Table 3.
Multivariate analysis for LSR reduction of lumber spine and femoral neck BMD
| Nagelkerke R2: 0.10 | Lumber spine | |
|---|---|---|
| Odds ratio | p value | |
| Average dose of prednisolone | 1.22 (1.02–1.50) | 0.03 |
| Type of anti-osteoporosis drugs | 1.52 (0.87–2.65) | 0.14 |
| CRP | 1.46 (0.95–2.25) | 0.08 |
| Nagelkerke R2: 0.08 | Femoral neck | |
|---|---|---|
| Odds ratio | p value | |
| Sex | 2.37 (0.82–6.90) | 0.11 |
| Disease duration | 0.95 (0.91–0.10) | 0.04 |
| Average dose of prednisolone | 1.13 (0.97–1.32) | 0.11 |
The stepwise binomial logistic regression detected high dose of prednisolone use is a risk factors for 12-month LSR of lumbar spine (≤ − 2.4%). The stepwise binomial logistic regression analysis detected short disease duration is a risk factors for 12-month LSR of femoral neck (≤ − 1.9%)
LSR least significant reduction, BMD bone mineral density, CRP serum C-reactive protein concentration