The built environment search identified 37 high quality reviews, 27 of which reported on outcomes relevant to this synthesis, of which 11 were systematic review, eight were meta-analyses, two were narrative reviews, and three reviews found no studies. Outcomes reported on most frequently included: injuries and safety (N=11), mental health (N=8), physical activity behaviour (N=7), and household air quality (N=5). Six of the 17 reviews included randomized controlled trails only, 14 included both RCTs and other types of studies (non-RCTs, quasi-RCTs, quasi experimental, controlled before-and-after studies, cross-over studies, controlled trials, or cohort studies), two reviews included only controlled before-and-after trials, five reviews include prospective cohort studies, as well as cross-sectional, interrupted times series, and/or retrospective cohort studies, two review did not clearly identify the types of studies included, and two reviews found no studies meeting their inclusion criteria. Twelve of the reviews scored 10 out of a possible 10 points on methodological quality, nine reviews scored nine, and six reviews scored eight. The majority of the reviews were conducted in the UK (N=15), followed by Canada (N=4), Australia (N=3), United States (N=2), and New Zealand (N=1). The number of primary studies included in the reviews ranged from no studies to 99. Based on the 27 included reviews, the populations varied greatly. Participants ranged from the general population to children, adults, older adults, and ethnic and low income populations. Settings included roadways, worksites, and homes. The interventions evaluated varied significantly across the reviews but can be classified into the following categories: traffic safety, occupational health, supportive housing, physical environment, falls prevention, home safety, child safety, and physical activity interventions.
Injuries & Safety
The evidence related to the built environment with respect to injury prevention included reviews evaluating traffic related injuries, work related injuries, and home related childhood injuries. One review also explored injury rates following privatization of industry (privatization of electrical/natural gas, mining companies).
Interventions to prevent and/or reduce the incidence of injuries to children occurring in the home were evaluated. Interventions included home safety education, as well as the provision of low cost, discounted or free equipment for increasing home safety practices, as well as parenting interventions. While home safety education increased safety behaviours among parents, fewer injuries among children in the home did not occur. For example, home safety education increased the proportion of families who had: safe hot tap water temperatures; functioning smoke alarms; fire guards; medicines, cleaning products and sharp objects stored out of reach; syrup of ipecac present in the home; poison control centre number accessible; fitted stair gates; and covered electrical sockets. However, home safety education and the provision of safety products was not associated with a reduction in thermal injuries, poisoning, and other medically attended or self-reported injuries among children, or in keeping hot drinks, food or matches out of reach; possession of a fire extinguisher; having window locks (although some evidence of effectiveness is observed for families with male children and older children); possession of non slip bath mats; using child restraints in cars; safe sleep practices. The evidence also suggests that parent education interventions are more effective for having poison control centre number accessible when one or more parent was not in paid employment, and when the home was not owner occupied for outcomes such as safe tap water temperatures, as well as the use of fitted stair gates.
In contrast, children of parents exposed to either a home visiting parenting program or a paediatric practice-based intervention compared to no intervention, experienced a lower risk of home related medically or non-medically attended injuries, burns, and scratches. There is also limited evidence demonstrating a reduction in emergency department visits following exposure to home visiting parenting programs. However, the findings must be interpreted cautiously given the outcome was measured using self-report.
The effectiveness of red-light cameras to reduce car crash casualties, the number of crashes, and red light violations was evaluated. While few studies were adequately controlled, the evidence suggests that red light cameras are an effective strategy in reducing total casualty crashes, but not right angle casualty crashes, total crashes or red light violations. The evidence also illustrates that the use of street lighting on roads previously unlit, or better lighting on roads previously lit, is associated with a statistically significant reduction in both total crashes and total injury crashes. Street lighting was found to reduce total crashes by 55% with the true value ranging from 31% to 71%, and total injury crashes by 22% with the true value ranging from a 3% to 37% reduction.
With respect to occupational health and safety interventions, generally the evidence suggests that interventions such as training plus a safety audit, training alone, and engineering interventions (i.e. ventilation systems) have mixed effects on work related injuries and illness. More specifically, some studies observed small positive effects, while an equal number of studies reported no impact. In particular the evidence illustrates that injury and illness prevention programs in the work setting are effective in reducing falls, preventing back injuries among nurses working on high risk floors, preventing eye injuries, and preventing hearing loss (only when hearing protection is mandatory rather than voluntary), but not in preventing logger injuries, hearing loss, and use of substances. The evidence is mixed with respect to repetitive strain injuries such as from computer use, with the tendency for there to be more studies reporting no effect on repetitive strain injuries in comparison to those reporting a significant reduction. Finally, there was some evidence that encouraging nurses with back injuries to engage in specific exercises was effective in reducing daily reported pain at 6 and 24 months in comparison to usual care (standard general practice physician).
With respect to privatization, the evidence did not illustrate an association between privatization and injury rates in any sector. For example, no difference in injury rates was reported where privatization had occurred in water, gas, electrical and mining industries.
Mental Health Outcomes
Evidence on the association between the built and natural environment and mental health outcomes has also been evaluated. For example, the evidence indicates there is a significant association between being born in an urban centre and developing schizophrenia. In addition, the findings suggest the greater the degree of urbanicity, the greater the risk of schizophrenia. Alternatively, living in sparsely populated areas is associated with higher suicide rates among males. The evidence from good quality studies is lacking to determine the impact of high household density on mental health outcomes among children and adults.
In instances where housing or neighbourhood regeneration is being implemented, there is some evidence of a positive effect on mental health outcomes among adults and male children, but not female children. Positive effects were observed in some studies as long as 2-3 years post intervention. In addition, there is some evidence, albeit not from rigorous studies, suggesting that poor neighbourhood and or housing quality is associated with poorer mental health outcomes and that access to green spaces is associated with better mental health outcomes. There is also evidence suggesting that interventions that target rehousing, refurbishment, and relocation are associated with improved mental and physical health outcomes in the longer term (18 months), although the majority of studies are not of high quality. Alternatively, in instances where the impact of witnessing crime or being a victim of crime was explored, the evidence illustrates poorer mental health outcomes among both adults and children. Similar results are reported for neighbourhood disorder. The relationship between population density and psychological outcomes was mixed, with some evidence reporting worse psychological outcomes for adults living in high density areas, and others reporting no association. With respect to noise, the evidence does not illustrate a relationship between aircraft noise and adverse mental health outcomes in children, while road traffic noise was found to be associated with greater risk of anxiety but not depression amongst adults.
Household Air Quality
Interventions to address household air quality include education about allergen exposure reduction, provision of allergen reduction equipment, energy efficiency measures, and dust control. Education about allergen exposure and the provision of allergen reduction equipment is associated with statistically significant reductions in physician diagnosed asthma in children, and number of days ill, but not asthma symptoms such as wheezing and lung function. The available evidence on dust mite control is not of high quality. Nonetheless, the evidence illustrates that the provision of mite impermeable bedding covers is associated with significant reductions in dust mite load. However education and dust control measures were not found to be effective in reducing dust levels in homes. Finally, there is a growing body of evidence to suggest that interventions targeting children's exposure to environmental tobacco smoke, particularly in the home, achieve some degree of success in reducing children's exposure to tobacco smoke. The evidence on energy efficiency measures is not of sufficient quality at this time to draw definitive conclusions on its impact on respiratory and other health outcomes.
Physical Activity Behaviours
Six reviews representing 101 studies explored the impact of built environment interventions on physical activity behaviours. One review did not find any studies meeting the author's inclusion criteria. Interventions evaluated include: travel behaviour change programs (i.e. aim to change travel behaviour from relatively inactive transport to more active transport - reduce single car use in favour of walking, cycling and/or public transport for travel to schools, tertiary institutions and workplaces); interventions to promote walking and cycling; environmental policy strategies (such as transit oriented development, street layouts, and the location of stores, jobs and school); point-of-decision prompts (such as motivational signs near stairwells); and interventions aimed to increase energy expenditure at work.
The evidence on travel behaviour change programs is mixed and is not of high methodological quality. While limited evidence exists illustrating a reduction in car use and an increase in walking as opposed to driving, an equal amount of evidence reports no impact on transport behaviour. There is limited evidence that worksite incentives (i.e. subsidy of employees who choose not to drive) have a positive impact on changing travel behaviour. However, the quality of many of the available studies does not allow for definitive conclusions on the impact of travel behaviour change programs to be identified at this time.
Interventions targeted at promoting walking and cycling generally (i.e. during leisure time), appear to having beneficial effects on behaviour. For example, brief, face-to-face counselling provided in the workplace, or by clinicians or exercise specialists in primary care was associated with increased self-reported walking at six weeks, and among some studies in the longer term. Interventions delivered via the telephone and/or internet to individuals, and those directed at groups through lay mentored meetings, led walks and group educational sessions were also associated with a statistically significant increase in walking. The use of pedometers to promote walking (i.e. 10,000 steps per day) was associated with a statistically significant increase in walking and/or step counts in the short term (4-12 weeks). However, these positive effects were not maintained after 24 weeks.
Community level interventions to promote walking produced mixed effects with some evidence suggesting a positive impact and others reporting no effect on walking. Furthermore, the most rigorous study evaluating community level interventions to promote walking reporting a positive effect also included a substantial mass media campaign.
Similar findings were observed for cycling. Interventions focused on promoting active communing (i.e. using bikes for transport to anywhere versus a car), along with educational activities and improving the cycle route network in cities, were found to increase the proportion of people cycling, the frequency of cycling per week and the distance travelled. However, particularly in relation to the proportion of people cycling and distance travelled, the overall effect size remains relatively small. While bike path usage increased with media and social marketing campaigns, and was sustained in the long term, this did not translate into increased population prevalence of cycling. In other words, those who cycled before the intervention, continued to cycle and utilized bike path systems consistently following their development, but the availability of bike paths did not lead to more people taking up cycling.
Currently there is limited evidence available to ascertain with any certainty if community-level cycling promotion activities promote sustained cycling, and at a level that achieves health benefit.
In terms of street-scale urban design and land use: improved street lighting or infrastructure projects that increase the ease and safety of street crossing, ensure sidewalk continuity, introduce or enhance traffic calming such as center islands or raised crosswalks, or enhance the aesthetics of the street area (i.e. landscaping) found positive effects on physical activity behaviour. While overall physical activity increased by 35%, the level of physical activity ranged from 16% to 62% and was based on moderate quality studies. It is important to note that all of the interventions were related to access, aesthetics, and safety.
Transportation and travel policies and practices (which strive to improve pedestrian, transit, and light rail access to increase pedestrian and cyclist activity and safety, reduce car use, and improve air quality), found no effect on physical activity behaviour. Interventions included creating and/or enhancing bike lanes, requiring sidewalks, subsidizing transit passes, providing incentives to car or van pool, increasing the cost of parking, and adding bicycle racks on buses.. This was based on one study of moderate methodological quality on the mode of choice for walking to school. However a statistically significant positive effective was not reported.
Point-of-decision prompts, such as motivational signs placed at or near stairwells, or at the base of elevators and escalators, encouraging people to use the stairs, was found to be effective. While the increase in stair use was small, 2.4% increase, it was found to be statistically significant in 15 of 21 studies. However, there was no significant relationship between baseline stair use and absolute change, or between the period of observation (leaving point-of-decision prompts in place and observed passersby for different lengths of time from 1 to 12 weeks) and relative change in stair use.
Workplace interventions to reduce sitting found no significant differences in workday sitting.