Table 7

Effect estimates from the model of length of stay in the CMP unit (n = 1225)

Variable
Multiplicative effect on LOS (95% CI)
P

Type of condition*


     Haemorrhage
1.00
<0.001
     Hypertensive
1.44 (1.29–1.61)

Age (per 10-year increase)
1.06 (0.98–1.14)
0.165
Surgical status*


     Surgical (admission from theatre)
1.00
0.036
     Nonsurgical
1.12 (1.01–1.24)

Past medical history presenta
1.04 (0.65–1.64)
0.883
Highest temperature (per 1°C increase)*
1.13 (1.06–1.20)
<0.001
SBP (per 10 mmHg deviation from normal)*
1.09 (1.05–1.13)
<0.001
Highest heart rate (per 10 beats/min increase)*
1.09 (1.06–1.12)
<0.001
Highest respiratory rate (per 10 breaths/min) increase
0.98 (0.91–1.06)
0.664
PaO2/FiO2 ratio (per 100 mmHg decrease)*
1.14 (1.09–1.19)
<0.001
Lowest pH (per 0.1 decrease)*
1.14 (1.06–1.23)
<0.001
Lowest sodium (per 5 mmol/l decrease)*
0.90 (0.84–0.97)
0.006
Potassium (per 1 mmol/l deviation from normal)
1.12 (0.94–1.33)
0.190
Highest creatinine (per 100 μmol/l increase)*
1.32 (1.20–1.45)
<0.001
Highest urea (per 5 mmol/l increase)
1.02 (0.94–1.11)
0.662
Lowest WBC (per 5 × 109/l decrease)*
1.12 (1.06–1.17)
<0.001
Lowest platelet count (per 50 × 109/l decrease)
1.01 (0.97–1.05)
0.563
GCS (per 1 point decrease)
1.01 (0.99–1.03)
0.370

aEvidence of any of a pre-specified list of severe conditions in the past medical history as defined in Acute Physiology and Chronic Health Evaluation (APACHE) II. CI, confidence interval; CPM, Case Mix Programme; GCS, Glasgow Coma Scale; LOS, length of stay; PaO2/FiO2, arterial oxygen tension; fractional inspired oxygen; SBP, systolic blood pressure; WBC, white blood count. *P < 0.05.

Harrison et al. Critical Care 2005 9(Suppl 3):S25   doi:10.1186/cc3542