Skip to main content
. 2007 Mar;97(3):509–514. doi: 10.2105/AJPH.2005.072447

TABLE 4—

Negative Binomial Regression Predicting Exercise Sessions per Week for Area Residents (n = 373): Los Angeles, Calif, December, 2003–May, 2004

Prediction of Exercise Sessions,a IRR (95% CI)
Residential distance to park
    > 1 mile (Reference)
    ≥ 1 mile 1.38* (1.04, 1.84)
Age (18–84 y) 0.98** (0.97, 0.99)
Gender (male = 1) 1.28* (1.01, 1.62)
Race/ethnicity
    Latino (Reference)
    African American 1.12 (0.84, 1.50)
    Other/Asian/White 1.25 (0.80, 1.94)
Indicator of safety (safe = 1) 0.93 (0.67, 1.28)
Parkb
    3 0.91 (0.60, 1.39)
    4 0.85 (0.55, 1.33)
    5 1.25 (0.94, 1.66)
    6 (Reference)
    7 1.20 (0.71, 2.00)
    8 0.95 (0.66, 1.38)

Note: IRR = incidence rate ratio; CI = confidence interval.

aIncreases in area residents’ exercise sessions per week.

bSurvey respondents of Parks 1 and 2 were not asked about the number of exercise sessions per week, so they were not included in the sample.

* P < .001; ** P < .05.