Study with the several resources on Docsity
Earn points by helping other students or get them with a premium plan
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Community
Ask the community for help and clear up your study doubts
Discover the best universities in your country according to Docsity users
Free resources
Download our free guides on studying techniques, anxiety management strategies, and thesis advice from Docsity tutors
The empirical evidence surrounding health inequalities, discussing various forms of inequalities such as gender, racial, and socioeconomic disparities. The author argues that certain health inequalities, particularly those related to race and income inequality at the top of the distribution in the US, are unjust and pose risks for other domains. The document also touches upon the relationship between health inequalities and education, income, and socioeconomic status.
What you will learn
Typology: Lecture notes
1 / 26
This page cannot be seen from the preview
Don't miss anything!
What does the empirical evidence tell us about the injustice of health inequalities?
Angus Deaton Center for Health and Wellbeing Princeton University January 2011
I am grateful to Anne Case and to Dan Hausman for comments on a previous version of this paper, as well as to participants at the Brocher summer academy, July 2010, for discussion and suggestions.
What does the empirical evidence tell us about the injustice of health inequalities? ABSTRACT
Whether or not health inequalities are unjust, as well as how to address them, depends on how they are caused. I review a range of health inequalities, between men and women, between aristocrats and commoners, between blacks and whites, and between rich and poor within and between countries. I tentatively identify pathways of causality in each case, and make judgments about whether or not each inequality is unjust. Health inequalities that come from medical innovation are among the most benign. I emphasize the importance of early life inequalities, and of trying to moderate the link between parental and child circumstances. I argue that racial inequalities in health in the US are unjust and add to injustices in other domains. The vast inequalities in health between rich and poor countries are arguably neither just nor unjust, nor are they easily addressable. I argue that there are grounds to be concerned about the rapid expansion in inequality at the very top of the income distribution in the US; this is not only an injustice in itself, but it poses a risk ofspawning other injustices, in education, in health, and in governance.
2. The birth of the gradient It is sometimes supposed that the gradient has always been with us, that rich people have always lived healthier and longer lives than poor people. That this supposition is generally false is vividly shown by Harris (2004, Figure 2) who compares the life expectancies at birth of the general population in England with that of ducal families. From the middle of the 16 th^ to the middle of the 19 th century, there was little obvious trend in general life expectancy. For the ducal families up to 1750, life expectancy was no higher than, and sometimes lower than, the life expectancy of the general population. However, during the century after 1750, the life prospects of the aristocrats pulled away from those of the general population, and by 1850–74, they had an advantage of about 20 years. After 1850, the modern increase in life expectancy became established in the general population. Johansson (2009) tells a similar story for the British royals compared to the general population, though the royals began with an even lower life expectancy at birth.
Kings, queens and dukes were always richer and more powerful than the population at large, and would surely have liked to use their money and power to lengthen their lives, but before 1750 they had no effective way of doing so. Why did that change? While we have no way of being sure, the best guess is that, perhaps starting as early as the 16 th^ century, but accumulating over time, there was a series of practical improvements and innovations in health, including inoculation against smallpox (not vaccination, which still lay many years in the future,) Razzell (1977), professional (male) midwives, cinchona bark against
malaria, “Holy wood” against syphilis, oranges against scurvy, and ipecacuanha against diarrhea, Johansson (2009). Many of these innovations were expensive; indeed, in sharp contrast to what would happen later, the “miracle” drugs and methods were imports from afar, Brazil, Peru, China, and Turkey. The children of the royal family were the first to be inoculated against smallpox (after a pilot experiment on condemned prisoners), and Johansson notes that “medical expert- ise was highly priced, and many of the procedures prescribed were unaffordable even to the town-dwelling middle-income families in environments that exposed them to endemic and epidemic disease.” So the new knowledge and practices were adopted first by the better-off—just as today where it was the better-off and better-educated who first gave up smoking and adopted breast cancer screening. Later, these first innovations became cheaper, and together with other gifts of the Enlightenment, the beginnings of city planning and improvement, the beginnings of public health campaigns (e.g. against gin), and the first public hospitals and dispensaries, Porter (2000), they contributed to the more general increase in life chances that began to be visible from the middle of the 19 th^ century.
Why is this important? The absence of a gradient before 1750 shows that there is no general health benefit from status in and of itself , and that power and money are useless against the force of mortality without weapons to fight. Link and Phelan’s (1995) “fundamental causes” hypothesis, that power and money seek out health improvements, but that these take different forms in different eras, is an important insight and frequently useful for thinking about changing patterns of disease. It also implies that there are periods when there may be
biologically determined and are not amenable to human action, so that they are neither just nor unjust. But biology cannot be the whole explanation. The female advantage in life expectancy in the US is now smaller than for many years, 5. years in 2008 compared with 7.8 years in 1979, and it has been argued that there was little or no differential in the preindustrial world, Vallin (1991). The contemporary decline in female advantage is largely driven by cigarette smoking, Pampel (2002); women were slower to start smoking than men, and have been slower to quit, so that the decline in associated mortality started much earlier for men. In some parts of Europe, female mortality rates from lung cancer are still rising. It might be argued that these gendered choices have no implications for overall well-being and are no more an injustice than it is unjust for women to choose Jane Austen over Dan Brown, while men choose the opposite. Yet these choices are not made in a social vacuum, nor without the constraints of economic or other circumstance, and injustice may (or may not) lie in these background arrangements. For much of the 20 th^ century, women were unjustly prohibited from smoking, and current outcomes are in part a reflection of that history. Yet that historical injustice of opportunity seems less important than other injustices, such as those of poverty and inequality, so that gender differences in smoking related mortality are surely of less ethical concern than differences in smoking related mortality between poor and rich.
It is hard to see health inequalities between men and women as a justification for differential treatment at the point of care, which would create a procedural injustice. Yet, at a systemic or research level, policymakers and administrators
constantly prioritize one set of conditions over another so that, among other considerations, one might argue that the injustice of men’s shorter lives calls for greater attention to diseases that are more likely to kill men. Alternatively, one might also argue that the male disadvantages in life expectancy reflect informed male choices of life style, and are fully compensated by the offsetting benefits of those choices. We might also argue for a broader view, that the inequalities that should concern us are those in overall well-being, not in its components. If men are favored in most domains of well-being, such as power, earnings opportunity, or morbidity, the superior mortality experience of women might actually reduce overall inequalities. On this argument, we would be much more concerned if women had higher mortality than men, just as we are particularly concerned about the higher mortality of blacks given that they suffer from a wide range of other disadvantages.
4. Children, race, and health care
Children have worse health outcomes when their parents have less income or less education. The differences are relatively small at birth, but widen throughout childhood; Case, Lubotsky and Paxson (2002) show that the income (but not the education) gradient of child health in the US steadily steepens with age. Similar results have been found for Canada and for the UK, Currie and Stabile (2003), Case, Lee, and Paxson (2008).
What seems to happen is that the disadvantages at birth from a wide range of conditions (income, housing, nutrition, health care) widen with age because of the
These childhood gradients, with their long reach into adulthood, are unjust inequalities that ought to be addressed. Sen (2002) writes “What is particularly serious as an injustice is the lack of opportunity that some may have to achieve good health because of inadequate social arrangements.” Heckman’s work, with its emphasis on dynamic complementarities through which investments in health and education have higher returns for better educated and healthier people, suggests high rates of return to interventions in early life, see Conti, Heckman, and Urzua (2010), so this is a case where justice and economic expediency are well-aligned.
Racial inequalities in the US are my second example of an unjust inequality although, once again, there is controversy about the cause of the inequalities, and the nature of the injustice. In 2006, life expectancy at birth was 4.1 years less for African Americans than for white Americans. There are also pronounced racial differences in treatment patterns, for example for cardiovascular disease, Smedley, Stith, Nelson (2002), or knee arthoplasty, Skinner, Weinstein, Sporer, and Wenneberg (2003). The conventional explanation for these inequalities, endorsed by a 2002 report of the National Academies of Sciences, Smedley, Stith, and Nelson (2002), is that the encounter between healthcare providers and patients leads to poorer treatment of African Americans by largely white providers. More generally, the daily stress of living in a racist society is itself thought to be a cause of poor health outcomes.
There is no doubt something to these accounts, but there is another, perhaps more obvious explanation, which is that African-Americans receive worse health
care because the hospitals and clinics that serve them are of lower quality than the hospitals and clinics attended by other Americans. Hospitals in the US are run on something close to an apartheid basis, with few white patients in the hospitals that treat mostly African-Americans, and vice versa , Skinner, Chandra, Staiger, Lee, and McClellan (2005); doctors and nurses are much less segregated, with many white doctors in “black” hospitals, and African-American doctors in “white” hospitals. The “black” hospitals have worse outcomes, are less well-provisioned, their pharmacies have fewer drugs, and their providers are less well-qualified, Bach, Pham, Schrag, Tate and Hargreaves (2004). In consequence, people who live in cities with large African American populations—both African-Americans and whites who live in those areas—have poorer healthcare and higher mortality rates than those who live in cities with small African American populations, Deaton and Lubotsky (2003).
Another disadvantage for African-Americans is that they are more likely to live close to environmental hazards. Currie (2011) has recently documented the claims of the environmental justice movement using data on 11.4 million births in five large American states. According to her calculations, 61 (67) percent of black mothers (without high school education), but only 41 percent of white mothers give birth within 2,000 meters of a site included in the US Environmental Protection Agency’s “Toxic Release Inventory.” There is good evidence that pollution from such sites can compromise health at birth.
These explanations for racial inequalities, like the explanation for early life inequalities, while recognizing multiple determinants, put more emphasis on
cardiovascular mortality are similar in different countries, much more similar than would be expected from different national patterns in economic and social environments, but exactly as would be expected from the spread of knowledge, drugs, and technology from one country to another, especially among rich countries where there are few barriers to adoption and implementation. Given the importance of these advances for mortality decline, and given that not everyone gets access at the same time—better hospitals adopt new advances more rapidly, and the use of drugs such as anti-hypertensives or preventive screening are more rapidly adopted by the more educated—it would indeed be surprising if the new innovations did not widen the gradients within countries, just as was the case for the first gradients in Britain in the eighteenth century. And the same argument applies here as there, that while we should like to reduce the inequalities, we must be careful that our policies speed up the widespread adoption of beneficial treatments and do not discourage their introduction or the discovery of new treatments, and thus kill the innovative goose that is laying the golden eggs. Wealth has a formidable record of generating new ways of improving health, and we need to harness its power, not muzzle it on the grounds that it generates temporary inequalities in health.
5. Socioeconomic status, education, income, and health
Much of the epidemiological and sociological literature describes and analyzes health inequalities in terms of differences in socioeconomic status, which is taken to be set by some amalgam of income, education, rank, and occupation, among
other things; indeed health inequalities are often described as “social” inequal- ities in health. The concept of socioeconomic status, although useful as a descriptive, portmanteau term, is unhelpful when we come to think about causation, and beyond causation, about policy. For example, there is much evidence, reviewed for example in Cutler, Lleras-Muney and Vogl (2010), that education directly promotes health, not just that those who are educated are also likely to be healthy because of some third factor, but because the things that are learned in school and college enable people to take better care of themselves as well as to take greater advantage of the healthcare system when they need it. We also know that being sick adversely affects the ability to earn—that is what the word “disabled” means.
In consequence, much of the evidence is consistent with the simple account in the Figure that focuses on health and education in childhood, with child health affecting both education and adult health, with education a major determinant of adult earnings which may also be limited by ill-health. Sick people earn less, they spend less time in the labor force, and they retire earlier. In this framework, little but confusion is generated by amalgamating a cause (education) and a consequence (income) into a single category labeled socioeconomic status
more difficult, but it is likely to be ineffective because it is based on a largely mistaken diagnosis of the problem.
6. Unhealthy behavior by the poor
One of the major causes of health inequalities is differences in behavior across income and educational groups; in rich countries, poorer people are more likely to smoke, are more likely to be obese (at least among women), are less likely to exercise regularly, are more likely to work in jobs that pose a risk of injury or disability (physical labor in a modern economy is more likely to be bad than good for health—think of a delivery driver carrying heavy packages), and are more likely to drink alcohol immoderately. While there is an element of choice in occupation and lifestyle, poor people lead heavily constrained lives, in terms of money, time, and choices, and some of these choices, even with their poor health consequ- ences, may not be easily avoided under the circumstances. For example, without human capital from education, or financial capital from inheritances, people must often rely on their physical capacities and energies, and adopt occupations, as well as consumption styles, that involve heavy wear and tear on their bodies and on their health, Muurinen and Le Grand (1985), Case and Deaton (2005).
The health inequalities that come about through these life-style and occupational “choices” are once again addressable, if only in part, by addressing early life inequalities in health and in education, the same prescription that runs throughout this essay.
7. International health inequalities
The differences in life expectancy between countries dwarf those between differ- ent groups within countries. (This is true for income inequalities too.) There is an eight year difference in life expectancy between Japanese women (86.1 years) and Japanese men (78.0 years), but both Japanese men and women can expect to live almost twice as long as a newborn in the lowest life expectancy countries in sub- Saharan Africa (Zambia, Angola, and Swaziland.) Infant mortality rates—which are the main drivers of differences in life expectancy between rich and poor countries—vary from three per thousand in Iceland and Singapore (who says the tropics must be unhealthy?) to more than 150 per thousand in Sierra Leone, Afghanistan, and Angola. In 1990, more than a quarter of children in Mali did not live to see their fifth birthdays, a marked improvement over 1960 when around half died in childhood—or put even more starkly, when median life expectancy at birth was only five years.
The children who die in poor countries would not have died had they been born in rich countries. The same is true of adults with AIDS, whose life expectancy is greatly prolonged in rich countries by the routine use of anti-retroviral drugs that are far from universally available elsewhere. At a medical level, we also know how to prevent the death of children in poor countries. They are not dying of exotic, tropical diseases for which there are no medicines, but from respiratory infect- ions, from diarrhea, from diseases associated with malnutrition, from lack of neonatal care, or from diseases like polio or measles, for all of which there are known, cheap, cures or preventions. Children in rich countries do not die of these
in those cases where first-world medical care and outcomes are found side by side with some of the world’s worst health conditions. But the remedy for this injustice falls, not on the international community, but on domestic governments, which sometimes seem to have little interest in or ability to address it. Of course, these arguments do not absolve rich people from an obligation to assist those who are suffering, to the extent that it is feasible for them to do so.
Whether or not we adopt a cosmopolitan position, it is clear that neither the international organizations, private NGOs, nor the governments of rich countries have more than a very limited ability to correct international health inequalities, so that the practicalities are against the cosmopolitans. This is not a matter of the citizens of the rich world being unwilling to pay the (relatively low) financial costs of the required vaccines, medicines, and health clinics. International health inequalities cannot be eliminated without the construction of well-functioning domestic healthcare systems that provide to the citizens of poor countries the preventative, pre- and post-natal and maternal care that is routine in rich countries. These systems cannot be constructed from the outside, but require domestic state capacity, institutions, and responsibility to citizens that is often missing in poor countries, and that may well actually be undermined by large financial flows from outside, Moss, Pettersson and van de Walle (2008), Epstein (2008). If this undermining is important, as I believe it is, there is a risk that a well-meaning cosmopolitan attempt to address international health inequalities might actually make them worse and cause even greater injustice and suffering.
None of this is to deny that much has already been done to improve health in poor countries by the application of first world knowledge (the germ theory of disease) and techniques (vaccination, smallpox eradication). Nor that rich countries cannot do more through basic research (e.g. that AIDS is a sexually transmitted disease, or the development of ARV drugs) or international legal arrangements (e.g. trade rules governing the international prices for vaccines and medicines.) Yet the leading sources of mortality in poor countries, especially among children—respiratory infections, diarrheal disease, lack of vaccinations among hard to reach populations—are not addressable by “vertical” health campaigns run by or with the assistance of international organizations.
8. Income inequality as a risk factor for health
The health inequalities literature frequently argues that differences in incomes cause health differences, a position that I have argued is largely mistaken. A related but different view is that differences in income are themselves a risk factor for the level of health (as well as for the levels of other good social outcomes), so that the rich as well as the poor are hurt by large income differences, Wilkinson (1986), Wilkinson and Pickett (2010). In effect, income inequality is a form of social pollution which, like actual particulate or chemical pollution, risks the health of everyone, rich and poor alike. That income inequality should be a risk factor is sometimes referred to as the “relative income hypothesis,” but this is a misnomer because it is possible for health to depend on income relative to others, on rank, or on status, without income inequality having
