Table 2. . Multivariable linear regression models of variables independently associated with urinary levels of β-2 microglobulin among 41 childhood cancer survivors post-treatment.

Variable	B	β	p-value	Partial R	T	VIF
Initial model: all covariates
Constant	-20,075	–	0.001	–	-3,877	–
Ifosfamide use (yes/no)	-0,858	-0,076	0.304	-0,201	-1,048	2,213
Cisplatin use (yes/no)	-1,079	-0,099	0.306	-0,2	-1,043	3,802
Cyclophosphamide use (yes/no)	-0,082	-0,007	0.903	-0,024	-0,123	1,552
Carboplatin use (yes/no)	-0,411	-0,031	0.645	-0,091	-0,466	1,893
Methotrexate (yes/no)	-0,098	-0,005	0.936	-0,016	-0,081	1,822
Follow-up (years)	-0,004	-0,035	0.666	-0,085	-0,437	2,777
Age at diagnosis (years)	-0,059	-0,047	0.583	-0,108	-0,556	3,004
Metastasis (yes/no)	0,043	0,003	0.96	0,01	0,051	1,573
Gender (male/female)	-0,494	-0,045	0.477	-0,14	-0,722	1,667
Plasma levels of magnesium (mg/dl)	1,811	0,059	0.394	0,168	0,867	1,968
Plasma levels of urea (mg/dl)	-0,063	-0,084	0.296	-0,205	-1,067	2,604
Plasma levels of cholesterol (mg/dl)	0,042	0,166	0.031	0,408	2,282	2,224
Plasma levels of creatinine (mg/dl)	19,878	0,821	< 0.001	0,919	11,91	2,003
Systolic blood presure (mmHg)	0,104	0,233	0.009	0,487	2,845	2,829
Final model: only covariates with p < 0.05
Constant	-15,472	–	< 0.001	–	-5,725	–
Plasma levels of cholesterol (mg/dl)	0,036	0,142	0.005	0,439	2,975	1,107
Plasma levels of creatinine (mg/dl)	19,784	0,817	< 0.001	0,926	14,968	1,451
Systolic blood presure (mmHg)	0,07	0,158	0.007	0,425	2,854	1,493

B: Unstandardized coefficient; β: Standardized coefficient; OSM: Osmolality; Partial R: Partial correlation coefficient; T: t-statistic; VIF: Variance inflation factor.