Skip to main content

Table 2 Eight bivariate logistic regressions analyses for calculating odds ratio for fracture predictors, and two multiple logistic regression analyses

From: Mandibular cortical bone structure as risk indicator in fractured and non-fractured 80-year-old men and women

Models

OR (95% CI)

Coefficient

P

I. Fracture/severe cort. erosion

2.32 (1.3–4.2)

0.846

0.0045

II. Fracture/thickness < 3 mm

2.00 (1.1–3.6)

0.692

0.0206

III. Fracture/bisphosphonate

5.57 (2.1–15.0)

1.717

0.0007

IV. Fracture/calcium

3.92 (2.0–7.8)

1.366

0.0001

VI. Fracture/corticosteroid

0 (0–> 1012)

− 12.489

0.9704

VII. Fracture/BMI

0.93 (0.91.0)

− 0.073

0.0516

VIII. Fracture/sex

2.36 (1.3–4.2)

0.857

0.0036

I. Fracture/erosion/bisphos

 Severe cortical erosion

1.95 (1.13.6)

0.670

0.0304

 Bisphosphonate medication

4.46 (1.612.3)

1.495

0.0040

II. Fracture/thickness/bisphos

 Cortical thickness < 3 mm

1.65 (0.93.1)

0.502

0.1101

 Bisphosphonate medication

4.68 (1.7–12.9)

1.543

0.0029

  1. Eight fracture risk factors were tested: severely eroded cortex with heavy endosteal cortical porosities (Fig. 1), a cortical thickness < 3 mm (Fig. 2), bisphosphonate medication, calcium, corticosteroid, BMI and sex