Inequality kills

Two books in the last decade have drawn attention to the health implications of inequality in society, The Status Syndrome1 (Michael Marmot, 2005) and The Spirit Level2 (Richard Wilkinson and Kate Pickett, 2009). Politicians have paid lip service to their revelations but the inequality has continued to grow. Michael Marmot’s new book, The Health Gap, has just been published and will be reviewed as soon as I’ve finished reading it. This review sums up the evidence so far.

In 2008, the WHO3 reported that life expectancy not only varied widely between countries (a girl in Lesotho has a life expectancy 42 years less than one in Japan) but within countries also (children born eight miles apart in the Glasgow area have 28-year differences in life expectancy). These facts come from the report of the WHO’s Commission on the Social Determinants of Health, a three-year investigation whose findings are truly shocking.

One commission member was Michael Marmot, Professor of Epidemiology and Public Health at University College, London. He has been involved in one of the longest-running studies of the health of workers, the Whitehall Studies, following civil servants’ health since 1967 (male workers) and 1985 (both male and female). This and many other studies throughout the world reveal a “social gradient” in health. We might expect the poorest sections of society to have worse health than the well-off but Prof Marmot shows in his book The Status Syndrome that the picture is rather more complex.

He shows that the social gradient in health is only partly due to absolute poverty. Also, it is only partly due to more unhealthy behaviours. His findings are the same as but predate those of the more well-known book The Spirit Level. Both conclude that it is our relative level in society that is crucial. Put simply, Marmot concludes that we all live in hierarchical societies and where you are in the hierarchy, at work or in society in general, affects your health. This is still true even after taking all other factors into account.

Now, in different societies, or in the same society at different times, different diseases are prevalent. In 19th century Britain, tuberculosis was a widespread killer: nowadays, it’s heart disease. But there is a social gradient in both. Then, 16% of gentlemen were “consumptive”, while 30% of labourers were. Tradesmen were in between at 28%. Now, heart disease affects those lower down in the hierarchy more than those in the next rank…and those in the next rank… and so on. The Whitehall studies divided civil servants into Admin, Executive, Clerical and Other. The last group, porters, doormen, drivers and so on, had the greatest risk of dying of coronary heart disease (CHD); then came the clerical staff; then the executives; and lastly the senior, administrative, grades. The lowest grade had an 80% higher chance of dying from CHD than the top grade. They also had the highest rates of smoking, blood cholesterol, blood sugar and raised blood pressure, all consequences of unhealthy eating. But, crucially for Prof Marmot’s argument, the health gradient persisted even after these were taken into account, the lowest grades still having a 50% higher chance of death from CHD.

The same pattern is found in the whole population: life expectancy increases steadily from Class V (lowest) to Class I. Of course, as health treatment has improved and people have become more health-conscious, life expectancy has increased. For men in England and Wales, Class V life expectancy increased from 65.5 to just over 68 in 20 years (between 1972-6 and 1992-6): for Class I, it rose from 72 to 77.5. But, while life expectancy rose for everyone, the health gradient not only persisted but even got steeper over a period that included Thatcher and her attacks on unions and workers’ rights.

There are a whole lot of social gradients which nearly all mirror the health gradient. They include education, social class of parents, job prestige, and income (The Spirit Level goes into more examples, like social mobility, violence or women’s status). Do these cause the health gradient? Or is it the other way round? Do one’s genes explain it? Marmot goes to great lengths to tease out the chain of causality.

While unequal access to health care is no doubt a factor, Marmot shows that it does not explain the health gradient. The Whitehall Studies showed that there was a social gradient in the incidence of disease (i.e. first occurrence, before health care was involved). In any case, thanks to our health service, civil servants received appropriate levels of health care. A similar social gradient was later found in women, despite the different ways that social status might be attributed.

So what explains this? For a scientist, it not enough to show a statistical relationship between two things. Does one cause the other – or is there a third factor that explains both? And the answer has implications about what, if anything, should be done. It is simplistically thought that it’s merely a matter of money – or of lifestyle choices. Marmot quotes a colleague reviewing the evidence for the social gradient in health as coming from two studies of free-living primates: British civil servants and baboons in the Serengeti! As Marmot points out, baboons don’t have money and neither do low-ranking baboons smoke, consume junk food, or fail to attend doctor’s appointments. But like civil servants, they do have a social gradient in health which mirrors their position in the hierarchy of baboon society.

Does this mean that, since hierarchies naturally form in primate societies, the health gradient is a fixed fact, determined by evolution? No, says Marmot. Hierarchies are a fact but “what it means to be high and low in a hierarchy varies”. Biology and society are interacting and can interact differently in different situations. In money-based societies, it is the relative income that reflects one’s status. This shows in self-report of happiness, with richer societies such as USA reporting no change in happiness in a period where the economy grew by 50%.

If it is not absolute income that determines happiness, what does? Or, more to the point, what is it about one’s position in a hierarchy that ultimately gives rise to the health gradient. Marmot identifies the degree of control one has over one’s life circumstances as the key factor, with stress as its inverse. There are five aspects to this – “control, predictability, degree of support, threat to status, and presence of outlets” – which can modulate the effects of a psychological threat. Marmot illustrates this with a hypothetical low-paid working couple, already poor, one of whom loses his job, becomes depressed and starts drinking, instead of helping to look after the children. The other’s job is also under threat as the factory she works in is being undercut by cheaper imports. Both face loss of status if they end up on benefit. Outlets to relieve the stress would cost money they no longer have. The boss, in this example a caring individual who does not want to lay off staff, may have to move production off-shore. He however has his community work which enhances his status, his support from a wife with a professional job, and his outlet of golf.

Can lack of control or power over life cause illness? Marmot gives diverse examples to show that it can. In 1981, a major earthquake struck Athens. The death rate from heart attacks shot up by 50% over three days. In 1991, in the Gulf War, Iraq bombarded Tel Aviv with Scud missiles. During the first week, heart attacks increased significantly. In 1996, 60% of the Dutch population saw on TV their team lose on penalties to the French in the European football championships. The rate of deaths from heart attacks and stroke on that day rose by 50% in men (but not in women).

Marmot as a scientist is not satisfied with the vague diagnosis of “stress” causing these excess deaths or the other health problems linked with status. He wants to understand how external factors (“control, predictability, degree of support, threat to status, and presence of outlets”) “get into” the body and cause illness. Here he refers to Robert Sapolsky’s studies of stress in animals on the East African savannah. Sapolsky describes a lion chasing a zebra, a life-and-death situation for both. Energy release must be maximised so that the muscles can work best. The sympathetic nervous system is activated and hormones released to increase heart rate, blood pressure and blood glucose levels. Unnecessary activities, energy storage, digestion (hence a dry mouth), growth, reproductive functions, inflammatory responses, tissue repair and immune response, are postponed.

Referred to as the “fight or flight” response, these changes involve release of the hormones adrenalin and cortisol. Adrenalin increases heart rate and breathing rate, and causes release of glucose from the liver and in the muscles. Cortisol suppresses the immune system, increases glucose levels in the blood, and suppresses inflammatory responses. This is appropriate for an immediate threat but, maintained over a long time, has the following effects. Insulin’s action is inhibited, risking diabetes; fat is deposited round the waist, rather than the hips; there are low levels of “good” cholesterol; blood pressure is high: these changes are associated with developing CHD. Marmot comments that, in people with CHD, acute stress, such as an earthquake or losing a football match, could easily trigger a fatal heart attack or stroke. To show the link, he refers to studies on rhesus monkeys, social primates that form hierarchical groups.

The monkeys were fed a diet high in saturated fats and cholesterol, known to cause atherosclerosis, narrowing or “furring up” of the coronary arteries that supply the heart muscle with blood, However, not all the monkeys developed atherosclerosis: those of lower rank were significantly more likely to get it. And this was only due to their rank as was shown by changing their groups: when top ranking monkeys were put in a group by themselves, a new hierarchy developed and those now lower down started developing atherosclerosis. Studies on baboons found that low ranking ones had higher cortisol levels and lower “good” cholesterol levels in their blood. Measurements were taken after shooting the baboons with tranquiliser darts. Marmot remarks that they decided not to “dart” civil servants but were able to show that, the lower the grade of job, the lower was the level of “good” cholesterol, the higher the blood glucose, and the more fat round the waist, changes linked to high levels of cortisol and predisposing people to CHD and diabetes.

Clearly, status affects health and the steeper the differences in status the more health is affected. Broadly, this is the same thing as inequality and, in The Spirit Level, it is pointed out that in both UK and USA inequality, as measured by the difference in incomes between richest and poorest, increased greatly in the 1980s. It is no exaggeration to say that the attacks on workers’ rights spearheaded by Thatcher and Reagan which have persisted until now are responsible for a great deal of disease and death.

Now Cameron’s government is cutting benefits and demonising claimants. Does the lesson of this need to be spelt out? Thus it would seem!

1Michael Marmot. Status Syndrome: How Your Social Standing Directly Affects Your Health. Bloomsbury, 2005.

2Richard Wilkinson and Kate Pickett. The Spirit Level: Why More Equal Societies Almost Always Do Better. Allen Lane, 2009.



Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s