Finally, we selected mastication disorder, preoperative weight loss, and energy intake (preoperatively and at the fourth week postoperatively) as variables in the multivariate Cox proportional hazards model because these factors had a marginal significance based on the univariate Cox proportional hazards model, correlation and VIF.
2
—
—
The sentences
However, owing to lack of marginal significance in the results of the univariate Cox proportional hazards model for these factors, we did not use these factors in the multivariate Cox proportional hazards model.