Table 2.
Fracture rates
| GR vs. other patients | |||||||
|---|---|---|---|---|---|---|---|
| GR patients | Other patients | ||||||
| n = 2,726 | n = 2,726 |
IRR (95% CI)a,b Unadjusted |
|||||
| Total number of fractures | Total patient-years of observation | IR of fractures, per 1,000 patient-years | Total number of fractures | Total patient-years of observation | IR of fractures, per 1,000 patient-years | ||
| Any site | 409 | 11,804 | 34.65 | 540 | 12,818 | 42.13 | 0.83 (0.70–0.97)* |
| Hip | 100 | 11,804 | 8.47 | 114 | 12,818 | 8.89 | 0.98 (0.69–1.40) |
| Pelvis | 28 | 11,804 | 2.37 | 43 | 12,818 | 3.35 | 0.71 (0.41–1.23) |
| Spine | 128 | 11,804 | 10.84 | 194 | 12,818 | 15.13 | 0.71 (0.54–0.95)* |
| Wrist/arm | 180 | 11,804 | 15.25 | 215 | 12,818 | 16.77 | 0.91 (0.73–1.13) |
| GR vs. alendronate patients | |||||||
| GR patients | Alendronate patients | ||||||
| n = 1,807 | n = 1,807 |
IRR (95% CI)a,b Unadjusted |
|||||
| Total number of fractures | Total patient-years of observation | IR of fractures, per 1,000 patient-years | Total number of fractures | Total patient-years of observation | IR of fractures, per 1,000 patient-years | ||
| Any site | 262 | 7,712 | 33.97 | 354 | 8,323 | 42.53 | 0.81 (0.66–0.98)* |
| Hip | 71 | 7,712 | 9.21 | 80 | 8,323 | 9.61 | 0.99 (0.65–1.51) |
| Pelvis | 16 | 7,712 | 2.07 | 26 | 8,323 | 3.12 | 0.68 (0.35–1.33) |
| Spine | 83 | 7,712 | 10.76 | 132 | 8,323 | 15.86 | 0.69 (0.49–0.97)* |
| Wrist/arm | 112 | 7,712 | 14.52 | 132 | 8,323 | 15.86 | 0.91 (0.70–1.20) |
Notes:
CI confidence interval, IR incidence rate, IRR incidence rate ratio
aIncidence rate ratios were estimated with generalized linear models with log link and negative binomial distribution with robust error variances to account for the matched pairs. Incidence rate ratios may differ from the ratio of the incidence rates because of adjustments for overdispersion
bAn IRR < 1 indicates a lower incidence of fracture per patient-year among women treated with risedronate GR
*Significant at the 5% level