Table 4.

The results of the two-piecewise linear model.

	Number (%)	OR (95%CI), P
Fitting model by standard linear regression	6249 (100%)	0.87 (0.79, 0.95) 0.0020
Fitting model by two-piecewise linear regression
The turning point of hemoglobin level (g/dl)
< 7.2 (slope 1)	321 (5.14%)	1.24 (0.92, 1.68) 0.1596
≥ 7.2( slope 2)	5928 (94.86%)	0.85 (0.78, 0.93) 0.0006
Slope 2 to slope 1		0.68 (0.51, 0.93) 0.0142
Predicted at 7.2		− 0.69 (− 0.80, − 0.59)
P for the log-likelihood ratio test		0.011

Model adjusted for: age; gender; HR; SBP; DBP; RR; AG; ALT, AST, bicarbonate, INR, total bilirubin; total calcium; creatinine; chloride; hematocrit; lactate; PLT; PT; TT; RBC; urea nitrogen; WBC; sodium; renal disease; CAD; diabetes; hypertension; SOFA; APAHCEII.

ALT, alanine aminotransferase; AST, aspartate aminotransferase; CAD, coronary artery disease; SBP, systolic blood pressure; DBP, diastolic blood pressure; HR, heart rate; RR, respiratory rate; WBC, white blood cells; PLT, platelet; RBC, red blood cells; PT, prothrombin time; TT, thrombin time; INR, international normalized ratio; AG, anion gap; SOFA, sequential organ failure assessment; APACHE, acute physiology and chronic health evaluation; OR, odds ratio; CI, confidential interval.