Evidence

Density and physical activity

Higher population and residential densities are critical in creating mixed-use neighbourhoods, because they provide the customers needed to support local businesses. Higher densities also generally result in more compact land use, thus decreasing the distances between homes and destinations. 

The viability of public transport also increases with greater population densities, with studies showing that the prevalence of using public transport (either for work or general purposes) is associated with higher residential density [22–24] and a greater availability of walkable destinations. [22] Indeed, studies have found that in low-density neighbourhoods with few local destinations, fewer people walk and more people drive. [25, 26] 

More recent evidence also points to a link between increased density, decreased automobile use and decreased sitting time, [22, 27–30] with easier access to destinations [22, 29] and better street connectivity [29, 30] also thought to play an important role. 

Studies have repeatedly shown that urban sprawl – as characterised by low densities, long and winding street networks, and separated land uses – decreases local walking and increases vehicle miles travelled.  [17, 25] In turn, this increases sedentary behaviour.  [31] 

There is a strong and consistent connection between higher residential density and mixed-use planning and walking, across all life stages.  [4–16] This is particularly evident for walking as a means of transport. Studies have found transport-related walking to be associated with increased residential density,  [23, 32–35] land use mix,  [32, 35] and access to services and destinations.  [34, 35] A recent, large, multi-country study not only confirmed this finding, but also revealed evidence of a potential threshold effect. According to the authors, walking was more likely when population density reached at least 7,500 dwellings/km2, and depending on the type of walking , as much as 12,000 dwellings/km2.  [36] 

Given that density was objectively measured and that the thresholds were comparable across a wide range of international study sites, the authors propose that their findings could potentially serve as a meaningful threshold point for future densification plans. 

A growing number of studies also support the association between higher residential density and adults’ engagement in active travel (most commonly walking and cycling).  [22, 27, 37–40] In some cases, the likelihood of active travel is 2.5 to 3 times higher for adults living in areas with higher residential density.  [38, 39] 

A similar trend can be observed among children and adolescents, with several studies identifying an association between higher residential density and active travel to and from school.  [41–43] Indeed, some evidence suggests that greater density around both the home and the school encourages habitual active transport to and from school.  [44] Other studies have demonstrated a relationship between children’s engagement in moderate to vigorous physical activity (MVPA) and increased residential and intersection density,  [45, 46] public transit density,  [45] commercial density, and number of parks.  [46] 

One notable exception to this trend, however, was found in a study of Canadian youths, in which urban sprawl (that is, lower density) was positively associated with active transportation and MVPA.  [47] As this finding contradicts the direction of association that has generally been reported for adults and children, it serves to reinforce the complexity of such relationships, and the need to study them across a range of different contexts and population groups. 

While not as comprehensive, there is further evidence associating higher residential density and mixed-use development with recreational walking, and with walking generally.  [9, 19, 48–51] For recreational walking, residential density and mixed-use development contribute to a varied and interesting environment, which can help entice walkers into the neighbourhood. 

However, recreational walking also tends to be associated with other neighbourhood factors, such as the presence nearby of high-quality, well-maintained public open space, and pleasing aesthetics. For example, a large, multi-country study found that greater residential density, land-use mix, street connectivity, proximity to parks, safety from crime, and pleasing neighbourhood aesthetics were all positively associated with adults’ recreational walking.  [52] 

The association for density was curvilinear, however, which means the likelihood of recreational walking increased up to a point, but then declined as density increased further. This could mean that in typical urban settings, density encourages recreational walking, but that in extremely dense settings – such as Hong Kong, one of the study sites – greater density may discourage recreational walking, due to increased congestion and a lack of opportunities for physical activity.  [52] 

Indeed, another recent study concluded that the social conditions of a neighbourhood are of vital importance, with indicators of physical disorder, which include graffiti and garbage, or a fear of crime having the potential to negate the benefits of increased density and land-use mix.  [53] 

Other evidence suggests that greater residential density and mixed land use may help prevent weight gain and obesity.  [54] Indeed, low-residential density is associated with men  [55] and children being overweight,  [56] while decreasing sprawl (indicating denser, more compact communities) has been associated with lower body mass index (BMI) among women.  [57] Growing evidence suggests that suburban sprawl is associated with increased weight,  [58] principally through a reduction in people’s energy expenditure by encouraging car dependence  [25] and more sedentary leisure behaviours.  [31] 

A study by Larry Frank and colleagues in the USA found that a person’s likelihood of being obese declined by 12.2 per cent for each quartile increase in mixed land-use and by 4.8 per cent for each additional kilometre walked, but conversely increased by 6 per cent for each hour spent in a car per day. [25] Physical activity is thought to be a significant way in which residential density influences a person’s weight. That is, denser, well-connected, walkable environments promote physical activity, with the resultant energy expenditure reducing body weight.  [56–57] 

Once again, the evidence indicates that density is not the only neighbourhood feature linked to weight, with a lack of parks or open spaces,  [56] increased traffic density, poorly maintained pathways,  [55–56] poor neighbourhood aesthetics, and increased distance to commercial facilities  [55] also being associated with increased weight or BMI. 

The ‘walkability index’ pioneered by Larry Frank and colleagues in the USA in 2009  [59] has been used to measure and quantify the presence of many features of the built environmental, by combining scores for variables that represent connectivity, density and land-use mix.  [59–62] Subsequent versions of walkability indices, including those developed in Australia, have modified the land-use mix computation by varying or adding new land-use categories. 

Evidence from several countries, including Australia, shows positive associations between neighbourhood walkability and walking for transport across the life course.  [10, 19, 63, 64] These indices highlight the combined influences of built environmental features and the importance of a combination of measures for creating active, walkable environments. Population, residential and employment densities underpin the provision, ease of access to, and effects of, diverse land uses and employment opportunities, public transport and destination. If density is too low, it is not possible to have mixed land uses, or nearby and accessible destinations and transit that encourage walking.