Table 4. Final Multivariate binary logistic regression for the occurrence of Parkinson’s disease.
| Variables | Coefficient | OR (95% CI) | p | |
|---|---|---|---|---|
| Final model | ||||
| Low frequency of toothbrushing / day) | 0.723 | 1.95 (1.39 – 2.96) | 0.006 | |
| Lower number of teeth | 1.837 | 2.21 (2.01 – 2.15) | <0.001 | |
| Bleeding on probing | 0.956 | 2.45 (1.62 – 2.45) | <0.001 | |
| Mean P. gingivalis counts | 1.230 | 3.39 (1.13 – 1.92) | 0.036 | |
| Periodontitis | 0.587 | 3.23 (1.92 – 5.90) | 0.003 | |
| Constant | -2.72 | - | <0.001 | |
OR = Odds ratio; 95%CI = Confidence interval. Final model: Pseudo R2 = 0.269; sensitivity = 74.81%; specificity = 85.34%; area under the ROC curve = 0.761. Significant p-values are shown in bold.