Destination accessibility and walking for different purposes in older adults

Background: This thesis examined the associations between perceived destination accessibility of 12 types of destinations (supermarket, café/restaurant, fruit and vegetable shop, fast food restaurant, public transport, public park, post office, library, primary school, childcare centre, chemist/drug store and doctor/medical centres) within a 5-, 10- and 20- minute walk from home and self-report measures of walking for transport and recreation in older adults (people aged 65 years or older) residing in Brisbane, Australia and Hong Kong, China. The overall objective of this thesis was to investigate and compare the relationships of perceived destination accessibility in the neighbourhood with walking for transport and recreation in older adults living in low- and high-density urban environments. The overall thesis objective was divided into three principal aims and each aim was addressed by an empirical study with distinct rationales, aims and statistical methods. Study One addressed the first principal aim of the thesis. The primary aim of this study was to characterise perceived destination accessibility within a 5-, 10- and 20-minute walk from home in the two cities and examine between-city differences in perceived access to specific destinations and mixes of destinations. Study Two addressed the second principal aim. This study extended the findings from Study One by examining associations between perceived destination accessibility within a 5-, 10- and 20-minute walk from home and self-report measures of total (location nonspecific) walking for transport and recreation in older adults residing in the two cities. Further, this study examined associations of perceived destination accessibility with selfreport measures of within-neighbourhood walking for transport and recreation in older adults within the context of Hong Kong. Study Three investigated the third principal aim of the thesis. This study extended the findings from Study One and Study Two by examining the moderating effects of nine perceived neighbourhood non-destination characteristics (physical barriers to walking; pedestrian infrastructure; aesthetics; the presence of people; traffic hazards; traffic speed; safety from crime; sitting facilities; and presence of bridges/overpass) on associations of perceived destination accessibility within a 5-, 10- and 20-minute walk from home with self-report measures of walking for transport and recreation in older adults within the context of Hong Kong. Methods: This thesis used data from two extant epidemiological studies on environmental correlates of physical activity conducted in Brisbane (N= 793) and Hong Kong (N= 484) with comparable measures of 12 perceived destination accessibility and self-report measures of walking for different purposes. The Brisbane data came from the Wave 3 (2011) of a multilevel longitudinal study—the How Areas in Brisbane Influence HealTh and AcTivity (HABITAT)—among adults of 45-70 years in Brisbane (conducted in 2007-2011), while the Hong Kong data came from a cross-sectional study (the Hong Kong Elderly Study) among older adults in Hong Kong (conducted in 2007-2008). The two studies were based on the socio-ecological framework of health behaviour and used similar sampling strategies —that is, older adults nested within neighbourhood environments varying in environmental characteristics — that maximise the variability in exposures within the study sites. A range of analytical techniques were used to address the thesis aims, including a variable-centred approach (advanced regression models) and a person-centred approach (latent class/profile analysis). All models were adjusted to account for neighbourhood-level clustering arising from the two-stage sampling design. Stata 15.1 was used to perform the regression and moderation analyses, while Mplus 7.4 and 8.0 were used to perform latent class and latent profile analyses respectively. Results: The findings suggest that older adults living in Hong Kong perceived higher levels of destination accessibility within a 5-, 10- and 20-minute walk from home than older adults in Brisbane. City-specific latent structures of perceived destination accessibility varied between the two cities and also influenced walking behaviours in older adults. Further, perceived neighbourhood non-destination characteristics independently or conjointly moderated the destination-walking relationships in older adults within the context of Hong Kong. Conclusion: This thesis suggests that providing neighbourhoods with higher levels of destination accessibility can help encourage walking for different purposes, especially walking for transport in older adults. However, other perceived neighbourhood nondestination characteristics such as safety from crime, sitting facilities, pedestrian infrastructures, connectivity, aesthetics and the presence of people in the street can moderate destination-walking associations.