Skip to main content
. 2019 Dec 27;9:19872. doi: 10.1038/s41598-019-56431-w

Table 2.

Propensity score-adjusted logistic regression analysis of choice of treatment with best supportive care alone.

Variable Univariate Analysis Multivariate Analysis
OR 95% CI p-value OR 95% CI p-value
Age (low → high) 1.08 1.05–1.11 0.0001 1.11 1.05–1.16 0.0001
Dementia (absent vs. present) 20.4 5.85–71.4 0.0001 37.0 6.54200.0 0.0001
Performance status (0 → 4) 3.77 2.83–5.03 0.0001 6.10 3.839.71 0.0001
Patient’s desire for best supportive care alone (absent vs. present) 4.02 2.22–7.30 0.0001 22.2 6.7176.9 0.0001
Pulmonary disease (absent vs. present) 2.13 1.34–3.39 0.001 3.83 1.648.93 0.002
Epidermal growth factor receptor mutation (mutant vs. wild type) 2.97 2.00–4.42 0.0001 2.68 1.295.56 0.008
Social background (absent vs. present) 3.39 1.46–7.81 0.004 5.44 1.2224.4 0.027
Psychiatric disorder (absent vs. present) 4.65 1.49–14.5 0.008 9.09 1.2666.7 0.029
Renal dysfunction (absent vs. present) 2.52 1.25–5.08 0.010 3.18 0.92–11.0 0.068
Blood disease (absent vs. present) 6.67 1.27–34.5 0.025 6.37 0.69–58.8 0.103
Aspartic acid aminotransferase (low → high) 1.02 1.002–1.03 0.031 1.01 0.99–1.03 0.366
Sex (female → male) 1.57 0.94–2.64 0.086 1.36 0.51–3.64 0.534
Histologic type (adenocarcinoma vs. others) 2.94 1.81–4.79 0.0001 NI
Albumin (low → high) 0.53 0.38–0.75 0.0001 NI
Lactate dehydrogenase (×103) (low → high) 83.3 12.4–500.0 0.0001 NI
Doctor’s recommendation about best supportive care (absent vs. present) 32.3 10.3–100.0 0.0001 NI
C-reactive protein (low → high) 1.16 1.10–1.23 0.0001 NI
White blood cells (×103) (low → high) 1.11 1.05–1.17 0.0001 NI
Stage (1 → 4) 1.27 1.10–1.46 0.001 NI

Social background includes the patient’s history of public assistance and psychiatric hospitalization.

Abbreviations: OR, Odds ratio; C.I., confidence interval; Variables with a p-value < 0.10 on univariate analysis were entered into the multivariate logistic analysis using the.

simultaneous method. NI, not included in the best multivariate logistic regression model.