Table 2

Summary of results from retrospective longitudinal studies (n=8)

Walking*Physical activity*Other travel behaviourConsistency score
TransportRecreationTotalCyclingPublic transportDrivingUnweighted +%Weighted +%
Recreation environment
 Parks/green space/recreation facilitiesAditjandra et al,47 (+)1/1 (100%)1/1 (100%)
Neighbourhood design
 Land use mix/destinationsMcCormack (+)McCormack (0)McCormack (0)1/3 (33%)1/3 (33%)
Transportation environment
 Overall transportation accessAditjandra et al,47 (+);
Cao et al,50 (+);
Handy et al,40 (+);
Handy et al,41 (+)
Handy et al,42 (+)Handy et al,41 (+)Aditjandra et al,47 (+)Cervero and Day, 66 (0,–)§;
Handy et al,40 (0)
7/10 (70%)5/7 (71%)
AestheticsAditjandra et al,47 (+);
Handy et al,40 (+);
Handy et al,41 (+)
Handy et al,42 (+)Handy et al,40 (0)4/5 (80%)3/4 (75%)
Crime-related safetyAditjandra et al,47 (+);
Cao et al,50 (+);
Handyet al,40(+);
Handy et al,41 (+)
Handy et al,42 (+)Handy et al,40 (+)6/6 (100%)4/4 (100%)
Social environmentAditjandra et al,47 (0);
Cao et al,50 (+);
Handy et al,40 (+);
Handy et al,41 (+)
Handy et al,42 (0)Handy et al,41 (+)Handy et al,40 (0)4/7 (57%)2/4 (50%)
Aggregated characteristics
  SprawlKlinger and Lanzendorf,64 (0)Klinger and Lanzendorf,64 (0)Klinger and Lanzendorf, (0)640/3 (0%)0/3 (0%)
  Overall accessibility (to destinations and transportation)Handy et al,40 (+);
Handyet al,41 (+)
Aditjandra et al,47 (+)Handy et al,40 (+)4/7 (57%)3/6 (50%)
Consistency score
  Unweighted +%18/19 (95%)3/5 (60%)2/4 (50%)2/3 (67%)2/8 (25%)
  Weighted +%10/11 (91%)3/5 (60%)2/4 (50%)2/3 (67%)2/7 (29%)
  • Bold entries denote objectively measured environmental attributes.

  • *The columns for recreation walking and transport physical activity were omitted because no study examined these outcomes.

  • †Unweighted consistency score: the percentage of associations coded ‘+’ out of the total number of associations; weighted consistency score: applied weighting to results from the same study by a factor of 1/total number of results from the same study in one cell. For this table, data from refs.40–42 50 were from the same study.

  • ‡Overall transportation access: access to a range of specific or non-specific transportation options, such as sidewalks, bike paths, public transport and roads.

  • §A study can contribute to more than one finding to each combination of built environment attribute and outcome when it involved different exposure measures for the same category of environmental attribute or different measures for the same domain of outcomes.

  • ¶Indirect effects mediated by car ownership.

  • +, statistically significant associations in the expected direction; –, statistically significant associations in the unexpected direction; 0, non-significant associations (expected direction is detailed in online supplementary table 2).