Published: 04/03/2004, Volume II4, No. 5895 Page 21

Much effort has been put into describing health inequalities, but there has been far less consideration of the mechanisms by which they develop or possible remedies. Ian Harvey welcomes a 'refreshing'contribution

Health Inequality An introduction to theories, concepts and methods By Mel Bartley Publisher: Polity Press.

ISBN: 0745627803£16.99

The words 'health inequalities' have enormous symbolic significance in British politics. Famously outlawed - certainly in official documents - during the 1980s and early 1990s in favour of euphemisms like 'variations in health', the concept - and its remedies - now sit, at least rhetorically, at the heart of government policy.

While there is a view on the political right that inequalities - like the poor - will always be with us, the public health commitment to describing and counteracting them is passionately held.

A great deal of effort has been devoted to pure description of the magnitude and time trends of inequalities, particularly between social classes.

There has been much less consideration of the mechanisms by which these inequalities develop, and discussion of the remedies.

Mel Bartley's enquiry into health inequalities is admirably thorough and balanced. Aimed 'at those with backgrounds in arts, humanities, social sciences, journalism and policy debate and formulation', it might be dismissed as a superficial account. But it is nothing of the kind.

What is particularly refreshing is the way in which the different implications for practical action of accepting different models of causation are highlighted.

Depending on which model or models is/are adopted could lead policy makers to focus on problems in areas ranging from knowledge of harmful health behaviours to excess job 'strain', poor social support, inadequate educational opportunity or environmental conditions.

The psycho-social model which emphasises workplace hierarchies has become well known. But it is unclear whether the critical risk factor is high work demand combined with low job control or an imbalance between effort and reward - or something else.

As Mr Bartley acknowledges, the hypothesis that job strain may exert its health effects through chronically raised adrenaline and cortisol levels is largely untested in large-scale studies involving a wide social class spectrum.

It is salutary to be reminded that social class gradients for coronary heart disease and diabetes were the reverse of those seen now as recently as the early 1950s, and that mortality rates for the 55-64 year age group in social class V only fell below 1931 levels for the first time in 1991. Cross-sectional data shows that social class inequalities are, surprisingly, at least as great in Scandinavian countries, with smaller income differences and better public provision, as in southern European countries.

Bartley also quotes evidence that during the economic recession of the 1980s inequalities increased less sharply in the more comprehensive welfare states of Scandinavia than in the UK or US, suggesting some general protective effect. By contrast, while public provision increased and income differentials reduced in the UK in the 1950s and 1960s, health inequalities did not diminish.

Not only does this excellent book present such complex and sometimes contradictory evidence with clarity, it also provides a comprehensive description of various approaches to measurement of social position. There is a concise and accessible chapter dealing with issues of confounding, adjustment and standardisation and both gender and ethnic inequalities are dealt with in detail.

Ian Harvey is professor of epidemiology and public health, School of medicine, health policy and practice, University of East Anglia.

Visit HSJ's regularly updated microsite on public health and health improvement at www. healthimprovement.hsj. co. uk