Richard Wilkinson shows how good health is linked to self-esteem and a person's perceived place in the social hierarchy

If toxic materials or infected food were responsible for as many deaths as social inequality, immediate action would be demanded. The link between absolute poverty and poor health is well established. But health research in affluent societies shows that relative deprivation is also highly significant. Research on death rates tells us about individual physical factors such as diet, air pollution or housing. But it also points to a broader picture, involving the entire social environment and its impact on human health.

The "Whitehall" studies of civil servants show that age for age, death rates are three times as high among the most junior as among the most senior office staff working in the same government offices. Such health inequalities run throughout the country, occurring at most ages and involving most causes of death. Rather than distinguishing merely between rich and poor, they constitute a social gradient in health so that even people near the top of society have slightly worse health than those immediately above them.

Ten or 15 years ago it seemed possible that health inequalities were primarily the result of social mobility moving the healthy up the social scale and the unhealthy down. Yet we now know that the overall effect of social mobility is to make the health differences smaller than they would be if people lived in the same social classes all their lives.

É«ºÐÖ±²¥

A combination of behavioural factors - smoking, diet and exercise - and exposure to different material hazards such as poor housing, air pollution or dangerous working conditions once seemed the most likely explanation. But the emerging conclusion is that these explain less than had been expected.

There are insignificant differences in fat consumption between different social classes. Inequalities in deaths from smoking-related diseases are not much bigger than from diseases unrelated to smoking. It is difficult to see how things like poor housing could cause increased rates of cardiovascular disease or cancer.

É«ºÐÖ±²¥

Instead findings from a variety of perspectives converge on the social environment as the culprit. Population studies repeatedly found that psychosocial factors like having no control over your work, anger, material insecurity, lack of social support, low participation in community life, low self-esteem and stressful "life events", all seem bad for health. Factory closure studies showed that health began to worsen when redundancies were first announced - often long before people became unemployed. The health effects of job insecurity illustrate the importance of psychosocial factors leading to anxiety.

Biologists have made progress in identifying the physiological pathways through which chronic stress affects hormonal and metabolic processes, including the immune system. This is linked to work on the health effects of social hierarchy among non-human primates. Studies of monkeys show that low status animals suffer the health effects of chronic stress. Not only do the lower status animals show more behavioural signs of stress, but most of the physiological differences are symptomatic of chronic stress. The studies ruled out other explanations: high and low status animals were given the same diet and kept in the same physical environment, and social status was experimentally manipulated. Several risk factors have been found to have strikingly similar social distributions in human populations.

Other evidence suggests that within developed countries, health relates less to absolute income and material living standards than to relative income. What matters is how your income compares with other people's. Instead of health being affected directly by material circumstances, it seems that the social implications of your position in society is the key.

The healthiest societies are not the richest developed nations, but the most egalitarian. Average life expectancy is two or three years longer in countries like Japan and Sweden, where income differences are much smaller, than in the United States or Britain. A developed country can be twice as rich as another without having much impact on health.

But the relationship between the extent of income inequality and death rates is close in most societies. The main reason for this is almost certainly that health standards in countries with bigger income differences are weighed down by a bigger burden of poor health linked to more widespread relative deprivation. Although the people of Harlem in New York are better off in absolute terms than the poor in some other developed societies, at most ages their death rates are higher than in a country as poor as Bangladesh. Death rates from drugs and violence are dramatically raised, but the biggest single cause of excess mortality in Harlem is heart disease.

É«ºÐÖ±²¥

It is likely that income inequality affects health in lots of different ways. Apart from the effects of unemployment and relative poverty, which are subsumed by measures of income inequality, there is strong evidence that less egalitarian societies are less cohesive. This in turn appears to contribute to health damage. Some clues can be found by looking at the main categories into which deaths are divided - cancers, cardiovascular diseases, infections, etc. Although they all tend to be more common in less egalitarian societies, the causes of death most closely related to income inequality are alcohol related deaths, accidents and violence. These "social" causes are indicative of differences in the social fabric and cohesiveness of more and less egalitarian societies.

In short, health inequalities tell us of the psychosocial damage of social hierarchy, while the effects of relative income and of income inequality show that the more hierarchical the hierarchy is, the worse the damage. It would be unreasonable to think that these effects are confined to health alone. It is hard to escape the conclusion that the results might look much the same if we were looking at crime rather than health. Research in the United States shows that violent crime and homicides in each of the 50 states are closely related to the scale of the income differences in each state. Homicide and inequality are related internationally. In Britain the record widening of income differences in the late 1980s was followed by record increases in crime. Japan has enjoyed diminishing income inequalities during most of the post-war period and a long-term decline in most categories of crime. Although white-collar crime did not decrease, it is indicative that the most rapid decreases were in crime associated with deprivation and the inner cities.

When I first became involved in research in this field I assumed that the causes of health inequalities were so deeply embedded in the basic social and economic inequalities that no government had the power to change them. But now it is clear that health inequalities change from one period to another (in Britain they seem to get wider with every decade that passes) and they are larger in some countries than in others. Within the broad framework of democratic market societies it is clearly possible to do better.

É«ºÐÖ±²¥

As research exposes the processes involved, what was once excusable government inaction becomes culpable negligence. Just as the recognition that some of the great infectious diseases of the past were water-borne led to the building of sewers and the provision of clean water supplies, so modern health research points increasingly towards a revolution in the social environment.

Economists used to argue that the big obstacle to increased equality was that it would mean sacrificing economic growth. But here too views are changing. Recent studies suggest that more equal societies have faster economic growth, leading to talk of a "new orthodoxy" arguing that equity is good for growth.

As well as the public costs of relative poverty, the corrosive effects of inequality produce an inefficient business environment. Social relations have always been central to the quality of life. Other human beings can be our most feared enemies and competitors, or our greatest source of comfort, support, love and fellowship. Which they are has always been crucial to our welfare. So human beings have lived in egalitarian societies for the vast majority of their existence. The food sharing and gift exchange practised in hunting and gathering societies served to keep social relations sweet. Evolutionary psychologists have shown that we have an evolved capacity for reciprocal altruism. Without a sense that we are contributing to the welfare of others or to some purpose beyond our narrow self interest, society lacks cohesion and it is hard to gain a sense of self-worth.

Richard Wilkinson is senior research fellow at the Trafford Centre for Medical Research, University of Sussex.

É«ºÐÖ±²¥

Unhealthy Societies: The Afflictions of Inequality, Routledge, Pounds 12.99.