Intended for healthcare professionals


Importance of monitoring health inequalities

BMJ 2013; 347 doi: (Published 05 November 2013) Cite this as: BMJ 2013;347:f6576
  1. Michael Marmot, director,
  2. Peter Goldblatt, deputy director
  1. 1Institute of Health Equity, Department of Epidemiology and Public Health, University College London, London WC1E 7HB, UK
  1. m.marmot{at}

In democracies, voters need to know what is going on, for better or for worse, in good times and in bad

Health inequalities have been recognised in official statistics in the United Kingdom at least since William Farr was active in the Victorian era. T H C Stevenson in 1911, with the introduction of the registrar general’s social classes, began a remarkable series of publications, continuing for a century, of “social class” inequalities in health. In 1980, the Black report summarised this evidence, gave an explanatory framework, and made recommendations.1 At the time of Black, observers from many countries scoffed at this health manifestation of the British class system. It was then perhaps a little shocking to some that—in the “egalitarian” United States, Australia, and the Nordic countries—there were socioeconomic inequalities in health of the same order as those found in the UK. Subsequent research has confirmed that all countries have inequalities in health, although the magnitude varies.2 3 The UK led the world, not in having the most striking differences in health between social classes, but in having the best data and, based on this evidence, in developing policies to combat health inequalities.

Regrettably, this world leadership in systematic and routine monitoring of health inequalities is now under threat. To make savings, the Office for National Statistics (ONS) is considering cutting back on the data it collects and the analyses that it produces. ONS data and analyses were absolutely central to the evidence we brought together, and the recommendations we made, in our report of health inequalities in England, Fair Society, Healthy Lives (Marmot review).4 For example, figure 1 in the review plotted life expectancy and disability-free life expectancy against measures of deprivation for neighbourhoods (technically, middle layer super output areas). The remarkable social gradient in health that was shown—the higher the socioeconomic level of the neighbourhood the more years of life and the more healthy years of life people could expect—laid the intellectual basis for everything that followed. The fact that these data will no longer be available undermines a key recommendation that we made: the importance of monitoring health inequalities and their determinants.

Among other cuts that the ONS is proposing will be ones that make it more difficult to monitor child health. For example, our understanding of infant mortality by ethnicity, class, or gestational age would be curtailed. It would be more difficult to monitor excess winter deaths and the impact of fuel poverty; houses in poverty (by middle layer super output area); aspects of lifestyle relevant to health; and various measures of health system performance.

One of the things the global financial crisis has done is change the way we think about money. For ordinary people, concern about the price of carrots is occasionally swamped by contemplating buying a capital asset such as a house, the price of which is orders of magnitude greater. With the financial crisis, blithely, we discuss borrowing tens of billions. In the context of a structural deficit of more than £100bn (€117bn; $161bn) a year, is the saving achieved by these cuts to ONS data and analyses worth the damage done?

The damage would be considerable. A central feature of an advanced democracy is that we know what is going on, for better or for worse, in good times and in bad. The UK can be proud of its record in relying on data and evidence in all spheres of public life. We love to complain that politicians sometimes seem to act despite the evidence, to distort the evidence, or even to doctor it, but the evidence matters. Evidence informs the debate—it tells us whether things are going in the right direction, and to what extent.

On the determinants of health and health inequalities, monitoring has to have a central focus. Documentation of the scale of health inequalities is vital, regardless of which government is in power. Under the last government we were involved in monitoring health inequalities. The report that health inequalities had not narrowed 10 years after the Acheson inquiry,5 which had been commissioned by the government,5 was one of the drivers in the decision to set up a new review of health inequalities.4

The present government, in the Health and Social Care Act, has given the secretary of state for health the statutory duty to have regard to the need to reduce health inequalities. This statutory duty sends a good signal right through the health and social care system. We must continue to have a robust monitoring structure, otherwise we may find ourselves unable to assess the effect of government policies on health inequalities over time.

An alternative option might be to move the responsibility of monitoring from the ONS to another government organisation, such as Public Health England. However, by transferring control of one of the key indicators of social justice in this country from the nation’s independent producer of official statistics to the agency with prime responsibility for implementing public health policies, an important element of independent scrutiny would be lost. To ensure that the reduction of health inequalities continues to be subject to independent scrutiny, it is vital that monitoring continues to rest within the ONS.

Among the freedoms worth going to the barricade to defend, continued production of disability-free life expectancy by middle layer super output area, classified by index of multiple deprivation, may not be the most eye catching. Similarly, cessation of the series on trends in social inequality in life expectancy will upset only some of us. That said, achieving better health for the whole population by taking the reduction of health inequalities seriously is a banner worth rallying behind. Monitoring of the health of the population, its determinants, and its distribution is central to that societal aim.


Cite this as: BMJ 2013;347:f6576


  • Competing interests: We have read and understood the BMJ Group policy on declaration of interests and declare the following interests: PG used to work at the Office for National Statistics (ONS). We both have made extensive use of the excellent data and analyses that the ONS produces.

  • Provenance and peer review: Commissioned; not externally peer reviewed.


View Abstract