Skip to main content
. 2024 Jul 6;29(11):e1523–e1531. doi: 10.1093/oncolo/oyae157

Table 4.

Associations between patient characteristics and chemotherapy toxicities, after overlap weighting

Univariate analysis Multivariate analysis
Moderate toxicities Severe toxicities Moderate toxicities p Severe toxicities p
Variable& OR 95%CI p OR 95%CI p OR 95%CI OR 95%CI
Female sex 0.88 0.51-1.49 .63 0.86 0.49-1.50 .59
>80 years of age 0.79 0.46-1.35 .38 0.95 0.54-1.68 .87
Metastatic status 0.91 0.54-1.55 .74 1.06 0.61-1.86 .83
ECOG-PS > 1 1.37 0.80-2.35 .25 1.68 0.96-2.94 .07 1.12 0.62-2.01 .71 1.26 0.68-2.33 .46
Polychemotherapy 1.38 0.82-2.33 .23 1.14 0.66-1.98 .63
Hemoglobin < 10 g/dl 1.82 0.75-4.38 .19 3.45 1.46-8.14 .005 1.78 0.71-4.47 .22 2.96 1.20-7.29 .02
Creatine clearance rate < 30 mL/minute 1.48 0.33-6.55 .49 2.80 0.68-11.56 .15 1.40 0.31-6.32 .66 2.42 0.56-10.33 .23
Albuminemia < 35 g/L 0.98 0.57-1.69 .94 1.49 0.85-2.61 .16 0.78 0.43-1.40 .40 1.03 0.56-1.89 .93
6-month weight loss 1.45 0.77-2.75 .25 1.43 0.74-2.77 .29
CIRS-G > 12 1.58 0.86-2.90 .14 1.87 1.00-3.48 .05 1.51 0.80-2.85 .20 1.62 0.84-3.14 .15
ADL < 6 1.18 0.54-2.61 .68 1.32 0.59-2.97 .50
IADL < 7 1.14 0.67-1.96 .63 1.39 0.80-2.43 .25
TUG < 20 s 0.52 0.25-1.05 .07 0.56 0.26-1.16 .12 0.53 0.25-1.11 .09 0.65 0.30-1.42 .28
Cognitive impairment 0.89 0.45-1.76 .74 1.20 0.61-2.38 .59
BMI
<22 vs 22-25 kg/m2
>25 vs 22-25 kg/m2

1.12
1.34

0.57-2.22
0.72-2.49

.74
.36

1.27
1.17

0.62-2.56
0.61-2.26

.51
.63
Gynaecological cancer* 0.87 0.45-1.68 .69 0.73 0.37-1.46 .38
GU cancer* 1.14 0.46-2.82 .77 1.34 0.54-3.34 .52
Lung cancer* 0.98 0.34-2.79 .97 0.84 0.28-2.55 .76
Other cancer* 0.73 0.33-1.64 .45 0.67 0.29-1.55 .35

&For continuous variables, the Youden Index30 was used to identify the cutoff with the highest sensitivity and specificity for classifying the presence or absence of toxicity. Variables with P < .1 in the univariate analysis and clinically relevant variables (single-agent chemotherapy or combination chemotherapy) were examined further in a multivariate logistic regression model.

*Reference: digestive cancer.

Abbreviations: ADL, Activities of Daily Living; BMI, body mass index; CIRS-G, Cumulative Illness Rating Scale for Geriatrics; ECOG-PS, Eastern Cooperative Oncology Group Performance Status; GU, genito-urinary; IADL, Instrumental Activities of Daily Living; TUG, timed up-and-go.