Influence of perceived neighbourhood environment and health behaviours on body mass index

Abstract

The obesity phenomenon in Malaysia is becoming increasingly worrying as it is closely linked to deteriorating health consequences. There has in fact been a significant growth in the literature identifying neighbourhood-level environment concerning food and built environment as influential factors of health-related outcomes. However, both of the environmental attributes mentioned have been rarely explored collectively in Malaysia to explain health behaviours and outcomes. Additionally, past studies tended to be based on objectively-measured data while ignoring subjectively-measured ones through self-reported perception of the environment. Self-reported perceptions were found to be as influential as objectively-gathered field data in studies of public health. As such, this study aimed to explain the obesity phenomenon through perceived neighbourhood environmental factors, health behaviours and Body Mass Index (BMI). Participants of the study comprised of 256 adults above the age of 18 years old in the district of Johor Bahru, Johor from 22 higher and 21 lower socioeconomic status neighbourhoods selected using the stratified random sampling technique. Data were collected via a 107 question-item questionnaire administered face-to-face with the assistance of six qualified enumerators. The data were analysed using Statistical Package for Social Sciences version 22.0 (SPSS v.22) and IBM SPSS Amos version 23.0 (AMOS v.23). Confirmatory Factor Analysis was then used to produce the measurement model, while Structural Equation Modelling was used to compute the causal model due to its suitability for quantifying latent variables. Analysis of the data revealed that the causal model linking perceived neighbourhood environment, health behaviours and BMI was a good fit (?²= 2164.5, df= 1356, ?²/df= 1.596, TLI= .903, CFI=.914, RMSEA=.048). Both Perceived Neighbourhood Food Environment (PNFE) (ß= -.303, r²= .092) and Perceived Neighbourhood Built Environment (PNBE) (ß= -.208, r²= .043) were found to have significant direct effects on BMI. Consistent with past studies, the former had a higher influence on BMI. As expected, Diet Quality (DQ) (ß= -.142, r²= -.142) and Physical Activity (PA) (ß= -.169, r²= -.169) were significantly associated with BMI as weight is basically the result of energy intake and energy expenditure. Other variables influencing BMI includes monthly household income, gender and education level. The study also found that full mediation was observed when PNFE and PNBE were tested as mediators between the relationship of DQ and PA to BMI. This suggests that the environmental factors were highly significant as predictors of BMI, warranting further exploration in future studies. Subsequently, a causal model explaining the obesity phenomenon from the aspects of perceived neighbourhood environment and health behaviours was developed. The model exemplifies the intimate implications that urban and regional planning has on public health. This study serves as a basis for inter-sectoral future research by linking more environmental-related variables to public health.