Table 2.
Significant variables in the logistic regression predicting a psychiatric admission on the basis of the analysed total population.
Variable | Beta | S.E. | Wald | df | p-value | OR | Cl (95%) |
Suicidality | 2.470 | 0.213 | 134.66 | 1 | 0.000 | 11.81 | 7.78–17.93 |
Diagnosis of dementia | 0.672 | 0.316 | 4.51 | 1 | 0.034 | 1.95 | 1.05–3.63 |
Female sex | 0.252 | 0.116 | 4.72 | 1 | 0.030 | 1.28 | 1.02–1.61 |
Two diagnostic measures | 0.513 | 0.239 | 4.61 | 1 | 0.032 | 1.67 | 1.04–2.66 |
One consultation | 0.476 | 0.156 | 9.35 | 1 | 0.02 | 1.61 | 1.18–2.185 |
Two consultations | 1.281 | 0.300 | 18.22 | 1 | 0.000 | 3.60 | 2.00–6.485 |
Care order in force | 0.528 | 0.224 | 5.54 | 1 | 0.019 | 1.69 | 1.09–2.633 |
Referral by a doctor | 0.809 | 0.221 | 13.37 | 1 | 0.000 | 2.24 | 1.45–3.462 |
Medication in the emergency department | 1.094 | 0.217 | 25.43 | 1 | 0.000 | 2.98 | 1.95–4.566 |
Patient reffered by a consultant | 1.273 | 0.158 | 65.17 | 1 | 0.000 | 3.57 | 2.62–4.863 |
-2 Log likelihood 1904.547, Cox & Snell R Square 0.387, Nagelkerke R Square 0.520