Neighbourhood walkability and home neighbourhood-based physical activity: an observational study of adults with type 2 diabetes

Table 2 Regression estimates for the association between neighbourhood walkability and neighbourhood-based total VeDBA (n = 97)

	Percent change in one SD of total VeDBA (95 % confidence intervals)^a,b	Corresponding change in daily steps (95 % confidence intervals)^c
Model 1	21.2 (12.8 to 29.6)	337 (203 to 470)
Model 2	17.6 (9.3 to 26.0)	280 (148 to 413)
Model 3	13.9 (5.2 to 22.6)	221 (83 to 359)
Model 4	10.0 (0.7 to 19.3)	159 (11 to 307)
Model 5	10.4 (1.2 to 19.7)	165 (19 to 312)

^aModel 1: Unadjusted. Model 2: Adjusted for age, BMI, sex. Model 3: Adjusted for age, BMI, sex, university, and season. Model 4: Adjusted for age, BMI, sex, university, season, car access and residential self-selection. Model 5: Adjusted for age, BMI, sex, university, season, car access, residential self-selection and valid wear-time
^bEstimates represent the percent change in one standard deviation of total VeDBA (95 % confidence interval) occurring within home neighbourhoods (excluding homes) for every one-standard deviation increase in the GIS-derived neighbourhood walkability index. Calculated by multiplying the original estimate by the standard deviation of the walkability index (i.e., 2.16), dividing the result by the SD of the outcome (i.e., 39,149.22) and multiplying by 100
^cCalculated using the following formula: daily steps = −548 + 0.0089*total VeDBA occurring anywhere)*(% change in one SD of VeDBA occurring in neighbourhood/100) where VeDBA occurring anywhere equals one SD of VeDBA occurring anywhere (i.e., 240,065.36)

ISSN: 1471-2458