Table 2.

Negative binomial regression analysis of 384 women with 309 falls for the 12-month follow-up

			Exp (B)a	95% CIb	p valuec
Crude valuesd
Study group
Intervention			0.84	0.61–1.14	0.26
Control			1
Fall risk category	Falls per person	Injury per fall
High fall risk (9–14p)	1.38	0.82	3.00	0.97–9.22	0.056
Substantial fall risk (6–8p)	1.73	0.53	4.00	1.93–8.30	< 0.001
Moderate fall risk (1–5p)	0.64	0.58	1.47	0.74–2.91	0.27
Low fall risk (0p)	0.45	0.54	1
Adjusted valuesd
Study group
Intervention			0.84	0.61–1.15	0.27
Control			1
Fall risk category
High fall risk (9–14p)			3.36	0.92–12.2	0.067
Substantial fall risk (6–8p)			3.77	1.77–8.04	0.001
Moderate fall risk (1–5p)			1.52	0.76–3.05	0.237
Low fall risk (0p)			1
Covariates
Single leg stance (seconds)e			0.995	0.980–1.010	0.49
Leg extension strength (Newtons)f			1.002	1.000–1.004	0.095
Grip strength (Newtons)g			1.003	0.999–1.008	0.13

^aExp(B) = Exponentiated values of the coefficients

^bCI = 95% confidence interval

^cp value < 0.05 was considered statistically significant

^dResults are shown for the crude values and for adjusted model with physical performance results as covariates

^eEyes open with the better foot, time measured continuing for a maximum of 30 s

^fMean of the best result for both legs from three attempts

^gHandheld dynamometer (Jamar, Sammons-Preston, Illinois, USA) the dominant hand with three attempts, the best result was used in the analysis