Health inequality and the causes of the causes

Boyer Lecture 1: Health inequalities and the causes of the causes

Professor Sir Michael Marmot

What good does it do to treat people and send them back to the conditions that made them sick?

Variations in the conditions that make people sick are in grim display in the US city of Baltimore. Such conditions do not lead only to ill-health. In early 2015 Baltimore erupted. Civil unrest broke out. The precipitant of the riots was the killing of a black man in police custody. Or should I say one more killing of a black man by the police. But the underlying cause of the riot was inequality of social and economic conditions. As Martin Luther King said: 'A riot is the language of the unheard'.

I said Baltimore erupted, but it was one part of Baltimore, the poor inner city, that erupted. Not the rest. I had been studying health inequalities in Baltimore before there was civil unrest. In the poor part, where the riots broke out, life expectancy for men was 63 years. In the richest part it was 83 years. A twenty year gap in one city. If you live in the richest part of Baltimore and want to see what it is like to live in a place with male life expectancy 63, you could fly to Ethiopia. Easier is to travel a short way across town. Life expectancy for men in the poorest part of Baltimore is the same as Ethiopia, two years shorter than the Indian average.

A link between riots and ill-health is not unique to Baltimore. In summer 2011 in London, there were riots. They started in Tottenham in North London. Eerily, the precipitant was the killing of a black man by the police. As with Baltimore, the underlying cause was inequality. I had been pointing to figures on health variations in London. For men, life expectancy in the most down at heel part of Tottenham was 17 years shorter than in the ritziest part of Kensington and Chelsea. No surprise that the riots should have broken out in Tottenham not in Kensington and Chelsea.

Why this link between ill-health and crime? I don’t think that ill-health causes civil unrest and I don't think that riots cause ill-health – except in the obvious way that people can be injured. No, I think the social conditions in which people are born, grow, live, work and age are strongly determinant both of risk of ill-health and likelihood of engaging in civil disorder. When the Tottenham riots broke out a politician commented: this is criminality, pure and simple. To paraphrase Oscar Wilde: the riots were not very pure and the causes were not simple. It was clear, though, that social deprivation was among the causes. The Guardian newspaper reported that of 1000 rioters going through the courts fewer than 9 per cent had a job or were in training; 91 per cent were not in employment, education or training - NEET as we say in the jargon. Nationally, at that time, the NEET figure was about 10 per cent. A stark contrast: 91 per cent of rioters not in employment education or training compared with 10 per cent amongst non-rioters. There were no doctors, lawyers, accountant among the rioters, no plumbers, drivers, shop assistants among the rioters, but people who had little to lose and uncertain futures.

The link between deprivation of social conditions, ill-health and crime is all too obvious in Australia. The life expectancy gap between the Indigenous and non-Indigenous populations of Australia is about 11 years. It had been said to be 17 years. Whether it's actually got smaller or it's a change in the calculation -either way it's enormous. Australian aborigine men are six times more likely, and aboriginal women eleven times more likely, to die of ischaemic heart disease than non-Indigenous men and women. The diabetes, the differences are more alarming. The diabetes death rate is nineteen times higher in aboriginal men, and twenty-seven times higher in aboriginal women, than in the non-Indigenous population.

Incarceration rates, too, fit with the pattern of ill-health and crime clustering together. Aborigines make up 2.5 per cent of Australia's population and 25 per cent of the prison population. Some of that appalling excess will be higher crime rates, linked to deprivation, and some will be discrimination through all parts of the justice system.

I have just described two phenomena that will be the subject of these lectures. The first, dramatic inequalities in health, is illustrated by the 17 and 20 year life expectancy gaps in London, Baltimore and other cities, and the inequalities in health between Indigenous and non-Indigenous Australians. The second theme of the lectures is what I call the social determinants of health: the conditions in which people are born, grow, live, work and age. And, I must add, inequities in power, money and resources that give rise to inequities in the conditions of daily life.

Health and inequalities in health are closely linked to the conditions in which we raise our children, the education we get, the neighbourhoods in which we live, the work we do, whether we have the money to make ends meet, our social relationships, our care for the elderly. In short, all the things that matter to us day to day and in the arc of our lives influence health. And these conditions of life that matter to us are strongly influenced by the decisions that societies make and, indeed, global decisions that influence our social environment.

At this point you might be thinking, I am interested in my health, all of us are, why has he not mentioned smoking and drinking and staying fit? More, aren’t inequalities in health the result of inequalities in access to health care? Put bluntly, you might be thinking, the poor probably don’t look after themselves and when, inevitably, they get sick won’t or can’t get access to medical care.

On the first, life style, let me summarise by saying that smoking, drinking, unhealthy patterns of eating and exercising, and obesity are indeed causes of ill-health. But we have to address 'causes of the causes'. Why in Australia, as elsewhere, are smoking and obesity more common the lower one is in the social hierarchy – people with lower income and education are more likely to smoke, or be overweight than people with more income and education. The causes of the causes are the social determinants of health and they influence not only life style but stress at work and at home, the environment, housing, transport.

As for health care, inequality in access to health care is not the cause of inequality in health. It is not lack of health care that causes someone to have a heart attack or develop cancer of the cervix. We know a great deal about the causes of those conditions and why they follow the social gradient. We need to address these causes to prevent disease from occurring. But when the catastrophe happens, health care is vital. We should not add the insult of lack of care to the injury of getting sick in the first place.

I am going to make a grand claim. So close is the link between the nature of society and health, that we can tell a great deal about how well a country meets the needs of its citizens – provides the conditions for them to lead flourishing lives – by the health status of its citizens. Even had I not spent decades of my life in Australia I would know that it is a well functioning society, because life expectancy in Australia is near the top of the world rankings. I know that Australia is doing somewhat better than the USA as a society, and much better than Russia simply by examining the health statistics.

But I know something else: there are big inequalities in Australia that show up in big inequalities in health. They are most obvious between Indigenous and non-Indigenous people. But they also track along lines of education, social standing, and income in the non-Indigenous population.

My concern is not simply to document inequalities but to draw on the best evidence globally to say what we can do about them by addressing the social determinants of health. Evidence shows that we need action from government – federal, state and local – from civil society, communities and families. The welcome news is that the action I am talking about would lead to less crime, a more cohesive society, and the opportunity for all of us to live healthy flourishing lives.

The first part of my grand claim is that the magnitude of health inequalities tells us a great deal about how well a society is functioning. The second part of my grand claim is that these inequalities are unjust. They are unjust precisely because we know what to do to put them right. Hence the title of these lectures: Fair Australia – Social Justice and the Health Gap.

This all sounds rather social and political. How did I, a doctor, and now President of the World Medical Association, find myself doing this?

The trail of things began when I was a medical student at Royal Prince Alfred Hospital in Sydney. As I lay out in my book, The Health Gap, I was sitting in Psychiatry Outpatients when the next patient arrived. She looked the very picture of misery. Her gait almost apologetic, she approached the doctor and sat down, huddling into the chair. The dreariness of the outpatient clinic, unloved and uncared for, could not have helped. Certainly it did nothing for my mood.

'When were you last time completely well?' Even in Australia psychiatrists were supposed to have middle-European accents. At that time, even in Australia, psychiatrists were supposed to have middle-European accents.

'Oh doctor,' said the patient, 'my husband is drinking again and beating me, my son is back in prison, my teenage daughter is pregnant, and I cry most days, have no energy, difficulty sleeping. I feel life is not worth living.'

It was hardly surprising she was depressed. The solution? The psychiatrist told the woman to stop taking the blue pills and try these red pills and come back in one month. That’s it? No more? To incredulous medical students he explained that there was very little that he could do.

The idea that she was suffering from red pill deficiency was not compelling. It seemed startlingly obvious that her depression was related to her life circumstances. The psychiatrist might have been correct that there was little he personally could do. I thought then, and am convinced now, that we should be paying attention to the causes of her depression. The question of who 'we' should be explains why I discarded my flirtation with psychiatry and pursued a lifetime investigating the social causes of ill-health and latterly advocating for action.

More than 40 years of research and action on social determinants of health led to my publishing The Health Gap. After the launch of the book at UCL, University College London, I hopped into a black cab clutching copies of my book.

The cabbie asked: 'Did you write that book? What’s it about then, guv.'

'The poor have worse health than other people.' I told him.

He started to laugh. 'Don’t misunderstand me,' he said, 'don’t take it amiss. But did you really need to publish a book saying that. Know what I mean?'

Challenged by the wisdom of the London cabbie, I told him about my Whitehall studies of British civil servants. Civil servants include neither the richest in society – no hedge fund managers; nor the poorest – no unemployed single mothers. Yet there is a remarkable gradient in health and life expectancy: the higher the grade of employment the longer the life and the healthier the life. I label this the social gradient in health.

'People in the middle of the hierarchy,' I told the cabbie, 'have worse health than those above them and better than those lower than them. It is as true in supposedly egalitarian Australia as it is in class-bound England.'

'Hey,' he said, 'this is much more interesting than what I first thought. Know what I mean?'

I was also able to reassure him that this remarkable finding of a social gradient in health was not unique to civil servants. In Australia, for example, we see a clear gradient: the fewer the years of education the higher the risk of death. Men and women in their 40s with fewer than 12 years of education, have 70 per cent higher mortality risk than the most educated.

The gradient changes everything. My taxi driver had no difficulty understanding that poor people had poor health - even if he got the explanation wrong. But the cabbie and the rest of us find it much more challenging to understand why people in the middle should have worse health than those at the top.

How do these gradients in health come about? We can start our search for explanations by looking at the extremes. To do that, let’s return to Baltimore. Remember, men in the worst off part have life expectancy of 63; men in the best off have life expectancy of 83. I was attending a meeting at Johns Hopkins University in Baltimore and a couple of young doctors kidnapped me. 'Come on', they said, 'we’re going to show you Baltimore'.

We started in leafy Roland Park, where male life expectancy is 83. It was quite lovely, trees, lawns, generous houses. We then went to Upton Druid, with life expectancy of 63. We saw streets where every second house has a diagonal red cross on the door. The cross means that the dwelling has been condemned as unfit for human habitation. If there is an emergency, the emergency services will not go there. Can you imagine what it means to live in a street where every second house is unfit for human habitation?

We can add some detail. In 2010, median family income was $17,000 in Upton Druid; $90,000 in Roland Park. In Upton Druid, half are single parent families; 7per cent in Roland Park. In Upton Druid, 90per cent of the kids don’t go to college; in Roland Park, 75per cent complete college. Wait for the next bit. In Upton Druid, one third of young people aged 10-17 are arrested each year for a juvenile disorder. One third each year! That means that the chance of getting to age 17 without a criminal record is quite slim. In Roland Park, it's not one third each year but one in fifty.

In theory the slate is wiped clean at 18, no criminal record. In theory. You apply for a job and you are asked have you ever been in trouble with the police. I suppose you could lie, but that is hardly a good qualification for getting a job. Or you could tell the truth which, given your record, is not a good qualification for getting a job.

Then there is the American disease: guns. In Upton Druid, over a 4 year period there were 100 non-fatal shootings per 10,000 residents and 40 homicides. In Roland Park, there were no non-fatal shootings and 4 homicides, one tenth of the rate in Upton Druid.

What these comparisons make clear is that we need to look at the whole of life from early childhood to older age and the life chances that differ so dramatically between areas even within one city.

A question I get asked in Australia, the UK, the US and elsewhere flows from the London cabbie’s response, and it is: why should I worry. The poor have poor health, but I’m not poor. It doesn’t affect me. If you are of one political persuasion you might think the poor have only

themselves to blame. If of a different view you may think that poor health for the poor is unfortunate and a mark of unfairness in society. Either way, you think: thank goodness it's not me.

The implication of the gradient is that it is you and it is me that are involved. All of us below the very top. We calculated in Britain, and the figures would not be very different in Australia, that the average person can expect 8 fewer years of healthy life than they would if they had the highest social position. Eight fewer years of healthy life means earlier onset of decline in grip strength, of difficulty walking, of cognitive decline and, of course, fewer years of life. The gradient involves all of us.. We want to level up, flatten the health gradient upwards – have the health of everyone approach the good health of those at the top.

I have used the word 'poor' to describe the young men growing up in the Upton Druid area of Baltimore. A pause is in order. The median household income in that area was $17,000. Is that poor? The simple answer is that $17,000 a year is poor relative to Baltimore. It is really quite rich on a global scale. Yet it does not guarantee good health - shorter life expectancy than in India. Let’s take a contrast from the Americas: Costa Rica – its average income per person is US$13,000, but it has a considerably higher male life expectancy of 77.

If you have little money, and can’t afford the basics, health will suffer. But in the case of the poor part of Baltimore, it is not just food and shelter, individual income cannot buy family security, a secure environment, good schools with motivated fellow pupils, job opportunities. In other words, where health is concerned, it is not only how much money you have but how much you have relative to the environment in which you find yourself. It’s not so much what you have but what you can do with what you have. Costa Ricans are much poorer on average than Americans, even allowing for differences in purchasing power, but their health is equivalent to the US average, male life expectancy of 77, and much better than that of poor Americans.

This emphasises two key themes of my lectures. First, health is a better measure of social progress than is national income. Second, we should be aiming to reduce the slope of the social gradient in health by levelling up, bringing the health of everyone up to the high level at the top of the social hierarchy.

At first blush, you might have thought I want to achieve this by giving everyone the same income. That, of course, has never been done since humans stopped being hunter gatherers and adopted more complex forms of social organisation. Is it then possible to narrow the health gap? Evidence shows it is.

We compared 15 European countries and looked at life expectancy at age 25. In each country there was a gradient, the more years of education, the better the life expectancy, but there were big differences among countries. Sweden had the longest life expectancy for men at 25 and the narrowest gap between university educated and those with basic education. Estonia and Hungary, and the other former communist countries of central and eastern Europe, had the shortest average and the biggest gap.

The message is that health inequalities are not inevitable. The magnitude varies enormously. Big in Hungary, small in Sweden, variable in Australia.

What can we do? I will come to that in the lectures that follow.

Let me anticipate, though, by saying that action will require greater equity of power, money and resources. Greater equity? I have been told that I live in fantasy land. When Martin Luther King rose in Washington and declared: 'I have a dream that on the red hills of Georgia the sons of former slaves and the sons of former slave owners will be able to sit down together at the table of brotherhood'. What if instead he said 'I’ve been told I live in fantasy land. We should accept the status quo and seek incremental change'.

There would never have been a civil rights act.

I say come and join me in my fantasy land and let’s seek a more just society.