Professor Sir Michael Marmot's work with the World Health Organisation and on post-Acheson research puts him at the forefront of modern thinking on health inequalities. Andy Cowper spoke to him

Health inequalities are not simply about poverty. In his groundbreaking long-term studies of Whitehall civil servants, world-renowned epidemiologist Sir Michael Marmot proved the existence of a social gradient in health and the importance of social determinants.

Professor Marmot and colleagues found a strong association between employment grades and mortality status, starkly shown in the finding that the mortality rate for those at the lowest grades, such as messengers and doorkeepers, was three times that of those at the most senior grades.

Nowadays Sir Michael is, to put it mildly, a busy man. As well as his day job as director of the International Institute for Society and Health at University College London, he is chair of the Department of Health's scientific reference group on health inequalities, which is leading the work on the Acheson 10-year review.

He is also looking at developments in England for the World Health Organisation's Commission on the Social Determinants of Health report. Professor Marmot chairs this WHO commission, which was set up in 2005 and is due to report in September with a DH-sponsored global conference event in November.

Renewed effort

"In the UK we had the Acheson Inquiry, reporting on developments in England in 1998 which reviewed the evidence of what to do about health inequalities and made its recommendations," says Sir Michael, who was on one inquiry's scientific advisory panel.

"Then after long deliberation and finally after a Treasury-led review, we got a national health inequalities strategy [Tackling Health Inequalities: a programme for action] to which government departments were asked to commit and sign up.

"We set up a scientific reference group on health inequalities which has issued a series of status reports monitoring developments against the strategy. Health secretary Alan Johnson has called for renewed effort and asked me as WHO commission chair to lead on what we've learned from the commission: how it can be modified, interpreted and added to our UK process."

Industrial hazards

Sir Michael is clear that evidence-based policy-making is essential for health inequalities but just as sure that implementation is crucial. "There's always got to be research. In the WHO commission, we were asked for evidence to support any recommendations. Predictably, the evidence base is inadequate, which could be a recipe for doing nothing. It's inadequate because the determinants of health are so wide-ranging and complex.

"For example, I was asked: 'Are trades unions good for health?' Now, you can't do a randomised control trial on that! So you use a chain of reasoning and the best evidence you can; look at the history of workplace collective action, which led to improved working conditions - industrial hazards are certainly bad for health.

"There's a good case that unions have been in the vanguard for improving working conditions. I think we have enough evidence for action on the social determinants of health but we must always re-evaluate and research while acting."

Sir Michael has made it clear that the government has made redressing health inequalities a priority. Does he think the NHS has matched the effort?

"On the whole, yes. It's been high on the NHS agenda to address inequalities since 1948, by making care universally available. To the extent that it's not perfect, we still want improvements.

"But before the NHS, care was unavailable to many. The NHS has made a big difference and more recently has explicit targets on inequalities."

Where should NHS managers target most attention to address inequalities? Sir Michael stresses the importance of working in partnership: "We've had lots of discussion on the WHO commission on the nature of cross-sector working.

"A primary care trust may formally have nothing to do with a local education body, but education is clearly a vital forum to make differences to health and inequalities.

"Look at [government programme] Sure Start; it's vitally important for health to be represented there but early child development is not just a healthcare issue. Managers must ensure equitable access and use appropriate evidence on health prevention and promotion, but ought to be taking a leadership role to co-ordinate other sectors; that would be really doing the job properly if you want to affect inequalities."

Graded phenomenon

The crucial public health function sits in PCTs. From his international experience, does Sir Michael think PCTs are organisationally big enough and strategic enough to address inequalities with commissioning?

"I can't really comment. International evidence suggests that what happens in any health service must be in concert with addressing inequalities in other sectors, so it's not just what PCTs can do in commissioning but what they can do in conjunction with other cross-sector actions. Organisations must adapt to local circumstances, but must work in concert."

The health gradient, as Sir Michael's Whitehall studies have shown, "does not simply mean that there's worse health at the bottom of the social ladder and the rest are OK (although for most areas, lower socio-economic indices are where you find worse health and vice versa). It's a graded phenomenon, with profound policy implications - it's not sufficient to focus only on the poorest of the poor.

"You asked why services are worse in poorer areas? The implication of the social gradient is that we should not focus simply on the poorest areas but look across the social spectrum. In the Whitehall studies no-one was really poor as in workless, but there is an evident social gradient for those lower in the hierarchy having higher mortality and morbidity.

"It's true outside the civil service, if you divide and classify local areas in England and Wales according to level of deprivation into a scale of twentieths, you see a remarkable gradient - each point on the scale as you go up the scale of affluence has lower mortality than the socio-economic area below.

"So we should not just focus on the very bottom but address inequality across the whole of society."

Health gradient

What practical strategies would he recommend for addressing the social gradient?

"Policy makers need as a starting position to think how policies will affect everybody, as the people in the middle of the scale matter too. We can't only focus on the very bottom of the social gradient, hugely important as that is."

Does he see perverse incentives arising from targeting the very bottom?

"Yes. Previous targets have reflected a focus on the most deprived. It's intrinsically harder to set targets relative to the social gradient, but we should think about all new policies in relation to the gradient.

"This is the point of the new Health Inequalities: progress and next steps report. We have two key aims. First, to improve health for everybody. The Tackling Health Inequalities: 2007 status report on the programme for action published in March this year showed that life expectancy and infant mortality for the bottom group are improving (as they are for the average), so we're meeting one key goal. The worst-off are doing better.

"The second is to narrow the gap between the better-off and the worse-off.

"We haven't achieved that yet, so there's much more to do".