Multivariable analysis of risk factors for fracture
| Risk factor . | Events/N . | Univariable (complete cases) . | Multivariable* . | |||
|---|---|---|---|---|---|---|
| HR (95% CI) . | P . | HR (95% CI) . | P . | |||
| Bone-protective medication† | ||||||
| No predisposing history‡ | 60/603 | 1.00 | .0022 | 1.00 | .0042 | |
| Predisposing history, no medication | 11/56 | 2.11 (1.08-4.12) | 2.12 (1.08-4.16) | |||
| Predisposing history and medication | 12/43 | 3.02 (1.58-5.76) | 2.84 (1.49-5.43) | |||
| DLBCL bone involvement | ||||||
| None | 46/559 | 1.00 | .001 | 1.00 | .0026 | |
| Involvement, no fracture | 18/94 | 2.23 (1.27-3.90) | 2.28 (1.29-4.03) | |||
| Involvement, fracture | 10/37 | 3.31 (1.65-6.63) | 2.76 (1.37-5.58) | |||
| Prephase steroid | ||||||
| No | 43/415 | 1.00 | .0320 | 1.00 | .021 | |
| Yes | 24/139 | 1.73 (1.05-2.84) | 1.84 (1.11-3.05) | |||
| Risk factor . | Events/N . | Univariable (complete cases) . | Multivariable* . | |||
|---|---|---|---|---|---|---|
| HR (95% CI) . | P . | HR (95% CI) . | P . | |||
| Bone-protective medication† | ||||||
| No predisposing history‡ | 60/603 | 1.00 | .0022 | 1.00 | .0042 | |
| Predisposing history, no medication | 11/56 | 2.11 (1.08-4.12) | 2.12 (1.08-4.16) | |||
| Predisposing history and medication | 12/43 | 3.02 (1.58-5.76) | 2.84 (1.49-5.43) | |||
| DLBCL bone involvement | ||||||
| None | 46/559 | 1.00 | .001 | 1.00 | .0026 | |
| Involvement, no fracture | 18/94 | 2.23 (1.27-3.90) | 2.28 (1.29-4.03) | |||
| Involvement, fracture | 10/37 | 3.31 (1.65-6.63) | 2.76 (1.37-5.58) | |||
| Prephase steroid | ||||||
| No | 43/415 | 1.00 | .0320 | 1.00 | .021 | |
| Yes | 24/139 | 1.73 (1.05-2.84) | 1.84 (1.11-3.05) | |||
Variables included: age, sex, Eastern Cooperative Oncology Group performance status (0, 1, 2, and 3-4), BMI group, smoking status, type 2 diabetes, alcohol use history, previous steroids within 12 mo, alkaline phosphatase, calcium (low, normal, or high), bone-protective medication (4 groups including comorbidities as in supplemental Table 3), bone involvement (3 groups as per univariable analysis table), prephase steroid, and non-lymphoma fracture at baseline. Stepwise backward selection with a cutoff of 0.05 was used.
Bone-protective medication includes vitamin D, calcium, and/or bisphosphonates.
Predisposing history: known osteoporosis, osteopenia, RA, or prior fracture.
Multivariable using complete cases from the backward selection model.