Elsevier

Social Science & Medicine

Volume 73, Issue 11, December 2011, Pages 1575-1584
Social Science & Medicine

Rising premature mortality in the UK’s persistently deprived areas: Only a Scottish phenomenon?

https://doi.org/10.1016/j.socscimed.2011.09.034Get rights and content

Abstract

In the international literature, many studies find strong relationships between area-based measures of deprivation and mortality. In the UK, mortality rates have generally fallen in recent decades but the life expectancy gap between the most and least deprived areas has widened, with a number of Scottish studies highlighting increased mortality rates in deprived areas especially in Glasgow. However, these studies relate health outcomes at different time points against period-specific measures of deprivation which may not be comparable over time. Using longitudinal deprivation measures where levels of area deprivation are made comparable over time, a recent study demonstrated how levels of mortality change in relation to changing or persistent levels of (non-) deprivation over time. The results showed that areas which were persistently deprived in Scotland experienced a rise in premature mortality rates by 9.5% between 1981 and 2001.

Here, focussing on persistently deprived areas we extended the coverage to the whole of the UK to assess whether, between 1991 and 2001, rising premature mortality rates in persistently deprived areas are a Scottish only phenomenon or whether similar patterns are evident elsewhere and for men and women separately. We found that male premature mortality rates rose by over 14% in Scotland over the 10-year period between the early 1990s and 2000s in persistently deprived areas. We found no significant rise in mortality elsewhere in the UK and that the rise among men in Scotland was driven by results for Glasgow where mortality rates rose by over 15% during the decade. Our analyses demonstrate the importance of identifying areas experiencing persistent poverty. These results justify even more of a public health focus on Glasgow and further work is needed to understand the demographic factors, such as health selective migration, immobility and population residualisation, which may contribute to these findings.

Highlights

► Uses a deprivation measure comparable over time for small areas across the whole of the UK. ► There is a specific focus on those areas within the UK which are persistently most deprived over time. ► Determines whether an increase in premature mortality found in Scotland exists for males and females and outside Scotland. ► Finds that premature mortality rates decline across the UK’s persistently deprived areas except for males in Scotland, especially in Glasgow where premature mortality rates rise significantly.

Introduction

Many studies find strong relationships between area-based measures of deprivation and all-cause and cause-specific mortality in the UK (Eames et al., 1993, Law and Morris, 1998, Romeri et al., 2006) and across many countries in Europe (Huisman et al., 2004) including Spain (Benach, Yasui, Borrell, Sáez, & Pasarin, 2001), Belgium (Lorant, Thomas, Deliège, & Tonglet, 2001), and France (Rey, Eric Jougla, Fouillet, & Hémon, 2009). Strong relationships between deprivation and mortality are also found in the USA (Singh, 2003), Canada (Auger et al., 2010, Pampalon et al., 2008), Australia (Butler, Petterson, Bazemore, & Douglas, 2010) and New Zealand (Pearce & Dorling, 2006). These relationships also exist for infant mortality (Norman, Gregory, Dorling, & Baker, 2008) and for morbidity (Boyle et al., 1999, Norman and Bambra, 2007).

Of particular concern in the UK is that subnational inequalities in mortality have been increasing over the last few decades. Although mortality rates have generally fallen, the decline has been slower in more deprived areas so that the mortality gap between the most and least deprived areas has widened (see, for example, Higgs et al., 1998, O’Flaherty et al., 2009, Rees et al., 2003, Shaw et al., 2004). Thomas, Dorling, and Davey Smith (2010) demonstrate that in Britain the ratio of standardised mortality ratios (SMRs) in the most deprived areas to SMRs in the least deprived areas increased from 1.61 in 1990–1991 to 1.88 in 2006–2007, an increase of just over 1% per year. During the same period, inequalities for mortality for ages 0–74 increased from a most to least deprived ratio of 2.17–2.79, an increase of 1.8% per year. For countries in Europe, Huisman et al. (2004) find that whilst inequalities in relative socio-economic mortality generally decrease with age, absolute inequalities increase. For Europe and selected countries in the rest of the world, Leon (2011) shows a general trend for longer life expectancies over time, but that differences between countries are increasing in many cases with the differences in life expectancy for males and females persisting. Similarly, in New Zealand, whilst overall life expectancy has increased during the last decades of the 20th Century, an increase spatial polarisation of health occurred along with a sharp rise in inequalities (Pearce & Dorling, 2006).

Compared with the UK and most of Europe, men and women in Scotland have relatively low life expectancy (GROS, 2010). In 2007, males in Scotland had life expectancy at birth 1.0 year lower than the European Union (EU) average of 76.1 while the life expectancy for females was 2.3 years lower than the EU average of 82.2. For males, the life expectancy gap between Scotland and the whole of the UK has widened by 0.4 years since 1997 to 2.5 in 2010. Across the EU, only Eastern European countries have shorter life expectancies than those experience in Scotland. For subnational geographies, a small number of Scottish studies have actually highlighted increasing mortality rates in deprived areas for: young adults (Leyland et al., 2007, McLoone and Boddy, 1994); certain causes of disease (Boyle et al., 2004, Leyland et al., 2007); and overall premature mortality rates in men in the most deprived areas (Leyland et al., 2007). For example, Leyland et al. (2007) found a rise in premature mortality rates for men in the most deprived areas of Scotland in 2001 compared to 1991 driven by rising mortality in alcohol, drug and mental health related causes amongst young adults particularly in the Glasgow region of the country. Life expectancy has also been shown to have declined slightly amongst men in the most deprived areas of Greater Glasgow from 65.3 years in 1981 to 64.4 years in 2001 (Hanlon, Walsh, & Whyte, 2006). Subsequently, Walsh, Bendel, Jones, and Hanlon (2010) found that Glasgow has a considerably worse level of health, compared with Liverpool and Manchester, which could not be explained solely by socio-economic circumstances.

However, these studies have used period-specific measures of deprivation. This means that deprivation scores from one period to the next are not directly comparable (Norman, 2010a), making it impossible to study areas that have been amongst the consistently most deprived in absolute terms over time. Recently, Exeter, Boyle, and Norman (2011) introduced a novel method for calculating Carstairs deprivation scores (Carstairs & Morris, 1989) that could be compared over time using areas in Scotland whose boundaries remained stable over the entire period 1981 to 2001. This analysis demonstrates that the persistently most deprived areas in Scotland (1981–2001) had seen premature mortality rates rise by 9.5% during the two decades.

Given that at a national level in Scotland, and across the UK, life expectancy has consistently risen year on year over this period, it is of concern that the average length of life for persons living in the persistently most deprived areas may be shortening. In this paper, we extend Exeter et al.’s recent Scotland analysis to the whole of the UK to assess whether, between 1991 and 2001, rising premature mortality rates in persistently deprived areas are a Scottish only phenomenon or whether similar patterns are evident in the other countries and regions of the UK. If this has only occurred in Scotland our results may provide some insight into the phenomena known as the Scottish and Glasgow ‘effects’, which refer to the inability of current levels of deprivation to fully explain Scotland’s high mortality rates compared to the rest of the UK (Hanlon et al., 2005, Walsh et al., 2010, Walsh et al., 2010).

Section snippets

Data and methods

Some fundamental difficulties restrict analyses of the relationship between mortality and deprivation for small areas across the whole of the UK; both cross-sectional and time-series. For this analysis we require mortality and population denominator data for small areas that are consistent through time between 1991 and 2001. The Consistent Areas Through Time (CATTs) which have been created in Scotland (Exeter, Boyle, Feng, Flowerdew, & Schierloh, 2005) and were used by Exeter et al. (2011)

Results

The 56 areas classified here into the most persistently deprived quintile in Scotland in both 1991 and 2001 represent 6% of the 935 Scottish small areas used in this work (Table 1). In 2001, these areas contained 265,982 persons aged 0–64 representing 6% of Scotland’s population of this age-group. Of the 35,495 premature deaths during 2000–2002 in Scotland as a whole, 4087 (12%) occurred in these persistently deprived areas resulting in a three year crude death rate of 5.12 per 1000 persons. In

Discussion

Here we have found that male premature mortality rates rose in Scotland over the 10-year period between the early 1990s and 2000s in persistently deprived areas. This is consistent with the evidence from previous studies using period-specific (Leyland et al., 2007) and longitudinal measures of deprivation (Exeter et al., 2011). Crucially we found that there is no significant rise in mortality in any of the other countries or regions of the UK and that the rise among men in Scotland is driven by

Acknowledgements

This research used Census data obtained via the MIMAS CASWEB facility and postcode directories and GIS boundary data obtained via the EDINA UKBORDERS facility, both services supported by ESRC and JISC. The Census, official Mid-Year Estimates and Vital Statistics data for England and Wales, Scotland and Northern Ireland have been provided by ONS, GROS and NISRA and the digital boundary data by OSGB and OSNI. These data are Crown copyright and are reproduced with permission of OPSI. Paul Norman’s

References (67)

  • R. Pampalon et al.

    Recent changes in the geography of social disparities in premature mortality in Québec

    Social Science & Medicine

    (2008)
  • F. Popham et al.

    Selective internal migration. Does it explain Glasgow's worsening mortality record?

    Health & Place

    (2011)
  • J. Wilkinson

    What we want to know is …

    Public Health

    (2010)
  • J. Adams et al.

    How accurate are Townsend deprivation scores as predictors of self-reported health? A comparison with individual data

    Journal of Public Health

    (2004)
  • N. Auger et al.

    Sex, age, deprivation and patterns in life expectancy in Quebec, Canada: a population-based study

    BMC Public Health

    (2010)
  • A. Bates

    The development of a postcode best fit methodology for producing population estimates for different geographies

    Population Trends

    (2008)
  • C. Beatty et al.

    The diversion from ‘unemployment’ to ‘sickness’ across British regions and districts

    Regional Studies

    (2005)
  • J. Benach et al.

    Material deprivation and leading causes of death by gender: evidence from a nationwide small area study

    Journal of Epidemiology and Community Health

    (2001)
  • P. Boyle et al.

    Suicide gap among young adults in Scotland: population study

    British Medical Journal

    (2004)
  • P. Boyle et al.

    The role of population change in widening the mortality gap in Scotland

    Area

    (2005)
  • P. Boyle et al.

    Migration and health

  • J. Brown et al.

    Establishing the potential for using routine data on incapacity benefit to assess the local impact of policy initiatives

    Journal of Public Health

    (2008)
  • D.C. Butler et al.

    Use of measures of socioeconomic deprivation in planning primary health care workforce and defining health care need in Australia

    Australian Journal of Rural Health

    (2010)
  • V. Carstairs et al.

    Deprivation: explaining differences in mortality between Scotland, England and Wales

    British Medical Journal

    (1989)
  • S. Curtis et al.

    Is urban regeneration good for health? Perceptions and theories of the health impacts of urban change

    Environment & Planning C

    (2002)
  • G. Davey Smith et al.

    Shrinking areas and mortality

    The Lancet

    (1998)
  • G. Davey Smith et al.

    Area based measures of social and economic circumstances: cause specific mortality patterns depend on the choice of index

    Journal of Epidemiology and Community Health

    (2001)
  • M. Eames et al.

    Social deprivation and premature mortality: regional comparison across England

    British Medical Journal

    (1993)
  • D. Exeter et al.

    The creation of ‘Consistent Areas Through Time’ (CATTs) in Scotland, 1981–2001

    Population Trends

    (2005)
  • D.J. Exeter et al.

    Shrinking areas and mortality: an artefact of deprivation effects?

    Journal of Epidemiology & Community Health

    (2005)
  • B. Galobardes et al.

    Is the association between childhood socioeconomic circumstances and cause-specific mortality established? Update of a systematic review

    Journal of Epidemiology and Community Health

    (2008)
  • I.N. Gregory

    Comparisons between geographies of mortality and deprivation from the 1900s and 2001: spatial analysis of census and mortality statistics

    BMJ

    (2009)
  • GROS

    Life expectancy for administrative areas within Scotland, 2007–2009

    (2010)
  • Cited by (54)

    • The geographic harmonisation of Scotland's small area census data, 1981 to 2011

      2019, Health and Place
      Citation Excerpt :

      This means, for example, a study comparing the association between area deprivation and mortality between 1981 and 2011 would use an area deprivation measure derived from the 1981 census to calculate mortality rates for that period and use the 2011 census-based deprivation index for the more recent period. Such results demonstrate how the social gradient has changed over time, but the ability to study areas whose socio-economic conditions have improved, worsened or remained broadly consistent over time is impossible (Norman et al., 2011). Previously, we found that although premature mortality had reduced in all period-specific deprivation quintiles in 1981 and 2001, when we used time comparable deprivation in the CATTs approach, mortality in the persistently deprived areas increased by nearly 10% over the 20 year period (Exeter et al., 2011).

    View all citing articles on Scopus
    View full text