Elsevier

Health & Place

Volume 13, Issue 1, March 2007, Pages 123-137
Health & Place

Neighbourhood inequalities in health and health-related behaviour: Results of selective migration?

https://doi.org/10.1016/j.healthplace.2005.09.013Get rights and content

Abstract

We hypothesised that neighbourhood inequalities in health and health-related behaviour are due to selective migration between neighbourhoods. Ten-year follow-up data of 25–74-year-old participants in a Dutch city (Eindhoven) showed an increased probability of both upward and downward migration in 25–34-year-old participants, and in single and divorced participants. Women and those highly educated showed an increased probability of upward migration from the most deprived neighbourhoods; lower educated showed an increased probability of moving downwards. Adjusted for these factors, health and health-related behaviour were weakly associated with migration. Over 10 years of follow-up, selective migration will hardly contribute to neighbourhood inequalities in health and health-related behaviour.

Introduction

Living in deprived neighbourhoods is associated with an increased probability of premature mortality, morbidity and unhealthy behaviour, and this association is only partly explained by the composition of neighbourhood residents in terms of individual-level socioeconomic indicators (Pickett and Pearl, 2001; Bosma et al., 2001; Martikainen et al., 2003; Borrell et al., 2004; Marinacci et al., 2004; Kölegård Stjärne et al., 2002; Davey Smith et al., 1998; Diez Roux et al., 1999, Diez Roux et al., 2001, Diez Roux et al., 2003; Duncan et al., 1999; Reijneveld, 2002; Van Lenthe and Mackenbach, 2002; Van Lenthe et al., 2005a, Van Lenthe et al., 2005b). The processes underlying this collective nature of poor health and unhealthy behaviour in more deprived neighbourhoods are still far from clear. A contextual explanation for these inequalities would imply that certain characteristics of the physical, social or cultural environment (such as physical deterioration or a lack of social cohesion) to which all residents are exposed are more prevalent in deprived neighbourhoods, and are also related to poor health and unhealthy behaviour. A growing number of studies aim at unravelling this mechanism by trying to identify the most relevant environmental characteristics (Yen and Kaplan, 1999; Van Lenthe et al., 2005b).

Another potential process, however, through which poor health and unhealthy behaviour can become distributed unevenly by the neighbourhood socioeconomic environment is selective migration based on health. According to a direct selection effect, health determines migration in such a way that those who are healthy migrate to less deprived neighbourhoods, and those with health problems migrate to more deprived neighbourhoods. For example, the latter could be mediated by a disease-induced reduction in income. Studies have shown an increased mobility out of employment for chronically ill persons compared with healthy persons (Van de Mheen et al., 1999). According to the indirect selection effect, the same processes are determined by risk factors of diseases. One way to understand how (un)healthy behaviour may determine migration is by regarding it as part of a broader lifestyle or culture. For example, those who remain in more deprived and unhealthy neighbourhoods may be more acculturated in these neighbourhoods, while those moving out of these neighbourhoods may do so with the aim of finding a neighbourhood with cultural values that are more in line with their own values.

The question of whether and to what extent selective migration determines neighbourhood inequalities in health and unhealthy behaviour is still largely unanswered. This may be partly due to the lack of longitudinal studies that are needed to explore these processes. Verheij et al. (1998) investigated whether selective migration based on health was responsible for urban–rural variations in health, but found no evidence for this. It was suggested, however, that such a process could be responsible for health disparities at other geographical levels. The results of this study also do not exclude the possibility of selective migration from affluent to deprived areas or vice versa. Boyle et al. (2002) using Scottish census data on 13 broad socioeconomically distinct regions found no evidence of an increased probability of limiting long-term illness in migrants in the more deprived areas and of a decreased probability in migrants in the less deprived areas.

The Dutch GLOBE study is a prospective cohort study, aimed at explaining socioeconomic inequalities in health. Multilevel analyses using data from the study confirmed neighbourhood inequalities in mortality, overweight, physical activity and smoking (Bosma et al., 2001; Van Lenthe and Mackenbach, 2002; Van Lenthe et al., 2005b; Van Lenthe and Mackenbach (submitted)). In the GLOBE study, addresses of participants were collected annually from the baseline measurement in 1991 until 10 years of follow-up. This allowed us to investigate whether selective migration based on health or health-related behaviour was responsible for neighbourhood socioeconomic inequalities in poor health and unhealthy behaviour. The following hypotheses were tested:

  • 1.

    The probability of upward migration, i.e. moving from more to less deprived neighbourhoods, as compared to staying in neighbourhoods with a similar socioeconomic environment is higher for those reporting good self-perceived health, absence of chronic conditions or healthy behaviour.

  • 2.

    The probability of downward migration, i.e. moving from less to more deprived neighbourhoods, as compared to staying in neighbourhoods with a similar socioeconomic environment is higher for those reporting poor self-reported health, chronic conditions or unhealthy behaviour.

Section snippets

Study population

Data for this study are from the Dutch prospective GLOBE study, an ongoing study aimed at investigating explanations of socioeconomic inequalities in health. Detailed information about objectives and design of the study are presented elsewhere (Mackenbach et al., 1994). For the baseline measurement in 1991, a random sample of 27,070 non-institutionalised subjects between 15 and 75 years of age and living in or near the town of Eindhoven were invited to fill in a postal questionnaire, including

Results

From the total study population, the majority of persons stayed within the same quartile of the neighbourhood socioeconomic environment (n=6501, 72.5%). Of these, the majority did not move (n=5449), and the rest moved within the quartile of neighbourhoods with the same neighbourhood socioeconomic environment (n=1052) (Table 1); 768 participants (8.5%) moved upwards, 597 participants (6.6%) moved downwards and 1100 participants (12.3%) moved outside the city of Eindhoven.

Compared to the other

Discussion

Neighbourhood inequalities in health and health-related behaviour could be the result of selective migration determined by health and health-related behaviour. This study found that age and marital status determined both upward and downward migration: 25–34-year-old persons and single and divorced persons moved relatively often towards neighbourhoods with both a higher and a lower neighbourhood socioeconomic environment. In addition, women showed an increased probability of upward migration,

Acknowledgements

The GLOBE study is carried out by the Department of Public Health of the Erasmus Medical Centre Rotterdam, in collaboration with the Public Health Services of the city of Eindhoven and region South-East Brabant. The authors express their gratitude to Dr. Basile Chaix of the Research Team on Social Determinants of Health and Health Care, French National Institute of Health and Medical Research for his comments on the pre-final draft of the manuscript. The authors are indebted to Ilse Oonk and

References (34)

  • D. Cooke et al.

    Self-reported alcohol consumption and dissimulation in a Scottish urban sample

    Journal of Studies on Alcohol

    (1983)
  • G. Davey Smith et al.

    Individual social class, area based deprivation, cardiovascular disease risk factors, and mortality

    Journal of Epidemiology and Community Health

    (1998)
  • A.V. Diez Roux et al.

    Neighbourhood differences in diet: the atherosclerosis risk in communities (ARIC) study

    Journal of Epidemiology and Community Health

    (1999)
  • A.V. Diez Roux et al.

    Neighborhood of residence and incidence of coronary heart disease

    New England Journal of Medicine

    (2001)
  • A.V. Diez Roux et al.

    Area characteristics, individual-level socioeconomic indicators, and smoking in young adults

    American Journal of Epidemiology

    (2003)
  • A. Fox et al.

    Selection and mortality differentials

    Journal of Epidemiology and Community Health

    (1982)
  • E. Idler et al.

    Self-rated health and mortality: a review of twenty-seven community studies

    Journal of Health and Social Behavior

    (1997)
  • Cited by (75)

    • Longitudinal effects of physical and social neighbourhood change on suicide mortality: A full population cohort study among movers and non-movers in the Netherlands

      2022, Social Science and Medicine
      Citation Excerpt :

      By studying both movers and non-movers we differentiated between effects of neighbourhood change as a result of residential moving and changes in residential environments itself. While some studies excluded frequent movers (Green et al., 2017; van Lenthe et al., 2007), we maintained them in our analyses which reduced the risk of selection bias as frequent moving was associated with poor health and socioeconomic disadvantage (Larson et al., 2004). Notwithstanding these strengths, our study had limitations.

    View all citing articles on Scopus
    View full text