Skip to main content

Table 4 Hierarchical logit model for predictors of readmission within 30 days among trauma patients 65 and older admitted for single-level fall

From: Hospitalization and readmission after single-level fall: a population-based sample

 

Odds ratio

95% CI

P

Trauma mortality prediction model§

1.03

1.01–1.05

 

Elixhauser comorbidity index for readmission

1.03

1.03–1.03

 

Female

1.29

1.26–1.32

 

Age, years

1.007

1.005–1.009

 

Palliative care consult

0.41

0.37–0.44

 

Injury categories

   

 Pelvis fracture

0.90

0.86–0.94

 

 Sprain or strain

0.87

0.80–0.94

 

 Upper extremity fracture

1.04

1.00–1.08

 

 Traumatic brain injury, ± coma

1.31

1.27–1.35

 

 Kidneys, ureter, bladder, urethra

1.25

1.04–1.50

†

 Eyes, ears, nasopharynx, oropharynx, teeth

0.87

0.81–0.93

 

 Vascular injuries

1.42

1.13–1.78

†

Comorbid conditions

   

 Alcohol abuse

0.88

0.83–0.93

 

 Deficiency anemia

0.88

0.86–0.91

 

 Cancer, lymphoma

0.81

0.73–0.90

 

 Cancer, metastatic

0.71

0.65–0.77

 

 Cerebrovascular disease

0.91

0.86–0.95

 

 Dementia

1.05

1.03–1.09

 

 Depression

0.92

0.89–0.95

 

 Diabetes mellitus without complications

0.94

0.91–0.97

 

 Drug abuse

0.85

0.78–0.93

 

 Neurological disorders affecting movement

1.05

1.01–1.10

†

 Neurological disorders, other

0.15

0.14–0.16

 

 Seizures and epilepsy

1.11

1.05–1.16

 

 Obesity

1.22

1.18–1.27

 

 Psychoses

0.89

0.84–0.94

 

Major therapeutic procedure

0.97

0.96–0.99

 

Minor therapeutic procedure

1.04

1.03–1.05

 

Discharge disposition

   

 Routine

Referent

Referent

 

 Transfer to short-term hospital

2.50

2.25–2.78

 

 Skilled nursing, intermediate care

1.35

1.30–1.40

 

 Home health care

1.06

1.01–1.10

 

 Left against medical advice

2.18

1.86–2.55

 
  1. *P value < 0.001 for all comparisons except where noted
  2. **0.01 < P value < 0.001
  3. †0.05 < P value < 0.01
  4. §Logit transformation of trauma mortality prediction model probability of death (pDeath) = \(\log ( \, p{\text{Death /}}(1 - p{\text{Death}}))\)