One size doesn’t fit all: cross-sectional associations between neighborhood walkability, crime and physical activity depends on age and sex of residents

Table 2 The main effect model of zero-inflated regression models^a of moderate-to-vigorous physical activity (minutes/day)^b as a function of percent green space, crime, and walkability index

	β (SE)	P
Neighborhood-level characteristics
Crimes in 2012^c	<0.01 (0.00)	0.53
Percent green space^d	−1.97 (3.17)	0.54
Walkability index^e	0.07 (0.04)	0.05
Individual-level covariates
Age (years)	−0.10 (0.00)	0.05
Male	0.89 (0.16)	0.00
Household annual income (per $1000)	<0.01 (0.01)	0.61
Education some college/bachelors versus less than college	−0.31 (0.15)	0.05
Married or living with partner versus living alone	−0.15 (0.15)	0.32
Any physical limitation walking 1 block	−0.62 (0.20)	<0.01
Car owner or access to a car	−0.15 (0.16)	0.32

Bold indicates statistical significance at p < 0.05
^a Zero-inflated negative binomial regression with “any limit walking 1 block”, “age”, and “male” in the inflate statement. Estimates are not presented but all were statistically significant at p < 0.001
^b Average time moderately-to-vigorously physically active based on 4–7 days of accelerometry data and a bout of at least 10 min of activity above the 100 mg threshold [41], where at least 80% of the bout was above the threshold of 100 mg
^c Counts of crimes in 2012 within 1 km network buffer of residence obtained from Pittsburgh Police Department
^d Percent green space defined by percent of area with green space within 1 km network buffer of residence
^e The walkability index [48] was composed of the following items: traffic signs at the intersection (4 points), pedestrian crossings (2 points), sidewalks (10 points), lighting (2 points), transit (2 points), and missed use (2 points). Items were summed for each street segment and the average of the summed items across the street segments constituted the walkability index. The index ranges from 0 to 22

ISSN: 1471-2458