Neighborhoods, psychological distress, and the quest for causality

https://doi.org/10.1016/j.copsyc.2019.06.009Get rights and content

Highlights

  • Neighborhoods may influence psychological distress but evidence for causality is mixed.

  • Causality has been examined by intervention, longitudinal, and twin studies.

  • Overall evidence suggests only limited support for causal neighborhood associations with psychological distress.

Neighborhood characteristics have been associated with psychological distress, but it is uncertain whether these associations are causal. The current article reviews data from interventions and quasi-experimental studies that have addressed the question of causality of neighborhood associations. Overall, data from neighborhood interventions, longitudinal studies, and twin studies have provided only limited and inconsistent evidence to support causal interpretation of neighborhood associations with psychological distress: very few findings have been replicated across different samples, and many associations have been observed only with some of the multiple measures included the studies. Studies that examine the effects of neighborhood change on people’s wellbeing are needed to improve causal inference and policy relevance of neighborhood studies.

Introduction

Metropolitan city centers, sparsely populated suburbs, remote rural towns, and other residential locations differ from each other in many aspects. Some neighborhoods have plenty of recreational opportunities, other have high crime rates; some invest in new bike lanes and urban amenities, others cannot attract money to cover pot holes; some attract young singles who soon move away, others are inhabited by families and retirees who tend to stay longer. It is reasonable to hypothesize that such differences in residential characteristics influence people’s mental health and wellbeing [1, 2, 3].

Dozens of studies have linked neighborhood characteristics with residents’ mental health problems, including depression, schizophrenia, and antisocial behaviors [4]. These associations are commonly labelled as neighborhood effects, but often the label promises too much; the majority of neighborhood studies have been cross-sectional, which makes it impossible to exclude the alternative explanation of selective residential mobility [5]. People with poorer mental health may, on average, end up living in different neighborhoods than those without mental health problems [2,6]. This could be caused directly by mental health problems (e.g. lower motivation to move, difficulties in deciding where to move) or indirectly by factors that influence both residential mobility and mental health (e.g. lower socioeconomic status constraining mobility options).

Yet it seems plausible that neighborhoods do influence mental health. The current review focuses on studies that have leveraged experimental or quasi-experimental study designs to identify potentially causal neighborhood effects on psychological distress, that is, symptoms of depression, anxiety, and unspecified somatic complaints that tend to co-occur in the general population. To cover the research literature as broadly as possible, I performed a literature search using Scopus database (scopus.com) searching titles, abstracts, and keywords for: (‘neighborhood’) AND (causal OR longitudinal OR experiment* OR quasi-experiment* OR twin) AND (depress* OR distress OR anxiety). After reviewing the titles and abstracts of 526 documents, I found 15 relevant studies and one review (not counting all the published articles from some of the individual studies) that formed the core of this review.

Section snippets

Community interventions

In the Moving to Opportunity (MTO) experiment, low-income families in five large U.S. cities were randomized to get housing vouchers that allowed them to move away from high-poverty neighborhoods [7]. This led to lower psychological distress among adult participants in the treatment versus control group [8], and some mental health benefits were observed even in the long-term follow-up 10–15 years after the experiment [9]. A further analysis showed that mental health improved only for those who

Longitudinal studies with fixed-effect regression

Longitudinal studies that measure mental health at multiple time points are an improvement to cross-sectional studies. But if they measure neighborhood characteristics only at baseline, they are still subject to similar confounding biases as cross-sectional studies. A better longitudinal study design uses repeated measurements of both neighborhoods and mental health, so that the participants can act as their own controls at different time points, adjusting for all individual characteristics

Twin studies

Twin studies can adjust for confounding that arises due to genetic and environmental influences that have made siblings more similar to each other. Three studies from the Washington State Twin Registry have examined neighborhood associations with mental health. The first [27] reported that neighborhood deprivation was not associated with depression after the shared genetic and environmental factors of twin pairs were taken into account (0.10SD versus 0.035SD difference per 1SD difference in

Conclusions

In an ideal situation, evidence from different study designs, samples, and methods would converge to a common conclusion when weighting all the available evidence for causality [31]. In the case of neighborhood effects, the causal evidence has not yet converged to robust conclusions [5,32]. The overall evidence from different types of studies of psychological distress is not particularly strong, but some of the results from experimental and quasi-experimental study designs do suggest possible

Conflict of interest statement

Nothing declared.

References and recommended reading

Papers of particular interest, published within the period of review, have been highlighted as:

  • • of special interest

  • •• of outstanding interest

References (40)

  • K.M. Fedak et al.

    Applying the Bradford Hill criteria in the 21st century: how data integration has changed causal inference in molecular epidemiology

    Emerg Themes Epidemiol

    (2015)
  • T.A. Glass et al.

    Are neighborhoods causal? Complications arising from the ‘stickiness’ of ZNA

    Soc Sci Med

    (2016)
  • A. Sariaslan et al.

    Schizophrenia and subsequent neighborhood deprivation: revisiting the social drift hypothesis using population, twin and molecular genetic data

    Transl Psychiatry

    (2016)
  • L. Colodro-Conde et al.

    Association between population density and genetic risk for schizophrenia

    JAMA Psychiatry

    (2018)
  • M. Jokela et al.

    Geographically varying associations between personality and life satisfaction in the London metropolitan area

    Proc Natl Acad Sci U S A

    (2015)
  • P.J. Rentfrow et al.

    Geographical psychology: the spatial organization of psychological phenomena

    Curr Dir Psychol Sci

    (2016)
  • P.J. Rentfrow et al.

    Regional personality differences in Great Britain

    PLoS One

    (2015)
  • Duncan DT, Kawachi I (Eds): Neighborhoods and Health, edn 2....
  • J.M. Oakes et al.

    Twenty years of neighborhood effect research: an assessment

    Curr Epidemiol Rep

    (2015)
  • M.A. Green et al.

    Using internal migration to estimate the causal effect of neighborhood socioeconomic context on health: a longitudinal analysis, England, 1995–2008

    Ann Am Assoc Geogr

    (2017)
  • Cited by (7)

    View all citing articles on Scopus
    View full text