Table 3.

Cumulative incidence of patients with ≥ 1 pooled clinical fractures (clinical vertebral and non-vertebral) and adjusted hazard ratios estimated with an extended Cox regression model at different time points of follow-up

	Teriparatide (n = 680)	Risedronate (n = 680)	HR (95% CI)	p value
At 3 months	6 (0.9%)	7 (1.0%)	0.70 (0.36, 1.37)	0.299
At 6 months	12 (1.8%)	17 (2.6%)	0.62 (0.37, 1.07)	0.085
At 7 months	12 (1.8%)	23 (3.6%)	0.60 (0.36, 0.99)	0.046
At 8 months	13 (2.0%)	25 (3.9%)	0.58 (0.36, 0.93)	0.024
At 9 months	14 (2.1%)	26 (4.0%)	0.56 (0.35, 0.88)	0.012
At 12 months	18 (2.8%)	34 (5.3%)	0.49 (0.31, 0.78)	0.003
At 15 months	22 (3.5%)	39 (6.1%)	0.44 (0.26, 0.76)	0.003
At 18 months	26 (4.1%)	49 (7.8%)	0.39 (0.20, 0.77)	0.007
At 21 months	27 (4.3%)	55 (8.8%)	0.35 (0.15, 0.81)	0.014
At 24 months	30 (4.8%)	61 (9.8%)	0.31 (0.11, 0.86)	0.024

BMD, bone mineral density; CI, confidence interval; HR, hazard ratio (teriparatide versus risedronate). Adjusted by treatment, antecedent of recent clinical vertebral fracture, recent use of bisphosphonate, age, baseline femoral neck BMD, baseline 25-hydroxy-vitamin D concentration, geographical region, and time-varying covariate treatment-by-time interaction