Logistic Regression Model in a Machine Learning Application to Predict Elderly Kidney Transplant Recipients with Worse Renal Function One Year after Kidney Transplant: Elderly KTbot

Table 6

Univariate and multivariate analysis of risk factors for CKD-EPI < 60 mL/min/1.73 m² one year after kidney transplant.

Univariate analysis					Multivariate analysis
Covariables	OR crude (95% CI)	OR adjusted (95% CI)	β (SE)		Covariables	OR adjusted (95% CI)	β (SE)
DM pre-KT	0.69 (0.20–2.35)	0.69 (0.18–2.21)	−0.37 (0.62)	0.56
Standard deceased donor
Yes	—	—	—	—
No	1.58 (0.57–4.37)	1.58 (0.56–4.35)	0.46 (0.52)	0.37
Mismatch HLA-A
0	—	—	—	—
1	1.84 (0.44–7.65)	1.84 (0.48–9.06)	0.61 (0.73)	0.40
2	1.34 (0.31–5.80)	1.34 (0.34–06.80)	0.30 (0.75)	0.69
Proteinuria >0.3 g/24 h	1.07 (0.45–2.56)	1.07 (0.44–2.53)	0.07 (0.45)	0.88	Constant	—	3.06 (1.72)
Age ≥ 60 years	4.23 (1.40–12.83)	4.23 (1.43–13.57)	1.44 (0.57)	0.01	Age ≥ 60 years	4.67 (1.52–15.55)	1.54 (0.58)	0.01
Haemoglobin	—	1.69 (1.04–1.32)	−0.28 (0.12)	0.03	Haemoglobin	1.35 (1.06–1.78)	−0.31 (0.13)	0.03

CKD-EPI: chronic kidney disease epidemiology collaboration; DM: diabetes mellitus; KT: kidney transplant; HLA: human leukocyte antigen; OR: odds ratio; CI: confidence interval; β: beta model coefficient; SE: standard error; multivariate analysis model: P < 0.002; Chi-square test (2 degrees of freedom): 12.64; determination coefficient (R²) of Nagelkerke: 0.15; C statistics: 0.70; and area under the curve (AUC): 70%. Crude odds ratio for haemoglobin was not calculated.