Does the Potential Benefit of Neighbourhood Green Space for Body Mass Index Depend upon Socioeconomic Circumstances and Local Built and Transport Environments? A Test of the ‘Equigenesis’ Hypothesis in Australia

Abstract

Background Recent studies have hypothesised that green spaces may help to narrow socioeconomic inequities in cardiovascular mortality and mental health (so-called “equigenesis”). To date, no testing of this hypothesis has been applied to body mass index (BMI), an important determinant of cardiovascular mortality and mental health. Nor have previous studies taken into account the potential for effect measure modification by measures of green space quality or other aspects of the built environment, such as public transport infrastructure. The aim of this study was to investigate the potential for equigenesis and other forms of effect measure modification with respect to BMI. Methods Multilevel models were used to analyze BMI in relation to an objective measure of neighbourhood green space quantity among 3684 mothers in wave 5 cohort k of the Longitudinal Study of Australian Children. A two-way interaction between green space quantity and neighbourhood disadvantage was fitted to examine the equigenesis hypothesis. Two-way interactions were also examined with respect to green space quantity and perceived indicators of green space quality, local services, access to public transport, presence of heavy traffic and indicators of area safety and social capital. Results Compared to mothers living in areas with 0 to 10% green space coverage, those in areas with 11% to 21% and >21% had lower mean BMI of -0.44kg/m2 (95%CI -0.96 to 0.07) and -0.90kg/m2 (95%CI -1.40 to -0.41), respectively. These associations were robust to controls for age, highest educational qualification, employment status, age and gender of the child, neighbourhood disadvantage and area remoteness. Green space quality was not associated with BMI and did not influence the association with green space quantity. Adjustment for other perceived neighbourhood variables similarly did not have an impact on the association between BMI and green space quantity. Fitting a two-way interaction between green space quantity and neighbourhood disadvantage did not result in meaningful evidence of effect measure modification. Conclusions The results from this study indicate that higher objectively-measured quantities of green space may be beneficial for BMI among women regardless of their socioeconomic circumstances. These findings do not support the equigenesis hypothesis. Green space quality may be less important than overall coverage. Perceived indicators of local built environment, such as busy roads and presence of public transport, similarly did not appear to modify the potential benefits of green space quantity for lower BMI.