AN V Atm HI per |m
Social (> ro u ps
C o ii s ii in p I i o n Class
I'd«y>
■ l(<n|>tat»tm m • |<i in "ii yea,
□ Any Ailment per IUOO • I * d<y ■
□ tliui
0 I i real ed Absent ■ *.
('» n mi 111 p t i o n Class
AiIiiicih-
Class and Social Group Inequities in Child
Social Groups
O n.Hr''■op1 A«iH
Hrnp *•
and Social Group Differences in
Mortality
140
IL Ezzi
100
SO-'
60
40
20-'
s^and Social Group Differences in
Nutrition Satus of Women
s and Social Group Differences in
Nutritional Status of Children
I
I
I
I
I.
Low
Standard o f Living
Social Gro tif
o ".women u 11:: li' II .
□
women with anjensi i
6
nd Social Groups Differences in
Reproductive morbidity
Other Inequities
40
35
&Age
30
I
25
20
Disability
□ AnyRTI
I
(%evci
15
inarricu
I
wolnvli I
Bl Any Rq'i
He uh li
J
Law
Medium
High
Problem i
IXilits
s tandard o f Living
G ro u ps
Adivnsis
OIIC
Ollier
Sot in I
Displaced persons
Migrants
& Habitat -slums
•J/ Conflict and communal situations
eurrci !•.
0 inurrk
Heinen ■
Conclusions
& Complexity of range of inequities in health in India
- Criticality of the rural-urban disparities
- Persistence of sharp class and social group differentials
- Patriarchal nuances
- Missing links of public health
Socio-economic level of development
& Financing issues - Tax : GDP ratios
7
Session on Mental Health and Human Rights, 9th July 2005
Objectives for the day:
Participants will be:
1. informed about the human rights problems and concerns in the mental health area
2. able to process and negotiate values and rights of diverse actors in the mental health
sector
3. consolidate a human rights perspective about persons with a psychosocial disability
Session I: Human Rights in Mental Health: An overview (1 /z Hours)
/z Hour: Ice breaking and brain storming on Mental Health and Mental Illness- What it is,
causes, etc. from personal sharing. Setting the framework and objectives for the day’s
discussion.
1 Hour: Lecture presentation covering: Need for Law concerning the mentally ill; Stigma
and HR; HR interventions in India; HR covered in Mental Health Act; International
Principles and Covenants; Case Law in India and other HR benchmarks.
Session II & Session III: Case Studies (3 Hours)
Case studies will be provided to participants covering:
• human rights issues in institutional context
• human rights issues in non-institutional treatment context
• human rights issues in traditional healing setting
• human rights issues in community
All case studies pertaining to the persons with a psychosocial disability
Discussion will pull together aspects of law in the Indian context, involuntary
commitment, forced treatment, and need for legal reform.
Session IV: The UN Convention on Rights of Persons with Disabilities (1 Vi Hours)
This session will consolidate learnings on Human Rights of Persons with psychosocial
disabilities referring to the UN Convention by readings of certain sections of the Draft
Convention.
Facilitated by:
Bhargavi Davar
Center for Advocacy in Mental Health
B 1/11, 1/12 KONARK POORAM
Kondhwa Khurd
Pune 411 048
Tel: 020-26837644; 26837647
Email: wamhc@vsnl.net; info@camhindia.org
Website: www.camhindia.org
1
I
I
•1
I
I
FBs8
V,
I Lecture covers
1. Definition
TORTURE
' 2. Aims
•
4. Doctors and torture
Amar Jesani
Centre for Studies in Ethics and Rights
Indian Journal of Medical Ethics
Mumbai
Torture: Definition (UN)
"Torture means any act by which severe pain or
suffering, whether physical or mental, is
intentionally inflicted on a person for such
purposes as obtaining from him or a third person
information or a confession, punishing him (her) for
an act he (she) or a third person has committed or
is suspected of having committed, or intimidating or
coercing him (her) or a third person, or for any
reason based on discrimination of any kind,...”
.3. Methods
;
5. Why doctors participate?
6. Is torture prevalent in India?
s
I
“
when such pain or suffering is inflicted
by or at the instigation of or with the consent
or acquiescence of a public official or other
person acting in an official capacity. It does
not include pain or suffering arising only
from, inherent or incidental to lawful
sanctions."
■M
■...............................
s
s
Torture: Aims
|r (I) To obtain information
| (2) To force confession
0 (3) To get a testimony incriminating others
Methods of Torture
I
(a) Physical Torture
s (b) Mental Torture
(c) Sexual Torture
■tJ (4) To take revenge
fI
I
(5) To spread terror in the Community
(6) To destroy the personality
I
hg
1
r
©
'
I :
a (a) Physical Torture
-I
sI
1. Physical torture that causes extreme and
excruciating pain:
beating, falanga, finger torture, suspension, cold torture,
heat torture, irritant torture, made to walk on sharp
objects, dental torture, ear torture, hair torture,
scratching with knife, forced position, twisting of body
parts, etc.
2. Physical torture causing fear of immediate death:
electrical torture, suffocation, sham execution.
ilfj
■■
1
(a) Physical Torture
1
.. 3. Physical torture causing extreme
exhaustion:
physical exhaustion, forced labour.
Physical torture causing disfiguration,
mutilation and permanent disability:
telefono, mutilation, disfiguration (acid).
;<<•
* V
■
I
•J
I
I1
Bl
I
A (b) Mental Torture
f'
ki
H
Deprivation techniques:
sensory, perceptual, social, basic needs.
Coercion techniques:
Impossible choices, incongruent actions.
Threats and humiliations:
Communication techniques:
misinformation, conditioning,
Pharmacological techniques:
“Truth Serum" - Sodium Pentothal
I
■
I' '■'^5
2
(c) Sexual Torture
Using instruments
Without use of instruments
z Using animals
i
Forced Feeding
ii
i
'r
s
| Doctor's participation in torture
r i Evaluating victim's capacity to withstand torture
J - Supervising torture through the provision of
r^.
medical treatment if complications occur
JI
Providing professional knowledge and skill to the
P
torturer
Ig £ Falsifying or deliberately omitting medical
information when issuing health certificates or
autopsy reports
1
fl. Doctor’s participation in torture
3
|P ^Providing medical assistance within the
torture system without either denouncing
F.
torture or resigning from such work
^Administering
torture
by
directly
p
participating in it
k
Remaining silent in spite of the knowledge
that abuse have taken place
(Chilean Medical Association)
3
!
i'< vJt;
/ <'
X.
Why do doctors get involved in torture?
•J
Those who get involved in torture NOT
"exceptionally or uniquely evil persons”, they
are often "your neighbour’s son"
Discrimination against and labeling or devaluing
of a victim group - "negative stereotypes", "out
group" - lacking basic human characteristics
Training, "learning by doing"
Obedience to repressive authority
q
Why do doctors get involved in torture?
®
■J
Existence of tendency to blame victims (a "justworld view" - in a just world, people get what
they deserve)
Identification with the cause of torturers
- Existence of ideology tolerant of torture
Fear of Consequences of refusing to cooperate
System of rewards for those engaged in torture
&
f.
|
Ki
LA
U Deaths in Police Custody
1.1985-November 1,1991 (6 yrs IQmths)
I AlJanuary
“India: Torture, rape & deaths in police custody 1992"
Why do doctors get involved in torture?
■’
a
i
"Bureaucratisation" of medical role - "Only
professional skills", NO "Moral frame-work" - "I
confine to my function, that is not my work, I am
a doctor"
% Cultural and societal processes which embody
tolerance to torture
Inadequate understanding of medical ethics
"Doubling'* (Litton, "Nazi Doctors"): Adaptation to extremity, "Division of
self into two functioning wholes, so that the part-self acts as an entire
self”
■A
Study
Al, 1992
1980-84
Not Covered
1985-89
1
70
W. Bengal, Orissa, Manipur. Assam
9°
Bihar, Haryana, Punjab, Sikkim, UP
170
Delhi, MP, Rajasthan
______________
63
__________________
21
I
Maharashtra
Total (Average 59 deaths/year)
j
102
1990-91
08
Not Covered
Total
21
.. 155
*MSW Dissertation (1992),
Karve Institute of Social Service, Pune
414
L
I
Deaths in Police Custody, 1980-89
Study (1992)*, Maharashtra State
Years
A '. ........................ —_——.————,—------ —
AP, Karnataka, TN, Pond., Kerala
.
■
H
Deaths in Police Custody, 1980-89
Study (1992), Maharashtra State
• > Of the 155 police custody deaths reported, in
only 15 some proceedings were done.
■'-> Of these 15 cases, the status in 1992 was:
• In one, police found guilty of torture
In four, police found not-guilty of torture
' In three, court proceedings were on
' In seven, government/department inquiry on
4
slisS
■
■
i
Deaths in Police Custody, 1980-89
Study (1992). Maharashtra State
_______
Causes
Number
56 (36.1)
J Natural causes (21), diseases • skin, kidney, abdomen,
r
burns, jaundice, giddiness, fits, TB, heart attack, alcohol
consumption <9>, neck wound, snakebite, car accident,
i bullet injury, wound, unconsciousness ....................
■
:
Due to police action
;
'
t
Hanging (45), self-immolation (3), Consumed poison (3).
self-inflicted injury, “Jumped"- tn well, under
autorickshaw/train/bus, from bed/window, on others___
Attributed to acts of public - beaten (9). strangulated
Cause not known/ nor record
\
15(09.7)
65 (42.0)
I___ 10 (06.4)
I
TOTAL
155(100)
5
TORTURE
Dr Raghunatham Opeh
People are being brutalised physically and wrecked mentally by various organs
of the state, society and family, to achieve one purpose or another. The
instrument used to bend and break a person is torture-physical mental and
emotional.
Definition
The definition of torture today according to the "UN Convention Against
Torture....or Punishment" of Dec 10 1984, in force from June 1987 is defined as:
Any act by which severe pain or suffering, whether physical mental, is
intentionally inflicted on a person for such purposes as obtaining from
him/her or third person, information or a confession, punishing him
for an act he or a third person has committed or is suspected of
having committed, or intimidating or coercing him or a third person,
or for any reason based on discrimination of any kind, when such
pain or suffering is inflicted by or at the instigation of or with the
consent or acquiescence of a public official or other person acting in
an official capacity. It does not include pain or suffering arising only
from, inherent in or incidental to lawful sanctions.”
Epidemiology
The UN convention against torture has been signed by very few of the world's
over 180 member states. According to Amnesty International, countries, some of
which are signatories to this convention. In India, violence, brutality and torture in
the form of blinding,, brandings, custodial deaths and other forms take place all
over the country.
Purpose of Torture
The main purpose of torture was to get information, confessions, to punish or to
terrorise. The goal is achieved when the victim can be described as a 'living
1
dead' - to destroy the mind without killing the body. Effect of Torture: Conditions
are created which effectively break down the victim's personality and identity and
his ability to live a fill life later with and amongst other human beings. The worst
is the mental sequalae - deep feelings of guilt and shame, for having given
information and thus harming friends, quite often to prevent family members from
being tortured.
Causes of Torture
1. PoliticalTorture by govt agencies applied to anyone identified as 'enemy' of the
o
state.
Torture and blackmail with threat of torture, to force compliance.
o
Torture and killing of people to force them to leave their lands.
o
2. War-time o
Against POWs.
Against spies and traitors.
o
Against civilians suspected of harboring enemies.
o
Against civilians of occupied territories to subjugate them.
o
3. PoliceTo get confession and or information.
o
To quieten rowdy/disobedient/mentally disturbed prisoners.
o
Of prisoners who try to escape.
o
Of suspects at Police stations.
o
Torture of kin or friends who try to help prisoners or those who try to
o
question the legality of arrest.
4. Organised Crime o To establish obedience through fear.
o To extort money.
o To avenge rival groups.
o To silence law enforcement agents.
o Torture of kidnapped persons to hasten ransom.
o Torture of witnesses to prevent reporting of crime.
5. Miscellaneous o Torture over religious matters.
o Over land/wealth disputes.
o As a reaction to injury/insult to self/family/community.
o As revenge for any reason.
o Family violence.
Torture as a Process
2
The torture process starts with arrest, usually at night, with a formidable display
of power and un-necessary use of violence. The "softening phase", which
follows, usually consists of a couple of days and nights of unsystematic violence
and beating, kicking and other humiliations. Later on, the real systematic torture
starts, when the torturers explore the weak spots of the victim to make him or her
break down. The final result may be a human being with little self-respect- a
broken down personality - with a medical certificate, generally false, denying any
form of maltreatment. The final result may also be death - the torturers disposing
of the corpse by hiding it in a mass grave.
Torture is a wound in the soul, so painful that sometimes you can almost
touch it, but it is also so intangible that there is no way to heal it. Torture
is anguish squeezing in your chest, cold as ice and heavy as a stone,
paralyzing as sleep and dark as the abyss. Torture is despair andfear and
rage and hate, it is a desire to destroy and kill, including yourself.
METHODS OF TORTURE
The methods may be classified into:
•
•
•
Physical
Psychological
Sexual
Physical Torture
The methods of physical torture are those which inflict pain, discomfort and
dysfunction in different parts of the body. Killing the victim is not the aim or
torture. The torturer also takes care that the torture inflicted upon the victim
remains undetected by an ordinary examination. Therefore, torturers are trained
to to torture in such a way that these two precautions are well taken care of.
However, despite all precautions, physical torture always leaves a trail that
eventually leads to its discovery. It is something like a crime. It always leaves
behind some clues that ultimately lead investigators to its discovery and to the
3
criminals. Similarly, medical science has now developed to such an extent that
the internal damage caused by such types of physical torture can be detected
even after several years of its incidence.
Types of physical Torture:
Physical torture can be inflicted in many ways, but the commonly reported ones
can be classified under four categories.
1. physical torture that causes extreme pain.
Beating and severe beating
o
It is the most common type of physical torture. Beating becomes sever
when it is carried out with sticks, cables, whips, iron rods chains, belts etc.
Likewise, punching, kicking etc are considered severe beating. So, simple
beating means slapping on less sensitive and delicate parts of the body
which does not cause significant external and internal damage in the body.
Boxing the ears cannot be considered a simple beating as it can damage
the ear drum and permanently affect hearing. The impact of severe beating
depends not only on the instrument used but also on the body parts
affected. So, simply saying 'severe beating' is not enough. It is essential to
describe the instrument used, body parts assaulted, and also the duration
and severity of beating.
Falanga
o
Falanga is the name for severe beating on the soles of feet. It does not fall
under the heading of severe beating, it has been described separately since
it is one of the most common types of systematic torture used in many
countries around the world. Besides, this type of torture has immediate
and long term consequences, sometimes making a person disabled for
several years. The pain continuously reminds the victim of the torture and
thus handicaps his/her recovery. Falanga can be given in many ways. The
victim may be suspended upside down, the victim's legs may be fixed on
to a table etc. Any technique that immobilises the legs and feet can be
used for applying Falanga.
Finger Torture
o
Pencil or a similar object is put in between two fingers which are then
pressed hard together against the objects. Similarly, fingers may be twisted
and pins pricked into fingers. Nails may be pulled out.
Suspension
o
The victim is suspended by his legs or arms or by his/her hair. It is usually
combined with other forms like Falanga, electric shock, heat, cold etc.
One of the special types of suspension is called 'Palestinian suspension'.
Cold Torture
o
The victim is subjected to varying degrees of cold in different ways.
He/she may be forced to sleep on a damp floor, may be forced to stay
4
naked in extremely cold weather. Some of the Bhutanese refugees have
alleged now they were forced to undress and roll on snow for long periods.
o
Heat Torture
The victim may be forced to stand for hours in the sun in temperature of
more than 30 degrees Celsius. Cigarette bum is the most common type
reported.
o
Irritant Torture
Irritants like chilli powder, table salts, etc are applied on delicate parts or
open wound. Chillies may be burnt and victim forced to inhale the smoke
in a closed room for several hours.
o
Sharp Objects Torture
Victim is forced to walk barefoot over thorny surface or over broken glass
covered floor. He/she may be forced to sit on an object with pointed and
sharp edges.
o
Dental Torture
Healthy teeth may be broken or pulled out. May be asked to chew stone,
wood, metal pieces etc.
o
Ear Torture
This is common the world over. Ears may be twisted or pulled to tear the
ear. Telephono, both ears hit with open palms simultaneously, to impair
hearing is also used.
o
Hair Torture
The victim is dragged by the hair, hair cut short, head shaven or hair
pulled out forcibly.
o
Scratching with knife
Scratches/cuts are made on different parts of the body with a shark object.
o
Tied down
One is tied down in many ways and kept in this position from several
hours to days. Kicks blows etc may be added to this torture.
o
Lower legs torture
Legs are tied together, a folded blanket is placed over the shins and a
round wooden log/iron rod is rolled up and down with a load or two
persons. Causes excruciating pain but leaves no signs.
o
Twisting body parts
The upper arms, lower limbs, neck etc are twisted to such an extent that
the ligaments in the joints are torn off causing severe pain even after the
twisting is over.
o
Poking
Poking the victim with a baton, rod or any similar object is quite common.
Any part of the body may be poked.
2. physical torture that causes fear of immediate death.
o Electric Torture
It is inflicted on the sensitive parts like nipples, genitals, etc. It is applied
inside the mouth which is quite painful and difficult to detect later. Some
times the victim is tied to a metallic bed and electricity applied affecting
the whole body.
5
Suffocation
The victim is suffocated till he/she feels as if on verge of death. The
torturer usually stops short of causing death. Sometimes the victim may
die during the process. By closing the nose and mouth with some object or
even with bare hands once the victim is tied down to the bed. When the
head is immersed in water, urine, vomit, blood etc it is called 'wet
submarino'. In 'dry submarino', the victim's face/head are tightly covered
with a plastic bag or some such article. Suffocation may also be caused by
strangulation with hands, rope, belt etc...
Sham Exectution
o
A well known form of torture. Here, the victim is blindfolded and made to
stand next to a wall. The victim is told that he/she is going to be run over
by a motor vehicle. The vehicle is driven at full speed towards the victim
and comes to a sudden halt close to the victim.
3. physical torture that causes extreme exhaustion.
Physical Exhaustion
o
Here the victim is forced to stand or do gymnastics for a prolonged period
of time. The victim is asked to stand on one or both legs. She/he may be
asked to stand on his/her head with legs in the air or supported by the wall
for a prolonged period.
Forced Labour
o
This is a very common type of physical torture reported by torture victims
in Bhutan and also in Nepal. The victim is made to work very hard
without food and water and wages. They clear forests, break stones, dig
ditches etc.
4. physical torture that causes disfiguring, mutilation and permanent disability.
Telephono
o
Simultaneous beating of both ears with the palms of hands to cause
rupture of ear drum, causing pain, bleeding and hearing loss. It is difficult
for a doctor to detect it.
Mutilation
o
Chopping off ears, nose, fingers etc..
Disfiguration
o
Acid or any other corrosive thrown on the face or other parts
o
Psychological Torture
Psychological methods of torture can be classified into the following categories:
•
•
•
•
•
Deprivation
Coercion or Compulsion
Threats and Humiliation
Communication Techniques
Pharmacological
6
1. Deprivation,
’Sensory Deprivation’ of light and sound disorients a victim with
o
reference to time and place.
’Perceptual Deprivation’ produces disorientation and confusion. It is
o
created by frequent transfers while blindfolded, very frequent disturbances
of sleep. Denial of letters, media and calendars.
’Social Deprivation’ includes not being allowed to see visitors, perform
o
religious rituals & solitary confinement.
’Deprivation of Basic Needs’ includes denial of food, water, toilet
o
facilities, sleep, medical facilities, clothes, space, immobilisation, restraint
and total darkness.
2. Coercion or compulsion.
In this, the victim is compelled or coerced to perform activities or to witness
actions that torture him mentally.
o Impossible Choices
The victim is compelled to choose between two alternatives that are
equally bad and cause mental torture.
Incongruent Actions
o
Signing of false statements. Disclosure of information. Forced to commit
blasphemous acts. Forced to violate social taboos. Forced to witness
torture of other victims. Forced to torture other victims.
3. Th reals and Humiliation.
Threats directed at the victim like threat of execution. Threats directed towards
the family members, kin, friends... Humiliating remarks and actions.
4. Communication Techniques.
o
Misinformation
The victim is given wrong information and tortured mentally.
Conditioning
o
The victim is further tortured whenever he does not comply with the
torturer's demands. Similarly, he is given better food, facilities etc
whenever he yields to demands.
o
Double Blinding
Brutal torture is followed by seemingly humane & sympathetic treatment
from another person.
o
Reverse Effect Technique
Torture is continued in spite of submission to every demand.
5. Pharmacological Techniques.
Various drugs are used to torture the victim, to facilitate torture, to mask the
effects of torture and also as a means of torture.
o
use of drugs to induce self disclosure, eg. alcohol.
o
use of curare/etc to a point of asphyxiation.
o
pain inducing drugs.
o
hallucinogens.
o
psycho-pharmacological drugs, etc.
Sexual Torture
7
Sexual torture has great social and psychological impact. It can be divided into
the following three categories. The most common and serious form of torture for
female is Rape.
1. Sexual torture using instruments:
-penetration of vagina or anus by batons, rods bottles etc.
o
-suspension of weights on penis and scrotum.
o
-electrical shock to sexual organs.
o
-mutilation of breast, genital organ, etc.
o
2. Sexual Torture without using instruments:
-verbal sexual abuse and humiliation.
o
-being undressed in front of others.
o
-rape by a person of the opposite sex, multiple rape and rape in the
o
presence of kin.
-rape by person of same sex.
o
-squeezing of breasts.
o
-forced masturbation.
o
-forced witness of sexual torture of others.
o
-forced to perform sexual torture on others.
o
-forced pregnancy.
o
-being photographed in humiliating positions and situations.
o
3. Sexual torture using animal:
o -rape by trained dogs, monkeys etc.
o -rats, mice, spiders, lizards introduced into vagina/anus.
Torture which leaves no visible signs
•
•
•
developing fear of the unknown
creating uncertainty and constant unpredictability.
causing disorientation and mental agony.
Conclusion
The afore mentioned details are not comprehensive as the torturers keep
devising newer methods using their mental aptitude and imagination. However,
the aim of all forms of torture is the same. The torturer may use one or all of the
methods of torture mentioned simultaneously to break the resistance of the
subject quickly as well as to satiate his own animal urges
8
TORTURE SESSION - AMAR JESANI
SESSION PLAN:
JULY 12, 2005.
Session: I: 2 pm to 3.30 pm.
1400-1430: 30 minutes: Lecture - "Introduction: Torture and its consequences"
1430-1440: 10 minutes: Clarification questions, and breaking into three groups
1440-1530: 50 minutes: 3 Case studies in group discussions and preparation of
reports
1530-1600: 30 minutes (Tea-break and group report preparation continued)
1600-1700: Group reports and discussion (report for 10 minutes, discussion 10
minutes on each report)
1700-1715: Warp up - short presentation
1715-1815: Film on torture and discussion (this is on assumption that the films on
torure at the CEHAT are still useable, if these films are not available, then I will
increase the time for my presentations.
Issues in Medical Ethics
Page 1 of2
Indian Journal of Medical Ethics
Home
Current Issue
Pastissues
Su|jp.ort
About MME
DISCUSSION
Response of mental health professionals in Gujarat
Harish Shetty
We visited Anand-Kheda(between Baroda and Ahmedabad). Sixty people died in the riots here,
and thousands were rendered homeless in approximately 125 villages. Victims have received
scant support from the government and voluntary organisations. Reports of insomnia, startle
reactions, fearfulness, intrusive memories and sadness are common. The commonest coping
method has been prayer. The women and the elderly appear depressed; young males appear
very angry. Children are also affected. Spiritual leaders have been largely silent. Victims fear
returning to their homes, and the state government's order to close relief camps on the eve of
the monsoons has aggravated existing stress.
The victims of communal violence in Gujarat have experienced extraordinary physical and
psychological trauma. These pains are compounded by government inaction against the
perpetrators of violence, and by the lack of support from outside their community.
The experience of a disaster can trigger off intense emotional reactions, sometimes highly
individual in nature. Sometimes the reaction is delayed. Sometimes it can lead to considerable
disability. Most people may not need professional help, but support from family and friends is
critical. Psychological symptoms can persist decades after the traumatic event.
Mental health response to disaster in India
Overall, the mental health response following large-scale trauma in India has been from a
psychiatric perspective, ignoring the multiple influences on people's experiences of, and
responses to, a disaster. What is needed is a psychosocial perspective viewing victims in their
cultural milieu. Further, there has been insufficient documentation of work, very little sharing
within the mental health community, and a preoccupation with research without an 'action'
component.
There were isolated psychosocial interventions following the 1992-93 riots in Bombay. Such
relief was provided for the first time through lay mental health workers, after the Kandla
cyclone. Relief after the Orissa cyclone and the Bhuj earthquake also had a mental health
component. Voluntary organisations played an important part.
There are many barriers to an adequate psychosocial response to large-scale trauma. The
country has only 7,000 mental professionals spread unequally across different states. The field
focuses on treating individual patients rather than communities. Mental health professionals are
not experienced with negotiating with NGOs, or working with the development model necessary
in the context of a disaster. Social work institutes may jump on to the bandwagon to make their
presence felt but abandon the field without setting a long-term process in motion. Some NGOs'
efforts are dictated by the availability of financial grants, preventing a long-term intervention.
Finally, there is a lack of awareness about mental health issues across all target groups.
People working in mental health will also have to reflect on the conflicts between religious
identity and professional responsibility. This is a requisite for post-riot counseling, also
important during times of peace.
Further, some activists have biases obstructing equitable counseling to all affected people For
example, after the Bhuj earthquake, some activists stopped me from going to homes of a
particular community stating that they were rich and required no help. Grief is universal; mental
health intervention must reach all hearts.
Mental health professionals in Gujarat
In Gujarat, mental health professionals from Ahmedabad have been working with violence
file://C:\Documents%20and%20Settings\Administrator\Desktop\2005\Health%20and%20H ... 5/8/2007
Issues in Medical Ethics
Page 2 of 2
victims in government hospitals and in relief camps. However, the response has been feeble
when compared to the Bhuj earthquake. The most important reason for this is the fear and
uncertainty of the situation. Second, the sanction of the violence by a large majority has
affected mental health professionals as well. They are also restricted by the absence of
institutional mechanisms to provide systematic relief. Gujarat needs a long-term plan for
community-based psychosocial intervention using volunteers. The key actors should be
disaster workers, primary care workers, teachers, medical professionals and social activists.
Training must include some counseling skills, briefing on social issues, team building, and
breaking stereotypes. This must be a government-mandated effort with inputs from specialists
close to the affected region.
Every riot victim has a right to mental health relief for a reasonable period, provided by the
State. Research or assessments must include an action component; victims should not be
viewed as a laboratory. Counsellors must plan to work on long-term conflict resolution. They
must move to the centre of the disaster zone, not hide in the periphery.
In Gujarat, the mental health fraternity was silent fearing the disruption of 'therapeutic
neutrality'. This is actually a denial of professional responsibility. Mental health professionals
need not be sloganeers, but they must raise sane voices during difficult times. A small minority
has made active efforts and taken stands, but on the whole, silence has transformed the
profession's empathy into apathy. This collective silence must be broken with concerted action
toward healing and prevention.
(Dr H Shetty, Dr S Shah, Dr H Sharma, and Ms P Pushkarna visited the camps. The team was
supported by Mr S Desai and Ms A Patel from the Charutar Arogya Mandal. Medical personnel
and social workers of the preventive and social medicine department of Pramukhswami Medical
College provided critical inputs.)
Dr Harish Shetty is a psychiatrist and visiting faculty at the Tata Institute of Social Sciences.
4A/11, Taxilla, Mahakali, Andheri (E), Mumbai 400093. Email: kannaad@hotmail.com
file://C:\Documents%20and%20Settmgs\Administrator\Desktop\2005\Health%20and%20H ... 5/8/2007
Response of mental health professionals in Gujarat
Harish Shetty
We visited Anand-Kheda(between Baroda and Ahmedabad). Sixty people died in the
riots here, and thousands were rendered homeless in approximately 125 villages. Victims
have received scant support from the government and voluntary organisations. Reports of
insomnia, startle reactions, fearfulness, intrusive memories and sadness are common. The
commonest coping method has been prayer. The women and the elderly appear
depressed; young males appear very angry. Children are also affected. Spiritual leaders
have been largely silent. Victims fear returning to their homes, and the state
government's order to close relief camps on the eve of the monsoons has aggravated
existing stress.
The victims of communal violence in Gujarat have experienced extraordinary physical
and psychological trauma. These pains are compounded by government inaction against
the perpetrators of violence, and by the lack of support from outside their community.
The experience of a disaster can trigger off intense emotional reactions, sometimes highly
individual in nature. Sometimes the reaction is delayed. Sometimes it can lead to
considerable disability. Most people may not need professional help, but support from
family and friends is critical. Psychological symptoms can persist decades after the
traumatic event.
Mental health response to disaster in India
Overall, the mental health response following large-scale trauma in India has been from a
psychiatric perspective, ignoring the multiple influences on people's experiences of, and
responses to, a disaster. What is needed is a psychosocial perspective viewing victims in
their cultural milieu. Further, there has been insufficient documentation of work, very
little sharing within the mental health community, and a preoccupation with research
without an 'action' component.
There were isolated psychosocial interventions following the 1992-93 riots in Bombay.
Such relief was provided for the first time through lay mental health workers, after the
Kandla cyclone. Relief after the Orissa cyclone and the Bhuj earthquake also had a
mental health component. Voluntary organisations played an important part.
There are many barriers to an adequate psychosocial response to large-scale trauma. The
country has only 7,000 mental professionals spread unequally across different states. The
field focuses on treating individual patients rather than communities. Mental health
professionals are not experienced with negotiating with NGOs, or working with the
development model necessary in the context of a disaster. Social work institutes may
jump on to the bandwagon to make their presence felt but abandon the field without
setting a long-term process in motion. Some NGOs' efforts are dictated by the availability
of financial grants, preventing a long-term intervention. Finally, there is a lack of
awareness about mental health issues across all target groups. People working in mental
1
health will also have to reflect on the conflicts between religious identity and professional
responsibility. This is a requisite for post-riot counseling, also important during times of
peace.
Further, some activists have biases obstructing equitable counseling to all affected
people. For example, after the Bhuj earthquake, some activists stopped me from going to
homes of a particular community stating that they were rich and required no help. Grief is
universal; mental health intervention must reach all hearts.
Mental health professionals in Gujarat
In Gujarat, mental health professionals from Ahmedabad have been working with
violence victims in government hospitals and in relief camps. However, the response has
been feeble when compared to the Bhuj earthquake. The most important reason for this is
the fear and uncertainty of the situation. Second, the sanction of the violence by a large
majority has affected mental health professionals as well. They are also restricted by the
absence of institutional mechanisms to provide systematic relief. Gujarat needs a long
term plan for community-based psychosocial intervention using volunteers. The key
actors should be disaster workers, primary care workers, teachers, medical professionals
and social activists. Training must include some counseling skills, briefing on social
issues, team building, and breaking stereotypes. This must be a government-mandated
effort with inputs from specialists close to the affected region.
Every riot victim has a right to mental health relief for a reasonable period, provided by
the State. Research or assessments must include an action component; victims should not
be viewed as a laboratory. Counsellors must plan to work on long-term conflict
resolution. They must move to the centre of the disaster zone, not hide in the periphery.
In Gujarat, the mental health fraternity was silent fearing the disruption of 'therapeutic
neutrality'. This is actually a denial of professional responsibility. Mental health
professionals need not be sloganeers, but they must raise sane voices during difficult
times. A small minority has made active efforts and taken stands, but on the whole,
silence has transformed the profession's empathy into apathy. This collective silence must
be broken with concerted action toward healing and prevention.
(Dr H Shetty, Dr S Shah, Dr H Sharma, and Ms P Pushkarna visited the camps. The team
was supported by Mr S Desai and Ms A Patel from the Charutar Arogya Mandal.
Medical personnel and social workers of the preventive and social medicine department
of Pramukhswami Medical College provided critical inputs.)
Dr Harish Shetty is a psychiatrist and visiting faculty at the Tata Institute of Social
Sciences. 4A/11, Taxilia, Mahakali, Andheri (E), Mumbai 400093. Email:
kannaad@hotmail.com
2
REPORT ON THE STATUS OF MENTAL HEALTH SERVICES IN MAHARASHTRA
Prepared by:
Center For Advocacy In Mental Health(CAMH)
36B, Jaldhara Co-operative Housing Society,
Near Kanya Shala, Narayan Peth, Pune-411020,
Maharashtra.
A. STATUS OF MENTAL HEALTH SERVICES WITHIN INSTITUTIONS:
L Case Story ofA Mentally III Patient from A Central Jail in MaharashtraMy name is Meena. I am 23 years old Brahmin girl. I came here because of mental illness, not by
my own will. I was given shock treatment at J. J. Hospital. It was a horrible experience. I was in
great pain. I started hearing voices in my head when I was around 10-12 years old. These
belonged variously to a girl, a boy, a man, a woman, an old man and old woman. They would say
dirty things to me and shout out my name all day. They would abuse me a lot and accuse me of
doing all kinds of things. They would tell me that they had a microphone by which they could
hear everything that was going on in my head.
I used to be in continuous tension all day. I could hear the voices at ail times. Unable to bear it
one day, I took 50 sleeping pills. Even after that twice I attempted to commit suicide. I survived
but the voices did not go away. My irritation with others had not gone down. I would feel like
killing everyone who came in front of me.
One Wednesday, I remember it to be my Jesus-fasting day, I awoke early had my bath and
finished my prayers. I told my auntie's daughter (with whom I stayed) to clean the house. My
brothers had a lot of books. I used to try to read a dictionary, since I had studied a little at the
'modern school'. The dictionaries were lying around. I told her to pick them up. She told me that
she was not going to do it. I felt extremely angry and at the same time I heard the sound 'kill her,
kill her' in my head. So, I picked up the grinding stone and threw it at her head. She started
bleeding and I knew she had died. I became senseless and hysterical for 10-15 minutes. I kept
shouting. I could still hear the voices in my head. My brother tried to come inside but his friends
held him back. I was totally maddened by this time. The police arrested me and kept me in the
lock-up. I felt very remorseful after that.
They later took me to a Private Hospital. There they gave me shock treatment. 4 shocks they gave
me with anaesthesia and three shocks they gave me without anaesthesia. It was extremely painful.
I would not allow them to put the cloth in my mouth. It was extremely painful, I suffered an
enormous amount. It was the most horrible pain in my head. I remember feeling those three
shocks while I was awake. I hadn’t fainted- I could actually experience the shock treatments. I
don’t remember anything else about the time. I was very rude with the doctors. The voices in my
head stopped after that. I kept feeling remorseful about how I had killed my auntie’s only child.
My aunty came to visit me in hospital also. I asked her whether she had told the doctors to give
me shocks. She didn’t know anything about it. I was unconscious for 3 days after the shocks. It
was very traumatic.
I was shifted to a Jail after that. Here in Jail I repented for what I had done. I had killed my own
sister, my own auntie's flesh and blood. I used to cry a lot out of repentance. I would feel like
dying. Here I didn’t eat food for 15 days.
From here they shifted me to a mental hospital. Here the medications started and I stopped
hearing the voices in my head. But I would feel very depressed because I realised what I had done
and I missed my aunty a lot. I wanted to die all the time. Here again, I drank Phenol, but 1
1
immediately vomited it out. The conditions in the mental hospital were very horrible. The patients
were made to wash all the barracks and the bathrooms and toilets. It was very dirty. I was made to
carry cannister of water. The barrack where mental people used to stay used to always smell bad.
Here I feel suffocated. I don’t feel like meeting with or talking to anyone. I feel like remaining
alone all the time. If I stay here for a little while more then I think I will die. I want to go back to
my auntie again and I want to make her happy. I have appealed from inside but nothing has
happened. I feel like dying here. I want to be put in a less crowded barrack but I don’t want to
live alone. I was fair and healthy before but now my blood has been burnt black with the shocks
and the medication.
Meena’s story raises 2 major issues:
the (mis)use of Electro convulsive therapy and its impact on a person.
i.
the impact of the insensitive attitude, towards a mentally ill imprisoned person by the
ii.
jail as well as mental health authorities, on the patients health.
2. Status of Mental Services Provided in A Mental Hospital:
CRUELTY/STIGMA
-Contra Mental Health Act (1987)
- Contra Supreme Court Judgement in the case of Sheela Barse Vs Govt, of India (1989)
-Contra Supreme Court Judgement in the Erawadi case (1999)
Insider's Views> The patients are treated in an inhuman manner here and a large number of them (around
20%) are mentally retarded. Other patients here sexually abuse a large number of MR
patients.
Observation by the CAMH team
> One patient was lying on the floor due to reaction of consuming tablets. His fists were
clenched close to his body and he had drawn his knees up to his chin. He had severe pain.
One attendant came and hit files on his knees. The hawaldar who is in charge there, not ask
the attendant for the stretcher. The attendant pushed the patient very ruthlessly towards the
hospital.
Insider's Views-
ENVIRONMENT
> The situation here is extremely unhygienic. In the morning women are strewn all over the
place lying in their own refuse and excreta.
WORK
-Contra Bombay High Court Judgement in the case of Shukri Vs Govt, of Maharashtra, 1988
-Contra Peoples With Disability Act, (1995)
> Ayahs and the sweepers make the patients do all the menial jobs such as clean toilets, pick
up filth and rubbish, clean the forest of undergrowth, bring cauldrons of piping hot food on
their heads so that the skin on the top of their heads are also singed.
> All the undesirable work is pushed to the class IV employees. Doctors who are supposed to
check the patients regularly push their work to the ayahs who are not even provided with
gloves. These ayahs complain that they might catch infections from the patient. They feel
angry about the apathy of the authorities towards their physical health. They push the more
menial tasks to the patients.
COST
2
> The psychiatrist gets approximately Rs. 30/- per ECT. This bill is charged to the patient’s
relatives. We were told that a case was charged an entire bill of Rs. 82000/- after a span of
13 years.
TREATMENT
> Sometimes women are discharged before their treatment and counselling is over. This often
leads to relapse among such women.
> If the patient gets too excited, she is given ‘solitary confinement’ at first and if still
unmanageable, then ECT. Each ECT order is prescribed by the psychiatrist, countersigned
by the Medical Officer (MO) and the deputy superintendent. However this procedure is a
formality and the psychiatrist usually decides single-handedly.
Some Observations by the CAMH Team- What happened to Right to Health Care With
Dignity?
The solitary confinement rooms are bare and damp. The floor is strewn with urine and
excreta. Many of the patients were huddled in corners rocking to themselves, murmuring or
singing. These rooms are reserved for patients who get too exited. There are no separate
latrines in these rooms. There is only a low sloping grill on the ground level, which is the
only source of light for that room. The patient has to urinate or defecate near that grill itself.
The passage outside has a few windows but the entire structure is exceedingly dark and
dingy. The condition of hygiene and dignity required for the sanity of any human being is
non-existent here. One woman within the cell was completely naked. Nobody had made any
efforts at clothing her.
> Patients who are moderately recovered are given clothes, which consist of a long shirt and
loose pantaloons. Others who cannot do self-care are just provided with a long shirt. Nobody
is provided any underclothes. This is done because they often soil their clothes. There is no
hygiene or dignity to their appearance.
> Patients who are recovered and live in the recovery ward are given soap, comb, powder and
other items, which they keep tied around their neck or waist in a pouch. They do not have
any space where they can safely keep their belongings.
>
B. Status of Mental Health Services available at Government Health Centres :
> INFRASTRUCTURAL FACILITIES♦♦♦ NERPINGLE PHC (AMRAVATI DISTRICT): As per the community outreach program,
the Rural Hospitals are supposed to communicate with the PHC's about the monthly O.P.D.
for the mentally ill and distribute mental health posters for display in the community.
However the PHC doctor told that so far, there was no such intimation from the Rural
Hospital. Thereby there is no mental health service or referral given at this PHC.
*♦* The Rural Hospital of Morshi Taluka, Amravati had transport facility in emergency
cases. There were jeeps to shift the patients to the district hospital in case of emergency.
However it was told that for last three years the posts of drivers are vacant. So irrespective of
the fact that there are two government jeeps, the patients have to organize their own transport.
Since February 2003 onwards, mentally ill patients are referred to the 'Psychiatric O.P.D.’,
which is open only on every first Thursday of the month. Psychiatrist, medical social
worker and the psychiatric nurse from the District Hospital run this O.P.D.
❖ The head psychiatrist of the Amravati District Hospital informed about the mental health
team consisting of him, a medical and psychiatric social worker and a psychiatric nurse. This
team provides O.P.D. service twice a week. However there is no inpatient service in the
hospital. Counselling the patient / relatives is done in a group, i.e. the doctor, the social
worker and the nurse. There is no separate room for counselling or checking the patient. It
3
was told that awareness about mental health is very low in the area. Usually people approach
the civil hospital as a last resort, after they have exhausted all the money in the private
psychiatric clinics.
♦♦♦ As a result of the Erwadi tragedy, the government (state mental health authority) has
launched a community outreach programme in the rural hospitals, to improve the access of
mental health service in the rural areas. As per this directive, the mental health team visits
four RHs in the district, once a month and provide O.P.D. service with the assistance of the
R.H staff.. However this programme too is bound by the limitations faced otherwise by the
mental health team. Many times the doctor is caught with other non-mental health duties. The
hospital jeep is also not regularly available. This results in cutting down in the O.P.D. time in
the RHs.
*:* The District Hospital , WARDHA does not provide mental health service on a day-today basis. OPD service is provided once a week between 9.00 a.m. - 1.00 p.m. In the inquiry
ward CAMH team was told that the doctor who was in - charge of this ward had resigned six
months ago and Dr. A was looking after the cases of mentally ill. During interaction with Dr.
A he clarified that he was not attending to the cases of mentally ill in any way. He suggested
the CAMH team to meet Dr. B, who attended to the cases of mentally ill patients. On meeting
him we discovered that he is a cardiologist by training and psychiatrist by interest. Dr. B
shared that since there was no specific department for mental illnesses, patients were sent
from one ward to the other and from there to the next ward. When no diagnosis could be
made, finally the patients were sent to Dr. B for treatment. He mentioned that he was the only
source for providing mental health service in the hospital and that there would be no mental
health service in the hospital after his transfer in Bhandara district, which is planned in the
next month. This raises 2 issues:
Who provides mental health services at the government health care centres?
i.
Isn't service provided by an unqualified person equal to no service?
ii.
>
TREATMENT
Dr. N , Morshi Taluka, Amravati used to counsel and prescribe anti - depressants to
depressed patients. He claims that anti-depressants are useful for depressed heart patients. He
mentioned that he keeps the hysteric patients on compose and later on counsels them. I would
like to draw your attention to the fact that compose (a benzodizepam) is highly addictive.
Besides he has not received any training in treating patients with mental illness.
❖ Dr. B of the Wardha District Hospital, who is a cardiologist by training, prescribes
medications in the first place and claims to prepares the patients for further counselling. He
randomly prescribes antidepressants such fluoxetine and sertraline for depression, neurotic
disorders, anxiety, hysteria and psychosomatic complaints for a period of three to six months.
For violent patients, he prescribes tranquillisers such as diazepam, lorazepam and Clonidine
for a period of three to six months.
> MANPOWER
<♦ Due to paucity of human resource, the mental health team of Amravati District Hospital
has to work on several fronts as directed by the civil surgeon. ‘Mental Health’ appears to be
merely an additional responsibility for the mental illness team.
Wardha District Hospital: Medical and Psychiatric Social Worker Ms. A mentioned that
she conducts recreational activities for the patients. She told that these were fluctuating
groups as patients kept on coming between 2.00- - 4.00 p.m. on every Friday.
Contradictorily, when our team visited the hospital on Friday, no such group activity was
found anywhere. Later on Ms. A herself confided that she was too busy to spare time any
group activity. Although the annual report showed the name of a clinical psychologist, Ms. A
shared that he too was too busy to spare any time for that, and in fact, he was not a trained
4
clinical psychologist. However we could not trace this doctor during our stay in the city.
When asked about community outreach programme, Ms. A told that in the presence of the
earlier doctor, she had visited the rural hospital once, however in his absence since last 6
months, the problem did not emerge.
> CRUELTY/STIGMA:
❖ In case of extremely violent patients, the patients of Wardha District Hospital are chained in
locked cells along with medication. Dr. B told that the hospital authorities were aware of the
Supreme Court directives about illegality of chaining the mentally ill, yet it was imperative to
lock such patients for practical reasons; especially considering the fact that there was no
separate cell for housing the mentally ill.
❖ Dr. N from Morshi Taluka, Amravati mentioned that the class 4 staff of the hospital ridicule
patients attending mental health O.P.D as "O.P.D.” of paagals depicting the need for
'perspective building' for them.
5
India Together: Gaining ground on mental health - January 2004
Page 1 of 11
free advertisement: I
i El
indiatogether.org
The news that matters.
Home
WOMEN
[x] Women's issues
page
3
> Authority
> Business
> Dowry
> Education
> Finance
> Health
> Laws
> Literature
> Media
> Opinions
> Policy
> Profiles
> Resources
> Reviews
> Violence
> Work+Life
a_
a_
s'
l,
Qp-Ed
Interact
States
Gaining ground on mental health
Mental health issues of women are gaining ground
in the women’s studies discourse. Vibhuti Patel
reports on current developments and advocates
progressive directions.
Services
Support
• Write the author
• Women
• Women's health
• Printer friendly version
January 2004 - Mental health issues of women are gaining ground in the wome
discourse. The women’s movement has provided fresh inputs in terms of indi
group counselling, popularly known as “Consciousness raising” exercise, a form
counselling that enables women as a group to share experiences, problems
dreams, utopia and action plan for rebuilding shattered lives. This process o
feminist consciousness allows women to recognise that what they perceive a
problems are shared with others in a non-threatening and non-power oriented al
It also enables women to realise what they think of as resulting of living in
society. “Consciousness raising can be seen as enabling women to overc
consciousness.” It empowers women to come to a realisation of their own potem
them autonomous, self-dependent in their decision-making power and emotio
reliant. It is an ongoing process that brings about personal and collaborative
opposed to structural change. The need for small groups and informal group di:
emphasised in this method.
Manifestations of depression
Depression in women manifests in headaches, sleepless nights, constan
detachment, irritability, loss of appetite, dryness of mouth, fear, self-blam
concentration, lack of interest in any kind of activity. Although chronic headach
be psychosomatic, they can be caused by depression or anxiety. Thus, counsellin
you identify and address emotional concerns and should be considered as pa
treatment.” There are two types of stress leading to mental disorder. Biochemic;
are hormonal fluctuations at the onset of puberty, pre-menstrual phase, post par
and menopause. According to Dr. K Ravishankar, Estrogens have an effeci
chemicals like serotonin and/or epinephrine that are involved in headaches. An
in serotonin levels has, in fact, been implicated in disorders like migraine and dej
RSS Feeds
S
i"
Psychosocial stressors originate from the external social environment such a<
inferior social position, lack of power, homelessness, economic hardships, ma
natural disasters. They create learned helplessness (women’s seeming passivity ii
of crisis such as domestic violence, accidents, etc) and reduce motivation to leac
life. Stress related mental health issues are illness in the family, death of one
file://C:\Documents%20and%20Settings\Administrator\Desktop\2005\Health%20and%20H ... 5/8/2007
India Together: Gaining ground on mental health - January 2004
[ent
1]___
[x| Manushi, Issue
145
Page 2 of 11
divorce, accident that might reduce or destroy women’s ability to shoulder rest
After marriage, women get displaced which brings about cultural loss and bei
loss of social networks and supports, loss of traditional healing sites. Psycholog
through which women pass are — enduring, suffering, reckoning, reconi
normalising. The successful completion of therapeutic cycle depends on how con
physical and emotional systems are. Sometimes hysteria can also open
opportunities and increased freedom/space with added costs. Women cope with
crying, talking it over, praying and engaging in creative work-music, art-era
studies-community work, team building.
Approaches to mental illnesses
s
SERVICES
s
One universalist approach uses diagnostic categories of mental illnesses such a
schizophrenia, psychosis, mania, phobia, paranoia so on and so forth. Psychiatri
does not take cognisance of material reality faced by women on a day-to-day e
levels. It obscures social reality such as riots, natural disasters, fire and acci<
dealing with phobia among women. While working with women victims of
should know that their phobia about men has a basis in the fact that they have
killings and rape. Hence, medicalisation of mental health in the Draft Natio:
Policy 2001-III has been criticized by women studies.
x|
> Advertise
> Calendar
> Letters
> Links
> Newsletter
> Space Share
> Submit
-f
The philosophical basis of psychiatry as a bio-medical discipline prevents me
professionals to take into consideration larger reality and macro issues resu
socio-economic and political factors. Psychiatry focuses on treating the
symptom while ignoring the disease. ‘Diagnosis’ frequently arouses protests of i
about labelling people as ill and treating them as impersonal objects. Limitatic
medical perspectives lie in their narrow focus on somatic and psychological fact<
diagnostic efforts, ignoring the impact of socio-cultural and socio-demographic
India, the focus is more on the treatment of the illness, not on preventive and
efforts. Marginalisation of mental health concerns results from the understa
mental distress is a manifestation of an individual problem, not directly relate<
oppression and not common to all women.
Another, different approach emphasises cross-cultural psychiatry and evaluates p
of mental illness from within a culture. Traditional treatment of the mental illne
be meditation, yoga, group singing and listening to the discourse. A culturally
counselling on mental health consequences of trauma takes into consideratior
socio-cultural environment.
The two approaches gave way to new cross cultural psychiatry where the emph;
the different contexts of mental illnesses in different cultures, not on b
categories. Now, there are no two opinions about the statement that psychothen
keep into consideration multicultural aspects of women’s existential reality.
Psychiatrists use chemotherapy i.e. administration of anti-depressants and sedati
therapy which induces shock, with or without convulsions, in a patient by means
or electric current through brain. Individual psychotherapy includes hypnosis, st
supportive therapy, re-education, desensitisation and other forms of consultat
psychotherapy, family therapy and psychoanalysis.
file://C:\Documents%20and%20Settings\Administrator\Desktop\2005\Health%20and%20H ... 5/8/2007
India Together: Gaining ground on mental health - January 2004
Page 3 of 11
Four phases of Healing Cycles are:
1. Enduring - anxiety, grieving and loss of past
2. Acceptance - reality testing, preparedness and reckoning of the future recoi
evaluation of self and resources, recuperating
3. Recovery - rebuilding life, maximising options, setting new goals, healing
4. Normalising - stability and routines, building relationships and community
Unequal relationship between professional counsellor, who is UP THERE and i
who is DOWN BELOW creates a communication gap. In case of women, this in
compounded by subjugation of women by the patriarchally structured psychiati
The focus on ‘feminine qualities’ pathologises all physiological changes of a ’
childhood, adolescence, reproductive age and menopause.
Worsening socio-economic, political situation and mental health of women
Experiences from both industrialised and developing countries have reveale
prevalence of common mental disorders or minor psychiatric morbidity is high
urban low income and marginalized population. Women among them are c
vulnerable. Globalisation, Structural Adjustment Programmes, increasing coi
neighbouring countries and ongoing sectarian violence on caste, ethnicity and
lines within the country have put the population of our country at high risk
illnesses.
Alert India is a large NGO with 550 community workers working among the m<
sections of Mumbai metropolis. Their women health workers found that women
to deal with financial hardship, experience tremendous stress. Moreover, won
community are affected differentially depending on their own place in the Inc
economic hierarchy. In this regards, female-headed households are most vul
mental distress. The mental health professionals are only geared for the episodi
and not the enduring disasters. Hence, there is a need for interdisciplinary me
interventions.
Professional counsellors act as facilitators in Self Help Groups (SHGs). The
technique of mutual counselling to identify areas of strategic interventions.
Need for Culture Specific Approach in Counselling
From respect for basic human rights it follows that the counsellor addresses
concerning cultural mindsets and behavioural variety that determine womei
responses to tragedies. If this variety is not appreciated, counselling will enc
reductionist and homogenising. Here the soft wear is not formal education, but li
health of women victims and survivors of tragedy demands multifaceted
Individual counselling by the professional counsellor can be helpful in breaking
same time, women with similar experiences can empower each other by narr
problem areas and finding solutions.
Patriarchal biases of the mental health establishments
The mainstream mental health professionals are unable to impart the required c
file://C:\Documents%20and%20Settings\Administrator\Desktop\2005\Health%20and%20H ... 5/8/2007
India Together: Gaining ground on mental health - January 2004
Page 4 of 11
to women due to misogyny. Stereotypical understanding about women’s role in
and society governs their psyche and if the so-called “mentally ill” woman doe
that mould, she is declared ‘socially incompetent’ woman. Witch hunting of lesb
mainstream psychiatrist is so strong that even All India Institute of Medical Sci
special package for counselling, “to correct deviant behaviour” of the
Subordination-domination relations between men and women are re-emphasi
mainstream counselling.
Sexual violence and mental health
Sexual assault, molestation, rape, sexual harassment at workplace, child sex
nuisance calls cause psychological disturbances among girls and women. The
sexual violence sparks off tension and anxiety at a dangerous level. Their me
problems are manifested in anxiety, fear, avoidance, guilt, loss of efficienc
coordination, depression, sexual dysfunction, substance abuse, relieving the
incidents through memory, suicidal attempts, eating disorders, disturbed sleep pa
of encountering such situation once again. It is found that “women who under^
sexual violence experience a loss of self and self-esteem following the shock ii
them. When there is a continuous period of traumatic stress, it becomes chronic
the individual’s ability to do any kind of constructive work.” Hence, this for]
violence towards women is an important issue that demands public attentio
women’s organisations have taken up this issue at a local, national and global lev
Domestic violence and mental health
Discourse on mental health of women in the family situation gained serious coi
in the context of campaign against violence against women. In domestic
situations, predicament of women is determined by their position in power-relat
vis the rest of the family members. Research which attributes deviant bel
adolescent girls to their working mothers place women on a guilt trip by
focussing on single parameter and ignoring factors such as peer-pressure, med
standards of morality in our society and power relations in the family. Such i
used by some counsellors to cage women into domesticity and divert the attei
generation of genuine support system for developmental needs of the daughters (
mothers. Women’s rights organisations which are doing support work for
distress have started giving due importance to counselling.
Adolescent girls and counselling
The most mind-boggling problems faced by adolescent girls are decision-mak
day-to-day life, self-dependence and career. Rapid changes in the socio-ecoi
cultural reality, parental expectations, values and norms, rising levels of comp<
pressure during examination time and a break down of traditional family stn
factors that accelerate this alarming trend. Examination related anxiety results
rise in girls hurting themselves deliberately, leaving homes or killing themselvi
failure is a root cause of all qualms. Large number of students and their parents e
professional help. Consulting a psychiatrist is no longer a taboo as the psychiatric
to cries for help from a crippling academic burden. According to them, we 1
problem parents than problem children.
file://C:\Documents%20and%20Settings\Administrator\Desktop\2005\Health%20and%20H ... 5/8/2007
India Together: Gaining ground on mental health - January 2004
Page 5 of 11
Providing good and healthy role models is very important. Parents who want the
to develop high self-esteem should make a point of treating them with respect ai
Concept of fiscal hygiene is important for girls to understand the value of ck
earned through hard work. Today’s adolescent girls are at the crossroads,
crossroad leads to new roads. Information revolution has made adolescent girls n
and precocious. They have to enhance their knowledge base. Broadening one’
horizons is a sure way of tackling the crisis within oneself. Today, girls find that
instilled in them since their childhood are hollow in real life. Romance is found
and consumerist. The economic security is bleak, emotional security is becomir
of uncertain times. Globalisation has led to the emergence of apparently homoge
styles, necessities and comforts through media-images, whereas the reality oj
significant variance with resources required to maintain such a life. This F
deepened the crisis of the youth. A dictatorial atmosphere in the family, e
institutions and in the community life, make adolescents feel left out of the
making processes affecting their lives.
Hence it is very important to understand that,
“Inclusion is trend,
Such as democracy,
Freedom and justice for all.
All means all,
No buts about it.
Inclusion is opposite of exclusion.
Inclusion is no to boycott.
Inclusion is a battle cry.
Challenge to the parents,
Child’s cry for his/her existence...
For welcome, for embrace,
To be remembered fondly.. .for award
For gift of love.. .like surprise,
Like treasure.
Inclusion means clean game,
General knowledge, courtesy, hard work.
Inclusion is great in its simplicity,
And surprising in its complexity.
Instead of investing in jails, mental asylums, hospitals, refugee camps,
To canalise resources for creating true homes,
True life, true human beings...
For humanising life.
— Mai
Both in private and in the public spheres, we need to give more space for deve]
the adolescent girls.
Counselling for substance abusers
file://C:\Documents%20and%20Settings\Administrator\Desktop\2005\Health%20and%20H ... 5/8/2007
India Together: Gaining ground on mental health - January 2004
Page 6 of 11
Support resources for substance abusers are counsellors, family members, signif
and school or treatment staff. Group therapy is an effective intervention me
abusers. It facilitates the process of recovery of addicts. Sharing of experien<
abusers shows them ways to empower each other. Self-help groups of abusen
effective as they avoid problems generated due to different wave-lengths.
Counselling for HIV-AIDS patients
This is very important issue faced the 21st century. Counselling for dealing \
stigma and creating an alternate support network are the most important
providing emotional support to the HIV-AIDs patients. The Lawyers Collective 1
Unit holds monthly drop-in meetings, with an objective of sharing information, e
and resolve mind boggling issues affecting the lives of HIV/AIDS patients. It als
legal aid and allied services to the needy. “The main objective of the unit is to {
promote the fundamental rights of persons living with HIV/AIDS who have bt
their rights in areas such as healthcare, employment, terminal dues like gratuib
marital rights relating to maintenance, custody of children and housing.”
Electronic media and mental health
People engage with the T.V. because there is vacuum outside. Different st<
morality for men and women, are present in messages from the film, television
advertisement industry. Boys and men who watch pornography are on the lo
innocent adolescent girls. These girls are the victims of pornography, blac
physical/psychological coercion. Adolescent girls working, as domestic workers
any emotional support, as there are hardly any television and radio programme
student urban youth. Dehumanisaton of women can be prevented by promotion o
agency in the media so that women can lead intellectually, psychologically and e
self-sufficient life.
Counselling in the shelter homes for women
The most promising solution to confusion and disorientation among the women
shelter-homes is a democratic space for brainstorming as autism is one of
problems faced them. Informal set up is more congenial to their personal ;
counselling. Workers at the shelter homes for women and girls need to be
understand that behind every behaviour, there is a story. It is important for thei
the story. Panel of psychotherapists and psychoanalysts in the shelter homes s
conduct the staff development programme so that the staff can handle post tra
disorder among the inmates with empathy rather than resorting to victim blaming
Mental health and reproductive rights
Societal attitude towards Indian women as son-producing machines creates pain
problems for women. A woman’s body is de-linked from her subjectivity. Pn
syndrome (PMS) and Post Partum depression (PPD) are regarded as general (
concerning women’s reproductive abilities. The weapon of premenstrual synd
debilitating factor has been used to run down women in the family and at the ’
PMS is a political category, which conveys that biology is destiny for women,
focusing on the genuine issues concerning premenstrual discomfort in terms
file://C:\Documents%20and%20Settings\Administrator\Desktop\2005\Health%20and%20H ... 5/8/2007
India Together: Gaining ground on mental health - January 2004
Page 7 of 11
headache, cramps, headaches resulting in to depression and crying spells, PM!
reductionist and reactionary explanation for women’s discontent. Women don’t
to decide, how many children should they have and at what interval.
New reproductive technologies (NRT) have robbed women of their individ
reduced them into spare parts for either scientific experimentation and/or sale. N
women only for their ovaries, uterus, foetus, that too preferably male. NRTs h<
tremendous psychological burden on women in the arena of sexual activity for f
or only for recreation without procreation with the help of contraception oi
Research over last 3 decades has highlighted mental problems associated wit
induced abortions, long acting hormone based contraceptives or conception indue
Instead of using humane healing techniques of music, fragrance, get to gathers t(
discomfort during pregnancy and post-partum depression, bio-medical inter
giving tranquillisers and electro-convulsive therapy are promoted by the psychia
is the most vulgar example of the medicalisation of the natural processes oi
bodies. Gender sensitive training programmes should be organised for medical
primary health centres and women health workers adopting perspective promo
UNFPA.
Menopause and mental health
Many psychologists have attributed harassment of daughter in law by her mothe
menopause. But it is not true for all women. Many women find it a liberating ex]
stay with their in-laws. It all depends on how society and family treat an agir
Pathologisation of menopause and negative attributes given to “old hag” (sadc
experienced by women are responsible for identity crisis and depression amo:
during this period. Here, the role of counsellor is to recommend activitiej
actualisation and a healthy diet and vitamin supplements to menopausal womar
is very important to increase conversion of androgens to estrogens.
Women and epilepsy
Disability and impaired quality of life caused by epilepsy can be reduced by “]
and psychosocial referral counselling on how to live with refractory seizures and
vocational rehabilitation.” Persecution and discrimination against epileptic won
be prevented by giving scientifically accurate public education through mass i
the curriculum of community workers training programme, module on epileps
convulsions should be incorporated.
Mental health of women senior citizens
The most talked about problem concerning mental health of elderly women is de
“loss of cognitive functioning, memory, language abilities, abstract thinking and
Dementia is often reversible. Irreversible dementia can arise due to amnesia,
disease and Alzimer’s disease(AD). Modern medicine treats this problem wit
replacement therapy, non-steroidal anti-inflammatory drugs and vitamins. Femi
citizens deal with mental problems of elderly women by providing spiritual!;
emotionally and intellectually stimulating group life to them. Discourses, singii
social service, meditation and mutuality and reciprocity in human relations r
file://C:\Documents%20and%20Settings\Administrator\Desktop\2005\Health%20and%20H ... 5/8/2007
India Together: Gaining ground on mental health - January 2004
Page 8 of 11
contribution towards their mental health. Vardhana, a group of feminists h;
women above 60 years of age as “Women of Age” and has provided a deme
development oriented platform to Women of Age.
Mental health of women in mental hospitals
Pathologisation of women by using diagnostic labels is a major cause of stigmati
ostracism for women. Women’s groups are demanding that pigeon-holing of f
set slots must stop. Interaction with the mental health professionals is used by
members and the community to declare the concerned woman “unfit” to live in i
be a parent, function as an autonomous individual or take up a job. Husband’s f
the episode to dispose her of or debar her from property right or right to
matrimonial or parental home. “Madness certificate” of the mental health profes
used by husbands to divorce, desert or throw out wives from their matrimon
Women are admitted in the mental asylum as per the directives of the Mental t
1987 and Lunacy Act, 1912.
Once dumped in a mental asylum, it is impossible for her to get out of it
complete recovery. “ Women in the mental hospitals have fewer visitors, are abz
tend to stay on longer than men within the institution. There are fewer volunta
among women than among men. Even in adjudication for a woman’s institutio
the official discourses are often coloured by the sex role stereotypes that the jud
officials and the staff in mental hospitals uphold.” Remarks of a social worke
visit to the mental hospital are apt, “The interaction with female patients made i
Almost all of them were abandoned/dumped by families or the police and cour
admitted after they hit the rock bottom. Most of them were forced to face violeni
in their lives and had painful and atrocious account to tell. In many cases, one
(although without an in-depth study, one can not claim and prove) that the ment
ill health had its roots not in a person’s biology or psychology, but in society, in
environment”.
The iron wall of secrecy about the administration of drugs, surgery and ECT anc
effects needs to be condemned by citizens initiatives and ethical medical practiti
long lasting side effects of biomedical approach need to be highlighted. C
hospitals need to focus on psychotherapy and counselling which involve the
produce positive results and no negative side effects.
Psychological problems of women in the police custody and prison
Activists working on prison reforms have demanded humane code-of-cc
governance of police custody and prisons, so that the inmates are not affl
permanent psychological scars. Solitary confinement of women prisoners t<
verbal articulation from them. Inter personal violence among prison inmat
reduced by counselling, group discussions and creative expressions. Wome
prisoners should not be forced to stay with hard-core criminals in the custody or j
Role of Support and Self-help Groups (SHGs)
SHGs provide democratic space for rebuilding broken lives. Non-power orieni
interest groups provide stimulus for canalisation of creative energy. Mutual c
file://C:\Documents%20and%20Settings\Administrator\Desktop\2005\Health%20and%20H ... 5/8/2007
India Together: Gaining ground on mental health - January 2004
Page 9 of 11
focussing on experience sharing without preaching or giving sermons
psychologically distressed women reorganise their life and enhance their poten
Out Centres can provide platform for community mental health intervention. H
the endorsement from an expert, “At the height of feminist activism in the
1980s, there was excitement in the air as women shared experiences about tl
their families, their lives and encounters. The growth of women’s confidenc
esteem knew no bounds as they challenged established theories about law, wo
equality and medicine. They talked late into the night, wrote pamphlets, stuck w;
spoke at public meetings, filed writ petitions. They felt reassured that theirs \
isolated or individual problem. The group’s endorsement and sharing <
experiences perhaps did much more for mental health than all medicines in tf
books. The women’s movement helped avert many breakdowns.” Enduring 1
engagement at community level can be group singing, festival celebrations, dis<
women’s issues and public meetings.
Developmental inputs
Cosmetic counselling offered by agony aunts is of no use. Breakthrough cou
need of an hour. To make women’s material reality more secure, liberating and
the only alternative to get out of repeated attacks of mental illnesses. Devc
counselling aims at removal of chronic conflict situation in women’s lives that is
with high mental health morbidity. It is more than a remedial service. It be
involvement, readiness and commitment on the part of the counsellor are nec<
basic conditions for counselling success.” It is concerned with the develoj
facilitation of human effectiveness. It increases self-direction and evolves bette
solving and decision-making abilities. This is the central axis around whic
therapy or counselling revolves. It emerged in the wake of the women’s movei
alternative to hegemonic patriarchal mental health establishments which depend
medical approach to deal with the innate feeling of unhappiness in women. A
time to evolve safety nets in the community and criminal justice system for prote
physical abuse.
Budgetary allocation for medical aid to treat mental illnesses of women
enhanced. Mentally ill women need legal protection in terms of property rights a
dwelling place. We need to create protective environment in personal and pul
prevent mental illnesses among women. E.g. efforts to prevent man-made disast<
riots, loot, rampage. Mental illnesses result into deskilling of the individuals <
Hence, there is a need to evolve a plan of action for the re-skilling baset
preferences and abilities. Half way homes should be created where, the mentally
can do productive work during the day and go home in the evening. After th'
from the mental illnesses, they should be employed. Financial security helps in
their sense of self-esteem. The most successful healer is one who avoids victii
and provides patient listening. After talking / catharsis, the seeker feels better.
Girls and women with communication disability need special help. At the s
“Reversing the process of alienation by consciously building community neb
must. Mental health professionals should be seen in the community rather tJ
secure institute or clinics” avers a well known psychiatrist. Dr. Harish Shetty.
Training programmes on counselling
file://C:\Documents%20and%20Settings\Administrator\Desktop\2005\Health%20and%20H ... 5/8/2007
India Together: Gaining ground on mental health - January 2004
Page 10 of 11
Sensitization and training of general practitioners and other health personnel
health, particularly, minor psychiatric morbidity (anxiety-depression) is a must,
need for social counsellors at health posts and public hospitals who are in touch '
providing institutional support to women in social distress. Sensitization o
community workers, youth groups, women’s organisations is extremely importar
sessions for professional and para-professional volunteers should focus on
networks, group cohesion and solidarity. Training should include modules on ir
skills, history taking, mental status examination. Electronic and print media
trained in sensitising the general public about psychological response to vi<
providing information about referral services as women and children affected b
violence, man-made or social disasters have special psychological needs.
Counselling ought to make women more aware about their problems and the
they face. Therapy can provide alternatives to deal with their problems. Counsel
used to bring to the fore the cognitive facility required to recognise danger and th
to assess the options and to leave if necessary, among women victims o
Counsellor’s have become astrologers. Counsellor should be proud with the ai
humble with the courteous. Supportive counselling alone won't do. At times, pro
necessary. Role-playing is an excellent procedure for learning about counsel
playing situations can be easily developed from the experience of people. Th
valuing and respecting others must be observed by the counsellor. Common che
required from the counsellor are concern, emotional investment, cognitive d
sensitivity and introspection. The counsellor should know that healing is
empowerment.
Conclusion
Civil society and the state should provide more and
Respect for diversity, ph
more opportunities to women of all age groups for
and multicultural outlool
self actualisation so that women can achieve high
democratic and tolerant i
level of mental health. Respect for diversity,
that is conducive for mei
plurality and multicultural outlook ensure a
health of women.
democratic and tolerant milieu that is conducive for
mental health of women. As compared to
institutionalisation based mainly on bio-medical intervention, community or fa
rehabilitation of mentally ill women based on human touch is far more effective.
Vibhuti Patel
January 2004
Dr. Vibhuti Patel is a Reader in the Department of Economics, Univeristy of M
Member Secretary, Women Development Cell. She is a founder member of For.
sex Determination and Sex Pre-selection, and a trustee of CEHAT.
References
1. Kramarae, Cheris and Dale Spender (Ed.s) Routledge International Encyc
Women- Global Women’s Issue and Knowledge, Vol.I, Routledge, New York
2000, p. 221.
2. Goel, Deepak “History of Headache”, Health Action, Special number-
file://C:\Documents%20and%20Settings\Administrator\Desktop\2005\Health%20and%20H ... 5/8/2007
India Together: Gaining ground on mental health - January 2004
n
Page 11 of 11
Neurological Disorders, Vol. 15, no. 6, June 2002, pp 13-17.
3. Times of India, 21-3-2002.
4. Davar, Bhargavi “Draft National Health Policy 2001-III, Mental Health
Misconceptions”, Economic and Political Weekly, Vol. XXXVII, No. 1, Jan 5
pp.20-22.
5. Amico, Eleanor Reader’s Guide to Women’s Studies, Fitzroy Dearborn 1
Chicago and London, 1998.
6. Shertzer, Bruce and Shelley Fundamentals of Counselling, Houghton Miffin C
1968, p. 14.
7. Kearney, M.H. (1999) Understanding Women’s Recovery From Illness an
New Delhi, Sage Publications.
8. Noonan, Ellen Counselling Young People, Methuen, London and New York, 1
9. Vindhya, U, A. Kiranmayi and V. Vijayalaxmi “Women in Psychologica
Evidence From a Hospital Based Study”, Economic and Politcal Weekly, Oct.
Vol.xxxvi, No. 43, pp.4081-4087.
10. Medico Friend Circle “Carnage in Gujarat- A Public Health Crisis”, Murr
2002.
11. Ali, Nasir and Surinder Jaswal “Political Unrest and Mental Health in Srin
Indian Journal of Social Work, Special Issue- Mental Health Consequences of
Vol. 61, Issue 4, October 2000, pp.598-618.
• Write the author
• Women
• Women's heal th
• Feedback : Tell us what you think of this page.
© Civil Society Information Exchange Pvt. Ltd., all rights reserved.
Home | About us I Support | Feedback | Disclaimer
IT
file://C:\Documents%20and%20Settings\Administrator\Desktop\2005\Health%20and%20H ... 5/8/2007
India Together: Special or segregated? - September 2003
Page 1 of 4
a
indiatogether.org
Information for the civil society
fx]
X
□
Home
a
EDUCATION
X
Education issues
page
Op-Ed
States
Campaigns
Special or segregated?
Lakshmi K argues that we must integrate mentally
disabled children in the mainstream schooling
system with modifications, as opposed to
segregating them in separate schools.
Events
Finances
• Wril^h^author
• Education
• Disability
• Printer friendly version
(al
H
> Policy
> Profiles
> Women
tel
|[3;
El
[enter e-mail ID
I
n
s’
SERVICES
a
> Events
> Feedback
> Advertising
> Newsletter
September 2003 - Roughly 10% of our population falls under the category of:
of disability, out of which around 2% are people with mental disability. A great
already been made in the field of education in recent years and efforts to
individual become literate has gained a lot of momentum. In this context, tl
special education is still in its infancy as far as mainstreaming and integrated edi
concerned.
If ever they do, children with special needs often receive services in a segre
special school. Though segregated schools work to help these children an
specialised services, the entire possibility of mainstreaming these children is los
with special needs often have to commute long distances to reach their schools a
only few centres in any city which offers specialised services. In small towns an
there can often be no local options. According to the Persons with disability
schools, private or government, are not to discriminate children with special neec
have to provide the services they need. The implementation of the rules thou
observed in their absence.
A serious social problem these children face as they grow is that they have to
their peer group why they go to a special school and not to a regular school. A
child irrespective of his/her condition. A child has not only to learn to cope w
individual condition but also the harsh reality of society's attitude towards hiir
human being, it is but natural to yearn for acceptance and recognition.
Consider the life of a mentally challenged child in a segregated setup or in z
special school. The most formative years of our life is our childhood, the time
faculties are sharper and learning through modelling and imitation is higher. Tl
who go to a special school often interact only with other special children and n
learning behaviours and skills that could have occurred had they been in the c<
normal children. For example, a child in a special set up uses language and comr
skills and behaviours that are used by other special children and which is most (
in nature.
On the other hand, as a special educator with 12 years in an integrated school, I
file://C:\Documents%20and%20Settings\Administrator\Desktop\2005\Health%20and%20H ... 5/8/2007
India Together: Special or segregated? - September 2003
> Submit
> I T -L inks
> Shop
la,
a
Page 2 of 4
that not only are there advantages for the children with mental disabilities
families, but there are gains to be made by the the normal kids too. When we
mainstream and offer services to children irrespective of their individual differen
equally promoting the rights of every individual.
The term "adaptive behaviour" as understood in the field of education, pla
important role in any person's life. It generally refers to our ability to copc
changing environment and adapt according to the situations we are placed ii
special child interacts with a normal peer he gets exposed to a wide variety o
inputs in terms of content and vocabulary. Importantly, age appropriate beha
often acquired through the company and the expectations which the society ii
normal children. The urge to be a part of normal peers adds to the zest in imprc
basic skills.
In an integrated environment, a disabled child gets to participate in all the activit:
conducted in the special classes and they as well get a chance to participate in
curricular activities alongwith their normal peers. This boosts their confidence
they become comfortable with their own condition. They receive the same p
express themselves as any other student. Several studies in the field of educ
shown that children learn a lot through imitation and modelling. Integrated educa
the opportunity for the special children to learn socially acceptable behaviours,
is also the satisfaction parents derive upon watching their child with special nee
part of a regular school.
Secondly, the special children can have as many choices of schools as their no
do. Long commutes can be avoided by having access to the schools in the
neighbourhood. It is much easier to use existing facilities for schooling rather
resources to set up special schools. If all the existing schools admitted just a f
children, the need for special schools will be eliminated. Just as we employ t
teach various subjects and include them in the curriculum, we need to empl
educators or people who are inclined to understand and work in the field of di
offer services for special children. Most parents of special children are willing
the services needed by their children.
Ideally, in the primary school, i.e. until the age of 12 - 13 and depending upon t
the child, s/he should be integrated in a regular classroom for a part of the day.
all the extra-curricular activities like P.T, Games, drawing/painting, socially
productive activities, cultural activities like music, dance, annual day, sports day
other festivals conducted in a school should have full participation by the speci.
along with the normal peers.
Academics is one area which needs close
monitoring by the special educator and this is
largely depends on the level of the child. Hence, the
focus should be child centered and a proper
Individual Educational Plan (IEP) should be
prepared while integrating these children into
regular classrooms. A special Educator has to be
specially deputed for foilow-up for special children
Contrary to the fears of tl
the normal students genei
to sit next to the special c
help him/her with activiti
colouring, teaching spelli
tutoring in academics anc
teaching
teaching simple
simPle games.
games,
who are placed in regular classrooms. My experience
tells me that no more than 3 to 4 students should be assigned to a special edu
file ://C :\Documents%20and%20Settings\Administrator\Desktop\2005\Health%20and%20H ... 5/8/2007
India Together: Special or segregated? - September 2003
Page 3 of 4
Special educator's job is to ascertain the level of the child, plan an IEP, place the
regular classroom and provide follow up services in all the areas that need develc
Therapeutics like physiotherapy, speech therapy, occupational therapy, pre
activities etc. should also be included in the lEPs. Facilities for these services s
part of the programme, if possible on a regular basis or atleast on consultal
teachers in the regular environment need orientation in terms of the condition of
children and also a thorough shift in their attitudes and beliefs about these child
this may not happen overnight, it can surely happen within a generation.
At some point of time, it becomes necessary to orient the parents of normal child
promote a healthy and more accepting attitude towards special children. It is a jo
the special children participating in the activities conducted in a regular classrc
normal child, it gives a great feeling of satisfaction that he/she is able to he
human being.
In fact, contrary to the fears of the adults, the normal students generally opt to
the special child and help him/her with activities in colouring, teaching spelling,
academics and help in teaching simple games. During snacks break or lunch brea
all often seen sharing their snacks, run to the nearest stall to purchase cool
snacks and often indulge in mischievous behaviour. Isn't growing up a lot of
way? I certainly believe that when we can enjoy life for what it is, the pain and s
alienation caused by our conditions are easily erased.
Segregation of the disabled leaves us unaware of the realities of disabilities ai
possibilities. Even reading material in this field is not to be seen in our schc
Attitudes, beliefs and values are often learnt and passed on from one generation
at our places of learning. The few facilities for the disabled which are available
are most often due to families with disabled members having invested their
effort and money despite disadvantages they may face themselves. Families w
disabled members seldom participate in institution building or providing se
people with disabilities. In this context, integration in school will be benefic
society at large as well.
As normal children become aware about disabilities, there is a greater chance the
grow into sensitive adults, regardless of whether their own families had specia
They are also more likely to contribute towards providing facilities for the disabb
they knew someone in their school. They will also take inspiration from the eff
disabled, and try their best just as they see the disabled do. Most importantly, ft
better prepared for possible disabilities in their own lives or families if it does oc
their adult lives.
The changes to our system are only a matter of adaptation and awareness, but <
they will go a long way to accommodate every one in the general population. Cl
very adaptive in nature and welcome changes with ease. We have to take advant
and facilitate normal children help and seek companionship with the special
accept him/her as a part of the group. I have seen this working successfully with i
younger ages when the concept of being normal or abnormal etc., is not yet fully
or fixed in a child.
file://C:\Documents%20and%20Settings\Administrator\Desktop\2005\Health%20and%20H ... 5/8/2007
Page 4 of 4
India Together: Special or segregated? - September 2003
Greater integration in schools will reduce the fear that the normal population
disabled, especially the mentally disabled. Our society needs solutions to the pro
it sees, not fear. It is really upto the schools to take the initiative. And schools
already led the way in successfully integrating special and disabled children
support and encouragement.
Lakshmi K
September 2003
Lakshmi K works as a special educator in Niraj Public High School, Hyderabad.
Afterword: Segregated facilities are advantageous for severely affected speci;
who need a lot of nursing care and extensive therapeutic intervention. Close moi
specialised clinical team (the physiotherapist, occupational therapist, ps;
psychiatrist, medical doctor, special educator, counsellor etc.) is done better in a
institution where all facilities are available under one roof.
• Writethe author
• Education
• Disability
• Feedback : Tell us what you think of this page.
© Civil Society Information Exchange Pvt. Ltd., all rights reserved.
About us | Finances | Feedback | Disclaimer
X
file://C:\Documents%20and%20Settings\Administrator\Desktop\2005\Health%20and%20H ... 5/8/2007
Special Procedures
An- either Special Rapporteurs, Special Representatlvcti of the Secretan-Gencral,
Introduction to Special
Procedures
independent Experts or Working Groups
CHR did not have powers of investigation until I97<ls
1970s: special mechanism for Palestine and then other country situations
1980s: thematic mechanisms begin
SOOTH ASIA HI MAN RIGHTS DoCUMENTATIc <N CENTRE (SAHRDCf TRAINING SERIES
AIHIUSTIISM
SOOTH ASIA HUMAN RIGHTS D< ICUMENTAT1I IN CENTRE ISAHRDCI TRAINING SERIES
AUGUST
Country-specific special procedures
Thematic special procedures
Special Rapporteurs monitoring the human rights situations in specific countries
•
Special Rapporteurs receives information from NGOs about these countries, may
visit them and report to the CHR
•
Most countries are not covered by specific special procedures, so there arc other
that cos er themes in violations
•
There are special procedures on:
In Asia, Myanmar, Cambodia, Afghanistan, Iraq all have country-specific special
w ide range of civil and political rights
procedures
some economic, social and cultural rights
some vulnerable groups
SOI ml ASIA HI MAN RIGHTS D< XTIMENTATI
TION CENTRE (SAHRDO TRAINING SERIES
AUGI 1ST
IT 3004
SOUTH ASIA 111 MAN RIGHTS I>i H 'llMENTATIi >N i.l-NTRH ISAHRIX'I TRAINING SERIES
Al;iil>ST3«M
Selected current mandates
Torture
Executions
Operations of special procedures
Receive information from NGOs and other sources
Disappearances
Freedom of expression
Religious intolerance
Racism
Urgent action appeals
Sale of children
Migrant ssorkers
General communications
Indigenous peoples
Education
Adequate housing
Food
Violence against svomen
Human Rights Defender
SOOTH ASIA HI MAN RIGHTS D<I<TIMHNTAT1< IN CENTRE ISA HRDCI TRAINING SERIES
AUOIIST 3<«M
Can either lie:
May undertake visits (upon invitation)
Report to the CHR
SOtmi ASIA HUMAN RIGHTS DOCUMENTATION
IN Cl
CENTRE (SAHRlX'l TKAIMNO SERIES
AIRIUHTMoa
Communications to special procedures
Output of special procedures
Special procedure* report* may:
You must:
consider developments in type* of violutions
identify the victim
through practice, extend the scope of the mandate
condemn governments
identify the suspected perpetrator
make recommendations for legislative amendments, the release of individual*
identify your organisation (you may request anonymity)
annex individual cases
provide a detailed account of the violation
he concise, factual and exact
St HTH ASIA HUMAN RIGHTS D< X-T'MENTATH «N CENTRE (SAHRDCl TRAINING SERIES
AUGUST 2004
S< >1 TH ASIA HUMAN RIGHTS DOCl MENTATION CENTRE (SAHRDCi TRAINING SERIES
AIKM 1ST 30tM
Information to include
Best practices
Provide as much relevant information as possible on:
Pros ide as much information as possible
Avoid political rhetoric and exaggeration
the victims (e.g., name; age; sex; nationality; identity number, if any; place of
Provide details of witnesses, where appropriate
birth; occupation; address)
Indicate local remedies attempted
the events (including dates, times, place of events)
Pros ide attachments! Local laws, medical certificates, police reports...
relevant attachments
St HTH ASIA HUMAN RIGHTS D« ■ 'I JMENTATR «N CENTRE tSAHRDO TRAINING SERIES
AIXJUST2OO4
St >1 TH ASIA HI MAN RIGHTS th w 1 JXfKNTATK iN CENTRE ISAHRIX3 TRAINING SERIES
AUGUST 2004
Limitations of special procedures
Urgent action appeals
Be .selective
Poorly resourced: do not over burden them
Provide as much reliable information as is readily available
Can not compel governments: only apply pressure
Intlk'iitv ItxjiI remedies
Can often not get invited to countries for vinitji
Provide follow up information
•
Succexx of the special procedure often depends on the mandate holder
Be realistic about what the .special procedures can do
lATK «N < "ENTRE (SAHRDCl TRAINING SERIES
St >1JTII ASIA HlJMAN KKIHTM IX X'IJMENT/
11ST 2«»4
AtKM
St H TH ASIA HI MAN RIGHTS Ih fCIMENTATH »N t.T-NTRfi iSAHRDC) TRAINING SERIES
AUGUST 2O»M
Strengths of special procedures
•
Remember...
Your govumment need not have ratified any treaties to be scrutinised by the
special procedures
•
Use your domestic remedies first
Can work svith them svithout ever leaving your office: do not need to go to Geneva
Be selective in your dealings with the special procedures: don't overburden them
Do not require your organisation to have ECOSOC accreditation
Chose the special procedure you use carefully
Reliant on NGO contributions
Be thorough and professional in your work
Anyone can contact them
*
SIIimi ASIA HUMAN RIUHTS D« h-UMENTATH 'N CENTRE(SAHRDCl TRAINING SERIES
AUUIIST BSU
Keep in mind that some situations svon't he covered by the special procedures
SIIITH ASIA HUMAN RIGHTS thCHMHNTATION CENTRE iSAHRHCl TRAINING SERIES
AUGUST 2I«M
Position: 1273 (4 views)