Elsevier

Social Science & Medicine

Volume 177, March 2017, Pages 239-247
Social Science & Medicine

Review article
Illuminating the lifecourse of place in the longitudinal study of neighbourhoods and health

https://doi.org/10.1016/j.socscimed.2016.09.025Get rights and content

Highlights

  • Neighbourhoods and health are multi-dimensional and dynamic constructs.

  • Changes in health may be intricately tied to neighbourhood change.

  • The lifecourse of place offers a framework to consider the dynamic nature of contexts.

  • Latent transition analysis (LTA) measures change across meaningful categorical states.

  • LTA can be used to model the lifecourse of place to study place-health.

Abstract

Place and health are inextricably entwined. Whilst insights have been gained into the associations between places, such as neighbourhoods, and health, the understanding of these relationships remains only partial. One of the reasons for this relates to time and change and the inter-relationships between the dynamic nature of both neighbourhoods and health. This paper argues that the lifecourse of place can be used as a conceptual framework to understand the evolution and ongoing development of neighbourhoods, and their impact on the geographies of health, past, present and future. Moreover, this paper discusses the capacity of a longitudinal form of enquiry – latent transition analysis – that is able to operationalise conceptual models of the lifecourse of place. To date, latent transition analysis has not been applied to the study of neighbourhoods and health. Drawing on research across a range of disciplines including developmental psychology, sociology, geography and epidemiology, this paper also considers praxis-based implications and recommendations for applications of latent transition analysis that aim to advance understanding of how neighbourhoods affect health in and over time.

Introduction

Time and change complicate the evident associations between neighbourhoods and health with longitudinal methods providing an apparatus to inform these relationships. In considering neighbourhoods from a longitudinal perspective, there is often a sense that they are constantly evolving. In many respects they are. But neighbourhoods can also display persistent qualities. This durability of character can function to pattern and shape health disparities in space and over time. For example, using an ecological analysis, Dorling et al. (2000) observed that within inner London, at the level of local government wards, poverty for many localities remained invariant between 1896 and 1991. Moreover, mortality attributable to diseases such as lung cancer and stroke in 1991, was predicted more strongly by the spatial distribution of a historic 1896 measure of poverty than a more contemporaneous expression of this exposure.

Noteworthy as this analysis is, it only affords a disjointed view having compared neighbourhood contexts at junctures distant to each other. Moreover, it does not capture changes in the nature of areas temporally proximate to the experiences and lives of current dwellers; though such studies are beginning to emerge. For example, using the US-situated Geolytics Neighborhood Change Database (NCDB), Do (2009) observed historic multiple-year measures of census tract poverty (three time points, each ten years apart) to be stronger predictors of self-rated health than single-point-in-time measures. Critically, these historic exposures also explained a greater amount of the identified racial disparities in health status. More recently, Mair et al. (2015), used data from the Multi-Ethnic Study of Atherosclerosis to explore changes in neighbourhood social cohesion, stress, violence, safety and aesthetic environment in a sample of 103 US-based census tracts across a 2.5-to-4 year period. Although estimated associations were imprecise and non-significant following statistical adjustment for covariates, changes in neighbourhood contextual conditions were implicated in corresponding changes in levels of depressive symptoms.

Whilst there is an interest in the health impacts of exposure to changes in neighbourhood attributes, research needs to delve into the dynamics of neighbourhoods and the associated “dynamics of person/place experiences” (Kemp, 2011, p.4). Some time ago Pred (1984) contended that studies of human settlements were at risk of reducing analysed contexts to fragmented and frozen scenes, arguing that settlements, such as neighbourhoods, never materialise fully formed. Nor do they lay dormant. Rather, neighbourhoods are spatiotemporal products. This perspective has been echoed (Cummins et al., 2007, Pearce, 2015, Robert et al., 2010, Tunstall et al., 2004), with a call for enquiries to assess the developmental history and temporal progression of neighbourhoods, and the impact of these dynamics on the geographies of health, past, present and future (Pearce, 2015).

Explicating the temporal ebb and flow of neighbourhoods, and identifying if and how this evolution gets under the skin to influence population levels of health is integral to strengthening the evidence-base concerned with neighbourhood effects. In addition, by considering how structural and socio-political determinants function to prime and condition residential contexts over time, prospective place-based interventions may be grounded in an enriched understanding. Longitudinal enquiry can benefit these purposes. Focusing on neighbourhoods, this paper has two broad aims. The first is to advance the lifecourse perspective to the lifecourse of place. The second, is to introduce and discuss latent transition analysis (LTA); a longitudinal method of analysis able to operationalise conceptual models of the lifecourse of place in order to inform the study of neighbourhoods and health.

Section snippets

Lifecourse of place

The lifecourse perspective provides an organising framework for the study of health and its development over time (Ben-Shlomo and Kuh, 2002). A central aim of the lifecourse approach is to inform understandings of how health at later time points, or periods of life, is impacted by earlier experiences, such as during gestation, childhood, adolescence, or adulthood. In this manner, time, timing, and sequencing are instrumental factors in lifecourse analyses (Kuh et al., 2003). Furthermore, the

Latent transition analysis

Longitudinal analysis involves attending to the manner and form of change over time. From a conceptual perspective there are two theoretical models of time-related change – continuous and discrete (Collins, 2006). Continuous models of change assume that the change process unfolds in a smooth, linear or curvilinear fashion (Kim and Böckenholt, 2000). In addition, the mathematical function that models the change process is assumed to be analogous for each subject, though inter-subject variability

Latent transition analysis and the lifecourse of place

Classifying neighbourhoods into meaningful and dimensional categories has been a feature of place-health research, although much of this applied study has been cross-sectional. Longitudinal analyses of neighbourhoods have been challenged by temporal shifts that arise in spatial (administrative) boundaries, making it difficult to compare data for given areas across time (Logan et al., 2014). They have also been constrained by the availability of diverse area-level indicators that have been

Assumptions of latent transition models and recommendations for practice

Attention is drawn to three model assumptions as they relate to the applied study of neighbourhoods. The first assumption is that neighbourhood sub-types exist and the manner in which neighbourhoods evolve, develop, and transition between states can be defined. This is relevant as finite mixture models, of which LTA is a variant, can serve two purposes (Bauer and Curran, 2003a, Bauer and Curran, 2003b). One purpose is to expose qualitatively distinct sub-groups that are not directly observed.

Conclusion

Neighbourhoods and health are multidimensional constructs. The inter-relationships between neighbourhoods and health are also heterogeneous and dynamic, transpiring in evolving socio-cultural and socio-historic circumstances. Understanding these associations is likely to be facilitated and enriched by longitudinal forms of enquiry. It is these notions that underpin the call for the application of LTA to the study of neighbourhoods and health across time. As a methodological tool LTA is adept at

Funding/disclosures

Mr Peter Lekkas was supported by: an Australian Postgraduate Award, Department of Education and Training, Australian Government; a University of South Australia Scholarship; and the School of Health Sciences, University of South Australia through the Research Chair, Social Epidemiology.

Dr Catherine Paquet was funded by a National Health and Medical Research Council (Australia) Program Grant (#0631947).

Prof. Mark Daniel was funded by a Research Chair, Social Epidemiology, University of South

Author contributions and acknowledgements

Mr Peter Lekkas initiated the idea for the research, conceptualised the manuscript, wrote the first draft of the manuscript, and critically revised subsequent versions of the manuscript.

Dr Catherine Paquet contributed to the conceptualisation of the paper, helped with methodological interpretations, provided critical feedback, suggested additional discussion and provided revisions to the manuscript.

Dr Natasha Howard contributed to the conceptualisation of the paper, provided critical feedback

References (78)

  • M.M. Weden et al.

    Neighborhood archetypes for population health research: is there no place like home?

    Health Place

    (2011)
  • J.S. Ahlquist et al.

    Model-based clustering and typologies in the social sciences

    Polit. Anal.

    (2012)
  • F. Bartolucci et al.

    Latent Markov models: a review of a general framework for the analysis of longitudinal data with covariates

    TEST

    (2014)
  • D.J. Bauer et al.

    Distributional assumptions of growth mixture models: implications for overextraction of latent trajectory classes

    Psychol. Methods

    (2003)
  • D.J. Bauer et al.

    Overextraction of latent trajectory classes: much ado about nothing? Reply to Rindskopf (2003), Muthén (2003),and Cudeck and Henly (2003)

    Psychol. Methods

    (2003)
  • Y. Ben-Shlomo et al.

    A life course approach to chronic disease epidemiology: conceptual models, empirical challenges and interdisciplinary perspectives

    Int. J. Epidemiol.

    (2002)
  • B.C. Bray et al.

    Modeling relations among discrete developmental processes: a general approach to associative latent transition analysis

    Struct. Equ. Model.

    (2010)
  • E.E. Bruch et al.

    Methodological issues in the analysis of residential preferences, residential mobility, and neighborhood change

    Sociol. Methodol.

    (2012)
  • A. Carpenter et al.

    Poverty and connectivity: crossing the tracks

    J. Space Syntax.

    (2010)
  • R. Chetty et al.

    The Impacts of Neighborhoods on Intergenerational Mobility: Childhood Exposure Effects and County-level Estimates

    (2015)
  • J. Chow

    Differentiating urban neighborhoods: a multivariate structural model analysis

    Soc. Work Res.

    (1998)
  • I. Cole

    Whose place? Whose history? Contrasting narratives and experiences of neighbourhood change and housing renewal

    Hous. Theory Soc.

    (2013)
  • L.M. Collins

    Analysis of longitudinal data: the integration of theoretical model, temporal design, and statistical model

    Annu. Rev. Psychol.

    (2006)
  • L.M. Collins et al.

    Latent Class and Latent Transition Analysis: with Applications in the Social, Behavioral, and Health Sciences

    (2010)
  • L.M. Collins et al.

    Repeated-measures latent class analysis and latent transition analysis

  • R.K. Dishman et al.

    Failure of post-action stages of the transtheoretical model to predict change in regular physical activity: a multiethnic cohort study

    Ann. Behav. Med.

    (2009)
  • P. Do

    The dynamics of income and neighborhood context for population health: do long-term measures of socioeconomic status explain more of the black/white health disparity than single-point-in-time measures?

    Soc. Sci. Med.

    (2009)
  • D. Dorling et al.

    The ghost of Christmas past: health effects of poverty in London in 1896 and 1991

    Br. Med. J.

    (2000)
  • A. Downs

    Neighborhoods and Urban Development

    (1981)
  • J.R. Edwards

    The fallacy of formative measurements

    Organ Res. Methods

    (2011)
  • G.H. Elder

    Time, human agency, and social change: perspectives on the life course

    Soc. Psychol. Q.

    (1994)
  • Z. Feng et al.

    Fuzzy geodemographics: a contribution from fuzzy clustering methods

  • C.S. Fischer

    Showing that neighborhoods matter

    City Community

    (2013)
  • S. Galea

    An argument for a consequentialist epidemiology

    Am. J. Epidemiol.

    (2013)
  • B. Guo et al.

    Using latent class and latent transition analysis to examine the transtheoretical model staging algorithm and sequential stage transition in adolescent smoking

    Subst. Use Misuse

    (2009)
  • R. Harris et al.

    Changing Suburbs: Foundation, Form and Function

    (1999)
  • K.L. Henry et al.

    Multilevel latent class analysis: an application of adolescent smoking typologies with individual and contextual predictors

    Struct. Equ. Model.

    (2010)
  • D. Hyra et al.

    The US Great Recession: exploring its association with black neighborhood rise, decline and recovery

    Urban Geogr.

    (2016)
  • C.B. Jarvis et al.

    A critical review of construct indicators and measurement model misspecification in marketing and consumer research

    J. Consum. Res.

    (2003)
  • Cited by (26)

    • Neighborhood deprivation and obesity: Sex-specific effects of cross-sectional, cumulative and residential trajectory indicators

      2022, Social Science and Medicine
      Citation Excerpt :

      These studies suggest that, for seniors, moving was associated with life events such as the death of a spouse or health problems, whereas individual income was not associated with moving. As for changes in neighborhood composition, many authors have shown that neighborhood SES is not fixed in time and can change according to its geographic, urban and initial SES characteristics (Ades et al., 2016; Lekkas et al., 2017; Séguin et al., 2012). In Quebec, most studies analyze the change in neighborhood SES in Montreal, the main metropolitan center.

    • Contextualising lifestyles: how socially contrasting places in Fife, Scotland influence lay understandings of lifestyle and health behaviours in relation to coronary heart disease

      2020, Health and Place
      Citation Excerpt :

      Contextual characteristics that can influence health and may be health promoting or health damaging (Macintyre et al., 2002) include physical features or the natural environment (Gatrell and Elliot, 2009); provision and availability of services, including healthcare, transport links (Kawachi and Berkman, 2003; Ecob and Macintyre, 2000); socio-cultural features – norms and values, neighbourliness, community spirit, trust (Macintyre and Ellaway, 2003) and reputation of an area (Boyle et al., 2004; Flowerdew et al., 2008). Geographical and socio-historical characteristics of place can also offer a collective explanation for place effects on health (Macintyre et al., 2002) and recent work by Lekkas et al. (2017) Pearce (2015) and Pearce et al. (2018) has revealed that the life-course of places – i.e. the dynamic historical, social, and material trajectories which occur in places, can impact on geographies of health and well-being past, present and future (Pearce, 2015; Cummins et al., 2007). These contextual understandings remain lacking in biomedical and public health understandings of health inequalities (Kearns, 1993; Jones and Moon, 1993).

    • Aging in place and the places of aging: A longitudinal study

      2020, Journal of Aging Studies
      Citation Excerpt :

      This involves looking at discrete parts, cases or contexts within the dataset and documenting something about those parts specifically (Mason, 2002). This approach enabled us to explore a complex set of experiences in detail and to recount the role of life events and neighborhood change over time (Feagin, Orum, & Sjoberg, 1991) as well as the impact of persistent qualities of neighborhoods (Lekkas, Paquet, Howard, & Daniel, 2017). Secondary data were derived from a panel study which represented one strand of a large-scale project called ‘Step Change: A longitudinal Qualitative Study on Travel, Transport and Mobility’ (Miles et al., 2018).

    View all citing articles on Scopus
    View full text