Table 3 Results of linear and logistic regression analyses to compare the primary outcome between the bicycle accident (n = 102) and non-bicycle accident (n = 773) groups

Length of hospital stay

 From linear regression analysis, number of admission days

  Model without covariates:

   Fracture caused by bicycle accident

0.88^a

0.78, 0.99

0.030

  Model with covariates:

   Fracture caused by bicycle accident

0.91^a

0.81, 1.03

0.125

   Age (reference: one year younger)

1.13^a

1.01, 1.02

< 0.001

   Females (reference: males)

0.92^a

0.85, 1.00

0.056

Hospital discharge before 8 days

 From logistic regression analysis, OR^b

  Model without covariates:

   Fracture caused by bicycle accident

0.64

0.42, 0.97

0.034

  Model with covariates:

   Fracture caused by bicycle accident

0.73

0.47, 1.12

0.148

   Age (reference: one year younger)

1.05

1.02, 1.07

< 0.001

   Females (reference: males)

0.78

0.58, 1.06

0.115