Construction of a risk prediction model for early postoperative recurrence in stage II/III colorectal cancer

Table 2 Univariate and multivariate logistic regression analysis for colorectal cancer patients in the training set

Variables	Univariate logistic regression analysis					Multivariate logistic regression analysis
	β	SE	Z	P value	OR (95%CI)	β	SE	Z	P value	OR (95%CI)
Size	0.9	0.16	5.67	< 0.001	2.45 (1.80-3.34)	1.43	0.36	3.93	< 0.001	4.17 (2.05-8.50)
ACJJ N
N0	-	-	-	-	1.00 (reference)	-	-	-	-	1.00 (reference)
N1	0.97	0.63	1.54	0.124	2.63 (0.77-9.05)	1.03	1.12	0.93	0.355	2.81 (0.32-24.98)
N2	2.49	0.61	4.1	< 0.001	12.10 (3.67-39.92)	3.59	1.12	3.19	0.001	36.05 (3.98-326.47)
NLR	1.29	0.25	5.23	< 0.001	3.64 (2.24-5.91)	1.33	0.39	3.43	< 0.001	3.80 (1.77-8.14)
PLR	0.01	0	1.89	0.059	1.01 (1.00-1.01)	0	0.01	0.7	0.483	1.00 (0.99-1.02)
Fatigue
No	-	-	-	-	1.00 (reference)	-	-	-	-	1.00 (reference)
Yes	0.6	0.43	1.41	0.158	1.82 (0.79-4.20)	-1.08	0.88	-1.22	0.221	0.34 (0.06-1.92)
Appetite loss
No	-	-	-	-	1.00 (reference)	-	-	-	-	1.00 (reference)
Yes	1.97	0.47	4.17	< 0.001	7.14 (2.84-17.97)	3.84	1.09	3.53	< 0.001	46.52 (5.53-391.38)

CI: Confidence interval; OR: Odds ratio; ACJJ N: American Joint Committee on Cancer Lymph Node; NLR: Neutrophil-to-lymphocyte ratio; PLR: Platelet-to-lymphocyte ratio.