Original Article
Associations between neighborhood effects and physical activity, obesity, and diabetes: The Jamaica Health and Lifestyle Survey 2008

https://doi.org/10.1016/j.jclinepi.2014.08.004Get rights and content

Abstract

Objective

To examine the impact of neighborhood disorder, perceived neighborhood safety, and availability of recreational facilities on prevalence of physical activity (PA), obesity, and diabetes mellitus (DM).

Study Design and Setting

Multilevel analyses were conducted among 2,848 respondents from the 2007–08 Jamaica Health and Lifestyle Survey. Neighborhood effects were based on aggregated interviewer responses to systematic social observation questions. Mixed-effect logistic regression models were created to assess the relationship between neighborhood indicators and DM and the modifiable risk factors PA and overweight/obesity.

Results

There was significant clustering in PA levels of 20 minutes at least once per week (intraclass correlation coefficient [ICC] = 10.7%), low/no PA (ICC = 7.22%), diabetes (ICC = 5.44%), and obesity (ICC = 3.33%) across neighborhoods. Greater levels of neighborhood disorder, home disorder, and counterintuitively recreational space availability were associated with higher levels of low/no PA among women. There was significant interaction by sex between neighborhood infrastructure and overweight/obesity with a significant association in men (odds ratio [OR] = 1.16; 95% confidence interval [CI] = 1.05, 1.28) but not women (OR = 1.01; 95% CI = 0.95, 1.07).

Conclusion

Differences in PA and obesity-related outcomes among Jamaicans may be partially explained by characteristics of the neighborhood environment and differ by sex. Future studies must be conducted to determine the mechanistic pathways through which the neighborhood environment may impact such outcomes to better inform prevention efforts.

Introduction

What is new?

  • This is the first national survey in the Caribbean to examine neighborhood determinants of physical activity (PA), measures of adiposity, and diabetes mellitus (DM) within a multilevel framework.

  • There was significant clustering in PA, overweight/obesity, and DM across Jamaican neighborhoods with neighborhood characteristics having a greater effect in women.

  • Greater levels of neighborhood disorder, home disorder, and counterintuitively recreational space availability were associated with higher levels of low/no PA among women.

  • Better levels of neighborhood infrastructure were associated with overweight/obesity in men.

  • The implications of these findings include the need to further understand the upstream factors leading to clustering and associations observed.

  • Public health efforts to increase PA and reduce both overweight/obesity and DM should include targeted neighborhood-level interventions.

The increasing prevalence of diabetes mellitus (DM) worldwide is thought to be a result of rising obesity and physical inactivity [1]. DM, obesity, and associated comorbidities are a major threat to public health and pose a significant economic burden in both developing and developed countries [2], [3], [4], [5]. Considerable success in decreasing morbidity from chronic noncommunicable diseases such as DM can be attributed to better treatment of these conditions—primarily through pharmacological agents. However, health promotion activities that address underlying causes of these conditions, such as obesity and reduced physical activity (PA), have been effective and are necessary. Social factors are thought to be important determinants of risk, and efforts to understand and address them must also be intensified to stem the worsening epidemic of DM and obesity [6], [7]. The 2007–08 Jamaica Health and Lifestyle Survey II (JHLS II) found the following prevalences among 15–74-year-old Jamaicans: DM was 8% (standard error [S.E.], 0.6%) (females, 9.3% [S.E., 0.6%]; males, 6.4% [S.E., 1.0%]) and obesity 25% (S.E., 1.2%) (females, 37% [S.E., 1.5%]; males, 12% [S.E., 1.7%]), respectively [8]. The prevalence of low/no levels of PA was 45% (S.E., 1.5%) (females, 62% [S.E., 1.6%]; males, 28% [S.E., 2.0%]) [8].

The Jamaican data reported prevalence of these outcomes at both the national and parish level and whether survey participants resided in urban vs. rural areas. No significant differences were detected at these geographical levels [8]. However, very little is known about the variability of these health outcomes in Jamaica at the levels of smaller geopolitical units, such as constituencies or enumeration districts (EDs). Researchers have recognized that the neighborhood may contribute negatively or positively to an individual's health and health-related behaviors.

Studies in the United States [9], [10], Canada [11], [12], [13], Australia [14], [15], and Europe [16], [17] have documented geographic variability in PA, obesity, and DM. There are also studies that have explored whether associations exist between the built environment and PA [12], [18], [19], [20], [21]. Greater neighborhood disorder has been significantly associated with less PA [22], [23]. However, the findings are less consistent regarding an association of the availability of recreational spaces/other spaces or play areas with PA [12], [18], [24], [25]. Multilevel conceptual models and statistical techniques have facilitated social scientists and epidemiologists studying the effects of neighborhoods on an individual's health. We are unaware of research from nationally representative samples from developing countries, which have examined clustering of PA, obesity, and DM at neighborhood level. We hypothesize that within Jamaica, there is significant variability in these outcomes at the smaller geopolitical levels.

The purpose of our analysis was to examine the variability in PA, obesity, and DM across small geopolitical units known as EDs in Jamaica using data from the JHLS II. Among other things, these outcomes may be influenced by neighborhood characteristics, such as access to open spaces and neighborhood disorder [22], [26], [27]. We also sought to explore whether home disorder, perceived safety, and neighborhood measures, such as neighborhood infrastructure and disorder, availability of recreational areas/spaces/playing fields (including whether they were within walking distance), were associated with these behavioral and health outcomes within a multilevel framework.

Section snippets

Design

Data for this study came from the JHLS II, a cross-sectional interviewer-administered nationally representative survey. Additional details on the study design, sampling methodology, and recruitment process are documented in the full technical report [8]. The home addresses of participants were geographically linked to EDs within each parish. An ED is the smallest geographical unit into which Jamaica is divided to facilitate data collection for censuses and surveys [28]. The results presented

Results

Table 1 describes the weighted summary statistics for the Jamaican population between 15 and 74 years old. The mean age of participants was 36.9 (±0.1) years. Motor vehicle ownership was significantly higher in men than women (P < 0.001). Women were significantly more likely to report low/no PA (45% vs. 28%), be overweight/obese (52% vs. 38%), and have DM (8% vs. 6%) compared with men. Women had lower perception of safety scores (ie, viewed neighborhood as more dangerous) and also had

Discussion

A better understanding of the factors that contribute to physical inactivity, overweight/obesity, and DM would facilitate the identification of approaches to arrest the rising levels of these outcomes and risk factors in developing countries, such as Jamaica. In this study, we examined (1) whether variability existed in PA, overweight/obesity, and DM across neighborhoods in Jamaica and (2) the impact of perceived safety, home disorder, neighborhood disorder, and recreational facility

References (45)

  • S.T. Broyles et al.

    Integrating social capital into a park-use and active-living framework

    Am J Prev Med

    (2011)
  • J. Mota et al.

    Perceived neighborhood environments and physical activity in adolescents

    Prev Med

    (2005)
  • Caribbean Commission on Health and Development. Report of the Caribbean Commission on Health and Development for the...
  • A.O. Abdulkadri et al.

    Economic burden of diabetes and hypertension in CARICOM states

    Soc Econ Stud

    (2009)
  • P. Hossain et al.

    Obesity and diabetes in the developing world—a growing challenge

    N Engl J Med

    (2007)
  • C.A. Monteiro et al.

    Socioeconomic status and obesity in adult populations of developing countries: a review

    Bull World Health Organ

    (2004)
  • D.R. Williams et al.

    Moving upstream: how interventions that address the social determinants of health can improve health and reduce disparities

    J Public Health Manag Pract

    (2008)
  • Wilks R, Younger N, Tulloch-Reid M, McFarlane S, Francis D. Jamaica Health and Lifestyle Survey 2007–08. 2008....
  • E.S. Ford et al.

    Geographic variation in the prevalence of obesity, diabetes, and obesity-related behaviors

    Obes Res

    (2005)
  • S.A. Prince et al.

    A multilevel analysis of neighbourhood built and social environments and adult self-reported physical activity and body mass index in Ottawa, Canada

    Int J Environ Res Public Health

    (2011)
  • B. Giles-Corti et al.

    Environmental and lifestyle factors associated with overweight and obesity in Perth, Australia

    Am J Health Promot

    (2003)
  • A. Ellaway et al.

    Graffiti, greenery, and obesity in adults: secondary analysis of European cross sectional survey

    BMJ

    (2005)
  • Cited by (28)

    • Who moves in vulnerable Caribbean neighborhoods? Positive deviance for physical activity: Findings from the Jamaica health and Lifestyle Survey 2017 (JHLS III)

      2022, Preventive Medicine Reports
      Citation Excerpt :

      Defining Vulnerable Neighborhoods. Given the inverse relationship between neighborhood crime and disorder with PAL (Cunningham-Myrie et al., 2015; Rees-Punia et al., 2018), vulnerable neighborhoods were those classified as being perceived by participants as having high crime and safety problems or physical or social disorder (Elo et al., 2009); based on summary scores for all subscales being at or above the 75th percentile. Perception of crime and safety problems was measured by seven questions regarding how worried the respondent was about crime and drug activity in their neighbourhood.

    • Perceived built environment and type 2 diabetes incidence: Exploring potential mediating pathways through physical and mental health, and behavioural factors in a longitudinal study

      2021, Diabetes Research and Clinical Practice
      Citation Excerpt :

      Although measurement error, recall bias, inadequate study design, and the long duration for T2D development are attributable for the modest evidence of mediation, only a few [29,37,39] employed causal mediation analysis (Supplementary Table S1). Most [27,28,30–36,38,40], but not all [29,37,39] studies also did not estimate the magnitude and significance of the mediated effect. None of the studies allowed for plausible exposure-mediator interaction which if incorrectly assumed, can lead to invalid inferences [16,41] and also almost all neither considered nor acknowledged the identification assumptions.

    View all citing articles on Scopus
    View full text