Close window to return

Issue 1, January 1998, pp.6-7

Area inequalities in health
Sally Macintyre

Area influence identified
In the mid-19th century, it was obvious to social reformers and public health physicians that place influenced health. Poor people lived in areas which were characterised by threats to health, and richer people in more salubrious places, with inevitable consequences for rates of disease.
In reports on the public health from this period, it was clear that differences between places were not simply because different types of people lived there. For example, Figure 1 shows differences between different types of towns and cities in the average age of death of three 'social orders': gentry and professionals, farmers and tradesmen, and labourers and artisans. The rank order by place of age at death was not the same for the different groups; the top group did best in Bath, the bottom did best in Rutland, and while the top and bottom groups did worse in Liverpool, the middle group did worst in Manchester. This suggested that both place and socio-economic position may influence health.

Figure 1
Age at death among different social orders, by district(1)

The response of social reformers in the 19th century was to tackle threats to health in the immediate environment by dealing with sewage, providing clean water, legislating against the sale of unsound foodstuffs and building better housing for the working classes.

With the decline in infectious diseases and the disappearance of such obvious external threats to health as contaminated food and water or overcrowded housing, public health interest in the role of place tended to decline in many industrialised countries. Since the Second World War, much public health activity has been focused on chronic diseases such as heart disease, respiratory disease and cancer, and on individual behaviours such as smoking, diet and exercise which contribute to such diseases. Rather less attention has been directed to the role of the local physical and social environment in generating inequalities in health.

Area as region, city, house
However, the observation that both place and socio-economic position predict life expectancy is as relevant at the end of the twentieth century as it was in the 1840s. There are marked and widening differences in health and longevity between regions, and between cities and neighbourhoods within regions in Britain.

Region. There was a strong North/South gradient in mortality in Britain in the nineteenth century, with death rates (standardised for age and sex) higher in the North and lower in the South.
This gradient still persists and appears to be getting steeper. During the period 1950 to 1985, death rates in Scotland were just over 10% higher than in England and Wales; by the early 1990s, they were 23% higher. Relatively low death rates are found in both the South East of England (e.g. Essex, Kent, and Surrey) and in the South West (e.g. Gloucestershire, Wiltshire and Somerset), whereas relatively high rates are found in the North West and North East of England (e.g. Liverpool, Manchester, Durham, Newcastle) and in Scotland, especially the West (Glasgow, Lanark). To give some sense of the scale of these differences, Daniel Dorling has reported that in the early 1990s a resident of Glasgow was 31% more likely to die than someone of the same age and sex living in Bristol, and 66% more likely to die than a resident of rural Dorset.(2)

City. Even within regions and cities, some neighbourhoods have illness and death rates which are substantially higher or lower than others. Variations within cities can occur between adjacent areas (for example, in Glasgow death rates in Drumchapel are around two and a half times higher than those in neighbouring Bearsden). These area variations have been widening. Between 1981 and 1991, differences in death rates between different types of neighbourhood (classified by census characteristics) increased, mainly because death rates declined faster in more socially advantaged areas than in more disadvantaged areas. While in 1981 death rates in poorer areas in Scotland were 40% above the average for Scotland, by 1991 they were 60% above average.

House. Death rates are highest for people living in council housing, and whose households do not have access to a car. They are lowest among those living in owner occupied households and whose households have access to two or more cars. In the 1980s, death rates among people living in homes rented from local authorities were 34% higher among men, and 32% higher among women, than among their counterparts in owner occupied accommodation. Information about housing tenure and car ownership, obtained from the census, is often used for planning purposes by private and public sector organisations (for example, food retailers take local levels of car ownership into account when deciding where to locate supermarkets).

Why are inequalities between areas increasing?
Are the differences between areas simply due to the way different types of individuals are located in different types of areas, or are there some features of regions or areas which influence health over and above the characteristics, such as poverty, of the individuals who live there? This question is important for social and health policy, since the answer might suggest whether area inequalities are more likely to be reduced by a focus on places (for example, by improving job opportunities, housing, transport, retail provision, recreation opportunities) or by a focus on individuals (for example, by income redistribution, changes in the welfare benefits system, health education and so on).

A further important question is whether different sorts of places influence the health of all their residents in the same way, or whether some groups are more sensitive than others to health promoting or health damaging effects of areas. For example, do men and women, rich and poor, and majority and minority ethnic groups, derive similar benefits from living in 'healthier' areas, or do deprived families do better or worse healthwise if they live in leafy suburban areas or run down inner city estates? And why do characteristics such as home or car ownership, so often used in classifications of areas, predict health and longevity?

Research in the Health Variations Programme
These are some of the questions about area inequalities in health which are being addressed by projects within the Health Variations Programme. Danny Dorling's research team at Bristol University is examining whether the increases in regional and local inequalities in health which occurred in the 1980s have continued, and the extent to which housing policy and the distribution of wealth have contributed to increasing regional and local differences in health.

Heather Joshi and colleagues at City University are using three large-scale data sets containing information about individuals over time to investigate whether the locality in which people live has an effect on health over and above the effect of a person's socio-economic circumstances. They are also examining whether such 'area effects' apply to all inhabitants equally.

James Nazroo at the Policy Studies Institute is analysing the Fourth National Survey of Ethnic Minorities to assess whether ethnic variations in health are partly due to the types of places in which ethnic minorities live. Jennie Popay and a multi-disciplinary team from Salford and Lancaster Universities are studying the relationship between material circumstances, the place in which people live, and the historical and cultural characteristics of localities.

My own group at the MRC Medical Sociology Unit and Department of Urban Studies at Glasgow University is examining the role of housing quality, residential environment and use of cars in influencing physical and mental health, and exploring why housing tenure and car ownership are so consistently associated with health and longevity.

Policy implications
These projects will not only help to advance our understanding of how inequalities in health are created and maintained, but will also have important policy implications. Examining places as well as people helps to draw attention to influences on population health outside the health care sector (for example, transport, housing, education, urban and rural planning, and policing), and to the potential value of health impact assessments of proposed programmes in a variety of sectors (for example, road building programmes, planning regulations for supermarkets, or policing in public parks). They will thus provide a broader perspective on causes of, and possible remedies for, social inequalities in health.
Sally Macintyre is Director of the MRC Medical Sociology Unit, Glasgow

References:

1. Chadwick, E. (1842) Report of an Enquiry into the Sanitary Conditions of the Labouring population of Great Britain.
2. Dorling, D. (1997) Death in Britain York : Joseph Rowntree Foundation.