Elsevier

Public Health

Volume 118, Issue 1, January 2004, Pages 11-20
Public Health

Health inequalities in Seville, Spain: use of indicators of social deprivation and mortality in small areas

https://doi.org/10.1016/S0033-3506(03)00141-0Get rights and content

Abstract

Objectives. To analyse the spatial distribution of social structure and mortality in Seville, and to examine the association between various social indicators and mortality.

Methods. Small areas of the city were typified on the basis of four social indicators, which were used to derive a social index. Overall mortality and cause of death were studied in two age groups (1+years and 1–64 years). Pearson's correlation coefficient was used to examine the relationship between the social indicators and mortality.

Results. Significant social and mortality differences, particularly in premature mortality of males, were found between the areas studied. However, when the basic health zones are grouped together by social level, these differences in mortality are not so clear. The social indicators that correlate most closely with mortality are unemployment and illiteracy. When the social index is used, the correlations are weaker. Premature death from trauma in males presents the highest coefficient of correlation with unemployment and illiteracy.

Conclusions. The social index used in the present study places less emphasis on material differences than those used by Townsend et al. and Carstairs and Morris. Also, it was not possible to study mortality by individual neighbourhoods in this study. Both factors could have influenced the finding that the correlations between both types of indicator are weaker with the social index than with unemployment and illiteracy.

Introduction

The analysis of social inequalities in respect of health is a priority, particularly at a time when there is much debate about the future of current healthcare systems and about the ways that these future perspectives will enable social demands to be met. It is essentially a question of debating how and how far individual and collective responsibilities should be delimited in the face of the existence of healthcare inequalities.1., 2., 3., 4., 5., 6. Nájera,7 when debating the age-old thorny question of whether individual health is or is not a social product, recalled the reference that Antonovsky8 made to the victims of the famous sinking of the Titanic in contending that ‘the time at which one dies is related to one's class’. However, the study of social inequalities in health should not be left as merely an analysis of the deficiencies in the application of the measures of the health system in different social groups;9., 10., 11., 12. that is, we must not only consider inequalities in services but should also use the analysis of inequalities as a conceptual and methodological tool to reveal the origin and social meaning of health, and to obtain evidence for the need to re-orientate the concepts and approaches on which health policies are based. In short, it is necessary to redefine the concepts of health and disease, and to discuss how the ‘social’ dimension is being used in health.13., 14., 15., 16., 17., 18.

The study presented here is set in the framework of a wider project which we have been undertaking in Seville, Spain.19 The purpose of this research is to demonstrate the relationships between the health profile of human groups and what our research group has been calling ‘the socio-ecological niche or nest’, understanding this as the ‘melting pot in which social factors are mixed and interact, where ‘ingredients’ such as work (if any), the level of education (if any), the living conditions (however squalid or luxurious), the human relationships (in all their diverse forms), …are all ‘cooked’ together with the physical environment and with biological and chemical components, etc.’ creating the conditions or processes for possible losses of health.16

In this context, it is clear that we must review whether the traditional parameters (person, place and time) of epidemiology continue to be useful in the analysis of social inequalities in respect of health. Without doubt, the most fundamental change should be that the person (the individual) ceases to be the most relevant parameter in the study of a particular health situation, and gives way to the community as the main unit of analysis. This change is closely related to one of the fundamental problems posed by the science of epidemiology, which is to define the concept of community or population. The current situation is characterized by the absence of a sufficiently clear idea of what concepts of community we are using in epidemiological research.

It is precisely when it comes to defining how we are going to configure these units of populational analysis that we must recognize that, conceptually, a community or population group does not necessarily have to be defined by a spatial principle that conditions it in its entirety. On the other hand, it can be very useful to concentrate first on the spatial component of a community to delimit the unit of analysis. In this respect, it must be borne in mind that, on the one hand, the population is not distributed at random in the various geographical spaces, and on the other hand, the empirical construction of community units of analysis is conditioned by the fact that a major part of the information used in epidemiology (such as censuses and records) is territorially based. In short, it is a question of using what have come to be termed ‘geographical-populational units’, which could be useful as the basis for investigation into socio-ecological niches. Taking into account the elements that would comprise the socio-ecological niche, the urban neighbourhood or quarter could be used in the configuration of these units. The urban neighbourhood is a low level of aggregation of individuals, in which it is realistic to consider that they are mutually linked by a series of bonds of co-existence, living together in close proximity within a relatively small common physical and social space. Under this definition, aggregation is much more meaningful than the mere summation of individuals. Our interest centres on taking these smallest units—the neighbourhoods of a city—and creating larger socially homogeneous zones that can serve in the study of differences in health patterns. The intention, then, is to apply, with a greater degree of coherence, the proposed ecological analysis of the cross-relationships between the social and mortality indicators.

To summarize, our general objective is to develop methodologies for the study of the health situation of an urban community. Our specific objectives are: (1) to conduct an analysis of the spatial distribution of the social structure of Seville, characterizing its historical neighbourhoods and basic health zones (BHZs) from the data of the 1991 population census; (2) to study the mortality in Seville during the 1990–1993 period, evaluating its spatial distribution; and (3) to analyse the ecological association between the indicators of mortality used and those of social deprivation.

Section snippets

Methods

For the configuration of socially homogeneous zones of Seville, the census section was chosen as the basic working unit, this being the smallest unit of division for the administration of the municipality, by numbers of inhabitants. From this, we obtained larger administrative and natural units: the BHZs (of which there are 32), the smallest unit of the health administration for planning and management; and the neighbourhoods (quarters or districts, known as ‘barrios’) (85 in number), which are

Results

The distribution of the indicator of unemployment in the neighbourhoods of Seville is positively asymmetric in character, such that most of the neighbourhoods present values below those for Seville as a whole in 1991 (19.49%). This asymmetry is much more exaggerated in the indicators of illiteracy of women, and of university graduates of both sexes. The distribution of the indicator of females in work, with secondary-level education, is also skewed to the right, with most neighbourhoods having

Discussion

Previously, population or ecological strategies that were configured using population data, either aggregated or separated by geographical areas, were mainly produced for divisions of regions or nations, and much less frequently for divisions of cities.26., 27., 28., 29., 30. As Carstairs and Morris stated in 1989,31 the almost universal use of a postal code as the basis for the classification of areas means that most events can be situated geographically, thus potentially solving many of the

Conclusion

Almost 20 years ago, Pringle44 complained that in the analysis of the geographical differences of health levels, the available studies were predominantly on the international and inter-regional scales; he attributed the lack of interurban studies to the non-availability of routinely published data at these levels. Oers and Reelick emphasized the need for the development of a local system of information to monitor the state of health of the large municipalities. Townsend et al.26 stated: ‘There

Acknowledgements

This research was funded by ‘Fondo de Investigaciones Sanitarias de la Seguridad Social’ and the ‘Instituto de Estadı́stica de Andalucı́a’, Spain.

References (45)

  • Nájera E. El enfoque epidemiológico en la formulación de polı́ticas de salud. First Chilean Congress on Epidemiology;...
  • Antonovsky

    Social class, life expectancy and overall mortality

    Milbank Mem Fund Q

    (1967)
  • G. Dahlgren et al.

    Strategies for equity in health: report from Sweden

    Int J Health Serv

    (1986)
  • M. Whitehead

    Inequalities in health in the 1980's

    (1987)
  • E. Nájera et al.

    La reorientación del sistema sanitario. Necesidad de un enfoque epidemiológico

    Rev Salud Publ

    (1989)
  • P.L. Castellanos

    Sobre el concepto de salud-enfermedad. Un punto de vista epidemiológico

    Cuadernos Médico Sociales

    (1987)
  • E. Nájera

    Investigación y desarrollo profesional

    (1988)
  • H. Waitzkin et al.

    A critical theory of medical discourse: how patients and health professionals deal with social problems

    Int J Health Serv

    (1989)
  • Garcı́a-Gil C, Cruz C, Morales JA, Alvarez M, Solano A. Análisis comparativo de la mortalidad en los distintos nichos...
  • Censo de la Poblation Española 1991

    (1993)
  • F. Garcı́a

    Introducción a la Estadı́stica en Sociologı́a

    (1985)
  • Cited by (15)

    View all citing articles on Scopus
    View full text