Objective To construct a frailty risk prediction model for postoperative elderly patients with colorectal cancer based on Gobbens frailty integration theory, and to provide an effective screening tool for clinical frailty intervention.

Methods Patients with colorectal cancer who met the inclusion and exclusion criteria were selected and analyzed. The Tilbrurg frailty Assessment Scale, Barthel Index (BI), Simple Nutrition Assessment Refined (MNA-SF), cancer self-management efficacy scale and Lubben social network scale were used to collect relevant patient data, and the prediction model of frailty risk of postoperative patients with rectal cancer was constructed. The model was analyzed by univariate and multivariate analysis. ROC curve was used to verify the goodness of fit and prediction effect of model.

Results The incidence of postoperative frailty was 55.12%. The results of univariate analysis showed that there were statistically significant in the age, educational level, perceived economic status, perceived sleep status, presence or absence of stoma, total score of BI, total score of MNA-SF, self-stress reduction, self-decision-making, positive attitude and social isolation between two groups (P ＜ 0.05 to P ＜ 0.01). The results of logistic regression analysis model showed that the abnormal self-perceived sleep status, high total score of BI, high total score of MNA-SF, high self-stress reduction, high self-decision-making, and social isolation might be the risk factors of frailty (P ＜ 0.05 to P ＜ 0.01). The results of ROC curve analysis showed that the regression model had a high predictive value for frailty, with an AUC of 0.936, a sensitivity of 87.1% and a specificity of 89.5%.

Conclusions The incidence of postoperative frailty in elderly patients with rectal cancer is relatively high, and clinical nursing should pay attention to it. The self-perceived sleep status, total score of BI, total score of MNA-SF, self-stress reduction, self-decision-making and presence or absence of social isolation are the influencing factors of the occurrence of frailty. Clinically, these several indicators should be paid attention to observe. The prediction model has good predictive efficacy. Frailty is a dynamic variable, early identification of risks and implementation of effective intervention measures can, to a certain extent, delay the development of frailty and improve the health status and quality of life of the body.