In-hospital fall prediction using machine learning algorithms and the Morse fall scale in patients with acute stroke: a nested case-control study

Table 3 Final logistic regression model after stepwise backward elimination

Variables	Odds ratio (95% Confidence interval)	p-value
Body mass index	0.95 (0.89–1.01)	0.092
Female	1.34 (0.91–1.95)	0.136
Albumin	1.47 (0.98–2.21)	0.059
Ward type
General	Reference
Integrated nursing care service	0.41 (0.27–0.63)	< 0.001
Intensive care unit/stroke unit	4.93 (2.59–9.41)	< 0.001
Admission route
Emergency department	Reference
Outpatient clinic	0.60 (0.34–1.08)	0.091
Admission state
On foot	Reference
Wheelchair	1.98 (1.06–0.36)	0.032
Bed	1.33 (0.68–2.60)	0.405
Hypertension
Diabetes	1.75 (1.14–2.67)	0.010
Dyslipidemia	0.60 (0.39–0.91)	0.015
Arrhythmia	3.20 (1.77–5.78)	< 0.001
Degenerative spinal diseases	4.30 (1.95–9.49)	< 0.001
Cerebral neurodegenerative diseases	3.17 (1.62–6.20)	< 0.001
Medication
None	Reference
Single	2.67 (1.75–4.07)	< 0.001
Poly	2.10 (1.12–3.94)	0.020
Alert mental status	0.35 (0.19–0.64)	< 0.001

ISSN: 1472-6947