The evidence, which is summarised in this editorial, shows that the circumstances in which people are born, grow, live, work and age are powerful determinants of health and of the unfair distribution of health – health inequity. Discrimination against people with mental illness shows how the relation can go the other way: for example, when people with mental ill health cannot get back into work.
Social injustice is killing on a grand scale. Put fairness at the heart of all policy-making, and health will improve and health inequalities will diminish. These bold statements were the respective judgements, based on deliberation and evidence, of the Commission on Social Determinants of Health (2008) and of the English Review of Health Inequalities (Marmot Review), which were published as Fair Society, Healthy Lives (Marmot 2010). The thrust of both reports (I chaired both, so the continuity of thought is hardly a coincidence) was the achievement of health equity through action on the social determinants of health. My colleagues and I, in the new Institute of Health Equity at University College London, are now involved in a third effort: the European Review of Social Determinants and the Health Divide (Jakab 2011).
Mental health is central to this agenda in at least three ways: mental health and mental illness are profoundly affected by the social determinants of health; psychosocial processes are important pathways by which the social environment has an impact on both physical and mental health; and discrimination against people with mental illness is an important component of health inequality. Adding to this last, there may be profound consequences of poor mental health and mental illness, including substance use disorder and personality disorder. Once mental disorder has arisen, it is associated with a range of further inequalities. These include increased health risk behaviour, reduced educational and employment outcomes, increased physical illness, and significantly reduced life expectancy, as well as discrimination. Since most mental illness arises by the time people reach their mid-20s (Royal College of Psychiatrists 2010), onset of mental disorder usually pre-dates physical illness by several decades, which means that mental disorder is an important driver as well as consequence of inequality. Examples include almost half of adult tobacco consumption in England being by those with mental disorder, almost half of smokers under the age of 17 having either emotional or conduct disorder, and adults with schizophrenia having a life expectancy that is reduced by 20 years.
Health inequalities between and within countries
In compiling evidence for the European Review (Jakab 2011), we have had to deal with the inaccurate comment that health inequalities are not an issue for a rich region like the WHO European Region. Quite wrong. Men in Russia have life expectancy of just over 60, in Iceland it is over 80 – a difference of 20 years. Among women, the spread across countries is somewhat less, although at 12 years it is still large. The rest of the 53 countries that make up the European Region are ranged in between these extremes.
Within countries, too, the inequalities are shocking. When we published the report Closing the Gap in a Generation, I highlighted the 28-year difference in male life expectancy in Glasgow between the most socially deprived and the most well-off (Commission on Social Determinants of Health 2008). In London, the spread is 17 years, from a low of 71 years in Tottenham to a high of 88 years in a part of Kensington and Chelsea. Fundamental to our thinking, however, is that health inequalities are not confined to ill health for the poor and reasonable health for everyone else. Health follows the social gradient. This was illustrated both for life expectancy and healthy life expectancy in Fair Society, Healthy Lives using neighbourhood income deprivation as a measure of socioeconomic position (Marmot 2010).
This appears to be the case, too, for mental illness. The Commission on Social Determinants of Health convened nine knowledge networks, one of which was devoted to priority public health conditions (Blass 2010). Reviewing the evidence on risk factors for depression, it concluded that the evidence was ‘very convincing’ for low socioeconomic position, low education, under-employment and unemployment, and ‘strong’ for food insecurity, gender inequity and low income.
Understanding causes and acting on them
The Marmot Review (Marmot 2010) made recommendations in six domains:
give every child the best start in life
enable all children, young people and adults to maximise their capabilities and have control over their lives – education and life-long learning
create fair employment and good work for all
ensure a healthy standard of living for all
create and develop healthy and sustainable places and communities
strengthen the role and impact of illness prevention, taking a social determinants approach.
These six summarise our understanding of the causes of health inequalities and what can be done about them. They are all, to greater or lesser extent, related to the three crucial areas I highlighted at the beginning: mental illness and mental health; psychosocial processes; and nondiscrimination against people with mental illness.
Early child development
The Commission on Social Determinants of Health emphasised the importance not only of physical development but also of cognitive/ linguistic development and social and emotional development – important both for the health of children and because they have a powerful effect on health inequalities in adulthood. Our own analyses of data from the Millennium Cohort Study (Kelly 2011) show that there is a steep social gradient on all measures of early child development – the lower the parents’ socioeconomic position, the worse the development scores of the children. An important contributor to the social gradient is parental input. As one simple example: reading to children is progressively less common the lower down the social hierarchy are the parents. Would reading to children be as effective if done by others than the parents? I am not sure that we have evidence on this. My own guess is that reading to children by caring parents combines emotional connection, cuddling and physical warmth with intellectual stimulation; all are likely to be necessary.
The lower the social position, the more common is post-natal depression. The effects on children’s development is likely to be profound.
As but one illustration of the importance of early child development for later life, analyses from the Dunedin cohort in New Zealand show that children with conduct problems at age 7–9 are subsequently more likely to experience depression, to die by suicide, to have drug problems or to be involved in crime (Friedli 2007).
Educational outcome – the percentage of young people who, at GCSE, gain at least five passes at C or above – shows a striking social gradient according to a standard measure of deprivation. Education is, of course, important, because of its close link with adult health inequalities. The key driver of this social gradient in education is the quality of early child development, which is related to family factors, degree of social deprivation and quality of schools. As a task group for the Marmot Review concluded: if you want to reduce inequalities in early child development and education, you have to reduce inequalities in society (Marmot 2010).
Employment and working conditions
Both employment and the nature of work are important. With the economic downturn, youth unemployment is now a major problem. Research from the 1980s’ downturn showed that unemployment led to a 20% increased mortality above the level of those in the same social class who remained employed (Moser 1987). Part of this excess can be attributed to drop in income among the unemployed and subsequent material hardship, but we know that unemployment is associated with mental illness and damaging blows to self-esteem – as is job insecurity. Worse, the concern from the 1980s’ downturn is that young people who left school only to find themselves out of the labour market never found their way into it.
A consequence of young people (18–24 years of age) not being in employment, education or training may be the prospect not only of worse health in the future, but other societal problems. The urban riots in the summer of 2011 began in Tottenham, north London. As indicated above, Tottenham Green has the lowest male life expectancy in London, 17 years shorter than the best. I do not think that poor health causes riots or that riots cause poor health. But one study of 1000 young people arrested for rioting showed that 91.4% were not in employment, education or training (Marmot 2011). Education, training and employment will be routes to better health and, possibly, less civil disorder.
Having a job matters, but so does the quality of work. For example, in the Whitehall II Study of British civil servants, we examined the effect on depressive symptoms of the psychosocial work environment (Head 2007). We showed that a combination of low control, high demand, low support at work, effort–reward imbalance and low organisational justice accounted for a third of the depressive symptoms occurring in this employed population. The clear implication is that action to improve management of the workplace is an important intervention.
Minimum income for healthy living
Drawing on the work of the late Jerry Morris, we said that having enough money for a healthy life is more than food and shelter: it includes enough to lead a life of dignity and to take one’s place in society (Marmot Review Team 2011). If an older person has insufficient income to buy presents for their grandchildren they cannot live a life of dignity. It is possible to calculate what that minimum should be for different family/household types. The tax and benefit system should be used to ensure that all in society have the minimum necessary for a healthy life.
Healthy and sustainable places
We have drawn attention to the fact that cold homes are bad for mental health (Marmot Review Team 2011), that exercising in green space is good for mental health, and that opportunities to exercise in green space are progressively fewer the lower in the social hierarchy one is (Marmot 2010). There are clear opportunities here for intervention.
In relation to prevention, our focus is not only on recognised unhealthy behaviours, but on why they are distributed socially as they are: the causes of the causes. Obesity is a cause of illness but what is the cause of the fact that, for women at least, obesity follows the social gradient – more as you go down. There is fertile ground for examining and addressing the social influences on behaviour.
Putting it together: the example of Glasgow
Above, I highlighted the dramatic health inequalities in Glasgow. Sir Harry Burns, Chief Medical Officer for Scotland, pointed me towards analyses that he and his colleagues have undertaken as a way of understanding the ‘Glasgow effect’ (Walsh 2010). They compared patterns of mortality rates in Glasgow, Liverpool and Manchester – cities with similar levels of deprivation and income inequalities. Glasgow has higher mortality rates than the other two. The four causes of death with the greatest relative excess in Glasgow are: drug-related poisonings, alcohol-related causes, suicide and external causes (violence). All are psychosocial in origin. Burns concludes that here lies the clue to Glasgow’s ill-health: low self-esteem, low control and low self-efficacy. These are themes highlighted by the Commission on Social Determinants of Health (2008).
To put it simply: to understand and act on health inequalities, we have to understand the effect that society has on mental life.
DECLARATION OF INTEREST