Article Text

Download PDFPDF

Inequalities II
Context or composition? Exploring the adult health and well-being of British children born in different regions
  1. S. Easton
  1. Human Geography, University of Sheffield, Sheffield, UK

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Background and Aim

    The aim of the research is to explore how multiple social and spatial disadvantage at birth and in early childhood may combine to affect health and well-being in adulthood at age 30 using the British Cohort Study (BCS70). Traditionally researchers examining the geography of poverty or disadvantage have bridged only two disciplines: sociology and geography, or have focused on single outcomes such as mortality or morbidity rates (Dorling, 1997; Tunstall et al, 2007). However this research takes a deeper view of “context”, based on Bronfenbrenner’s ecological systems theory, where the development of a child depends on the interaction of complex layers in their environment (physical, familial, peers, neighbourhood, wider society). This research therefore crosses several disciplinary boundaries by utilising indicators of early health, cognitive ability and behaviour in the modelling process as well as socioeconomic ones. Some of the variables explored include: birth trauma, breast-feeding, maternal “malaise”, antisocial/hyperactive behaviour at age 5, an index of early adverse events and child health problems.

    Adult Outcomes

    Three adult outcomes at age 29–30 were modelled taking into account individual/household socioeconomic and spatial variables. These are 1. Satisfaction with Life so far; 2. Self-reported health; 3. Presence of a limiting long-term illness.


    Multilevel Modelling: Data were nested by region and the social rating of the neighbourhood at the higher levels and household/individual characteristics at birth and age 5 at the lower level. Spatial level variables such as regional poverty rates at the relevant time (1975) were also included in the models. Multilevel logistic regression and multinomial category response models with MCMC estimation were used to fit the data using MLwiN.


    The presence of a limiting long-term illness in adulthood was found to be associated with being female, and low birthweight (linked to socioeconomic position at birth, maternal smoking during pregnancy). The association between low birthweight and having more health problems in childhood was significant. Interaction was found among maternal mental state, living in a poor neighbourhood and child antisocial behaviour. Lower life satisfaction in adulthood was found to be linked to the presence of a congenital abnormality and (socioeconomic position, smoking and) low birthweight, antisocial behaviour in: boys and children from large families, and being poor in a poor neighbourhood. There were interactions between socioeconomic position, cognitive ability and birthweight across the subgroups. Further investigation of Self-reported health at 30 is currently underway.