Effects of neighbourhood environment on physical activities in gated and non-gated neighbourhoods in Karachi, Pakistan

Abstract

While it has been established that neighbourhood environment influences physical activity, gated neighbourhoods, which are a relatively new phenomenon, have been reported to affect the whole urban morphology with their distinct built and social environment as well as the number of physical activity facilities (PAF). However, limited research has been carried out on the effects of neighbourhood environment of gated neighbourhoods on physical activity (PA), especially in the context of developing countries. The aim of this study is to examine the effects of neighbourhood environment (built and social), PAF, and demographics (age, gender, and employment status) on PA in gated and non-gated neighbourhoods of Karachi, Pakistan. Data were gathered using the quantitative research methods: the neighbourhood environment indicators (walkability index, land-use mix (LUM), streets connectivity, housing density, perception of crime (POC), perception of traffic hazard(POTH)), PAF, and demographics were treated as independent variables, whereas physical activity variables (vigorous physical activity (VPA), moderate physical activity (MPA), and walking) were dependent. Data on LUM, streets connectivity, and housing density were collected through objective methods using GIS database and Space Syntax technique, while the number of PAF was enumerated. The data on POC, POTH, PA variables, and demographics were collected through questionnaire. A total of sixteen neighbourhoods eight for each neighbourhood type were studied, involving a total of 1042 participants. The collected data were statistically analysed by using SPSS. A walkability index was developed using data on LUM, streets connectivity, and housing density. The differences between variables of gated and non-gated neighbourhoods were analysed through independent sample t-test. The association of neighbourhood environment indicators and demographics with PA was measured through binary logistic regression. The comparison of walkability index with walking and PAF with VPA were done through -tests. Results show that the walkability index, LUM, streets connectivity, POC and walking were higher in non-gated neighbourhoods while the numbers of PAF, POTH, total PA and VPA were higher in gated neighbourhoods. Results also show that there was a significant positive relationship of walkability index and numbers of PAF with PA, but a significant negative relationship of age and employment status with PA. Other results show that the PAF had association with VPA in both types of neighbourhoods while walkability index had association with walking only in non-gated neighbourhoods. It is concluded that PAF is an important attribute to increase VPA, and consequently total PA; therefore more PAF should be built in non-gated neighbourhoods as well. Gated neighbourhoods have less LUM and street connectivity, and consequently less walkability, therefore, gating the communities should not be encouraged at policy level.