Research Article
Design of a Clinical Decision Support System for Fracture Prediction Using Imbalanced Dataset
Table 5
Association between prescribed oral or intravenous corticosteroid and fracture.
| | Fracture | value | No () | Yes () |
| Medication (oral or intravenous corticosteroids) | 10511 (90.3%) | 1134 (9.7%) | | Cumulative dose (mg) | | | <0.001 | Mean ± SD | 1345.77 ± 2831.30 | 2355.86 ± 4049.43 | Median | 390 | 810 | Mean follow-up daily dose (mg/d) (cumulative dose/follow-up duration) | | | 0.979 | Mean ± SD | 59.74 ± 218.39 | 59.56 ± 285.58 | Median | 4.54 | 3.275 | Mean exposed daily dose (mg/d) | | | 0.001 | (Cumulative dose/cumulative exposed duration) | | | Mean ± SD | 67.65 ± 235.62 | 94.04 ± 314.98 | Median | 13.45 | 15 | Duration (day), median (range) | | | | Follow-up duration | 706.77 ± 907.35 | 999.65 ± 997.09 | <0.001 | Median | 240 | 731.0 | Exposed duration | 90.11 ± 194.80 | 154.04 ± 320.91 | <0.001 | Median | 28 | 38.5 |
|
|