Construction And Validation Of A Risk Prediction Model For Body Image Disturbance In Patients With Head And Neck Cancer

Objective:The purpose of this study was to comprehensively extract and summarize the risk factors of body image disturbance in head and neck cancer patients through systematic evaluation,and to construct a prediction model of body image disturbance in head and neck cancer patients based on Logistic regression analysis,and to verify the prediction effect of the model,so as to provide a reference for clinical nurses to assess the risk of body image disturbance in head and neck cancer patients,screen high-risk groups and formulate intervention programs in time.Methods:1.We systematically retrieved relevant research reports on factors influencing body image disturbance in head and neck cancer patients,conducted literature screening,information extraction and quality evaluation to clarified risk factors for body image disturbance in head and neck cancer patients.2.Used the convenience sampling method,332 patients with head and neck cancer who were hospitalized in the Department of Otolaryngology,Head and Neck Surgery,Department of Stomatology and Department of Radiotherapy in a tertiary care hospital in Taiyuan City from June to November 2022,and patients with head and neck cancer who were hospitalized in the Department of Head and Neck Surgery in a specialized oncology hospital in Taiyuan City from July to September 2022 were selected as study subjects,and a self-administered questionnaire on risk factors for body image disturbance in patients with head and neck cancer was used to collect patients’relevant information.The patients were divided into the modeling group and the validation group according to the ratio of 7:3 by the random number table method.The patients in the modeling group were divided into the group with body image disturbance and the group with body image normality based on the occurrence of body image disturbance after treatment,and the risk factors for body image disturbance were explored by univariate analysis and logistic regression analysis to establish a prediction model and draw a nomogram.Internal validation of the model was performed using data from the validation group.The area under the ROC curve and the Hosmer-LemeshowX~2 test were used to evaluate the discrimination and calibration of the model,respectively.Results:1.Eighteen studies were eventually included in the systematic evaluation,including1 prospective cohort study,1 prospective longitudinal study,1 retrospective study,and 15cross-sectional studies.Risk factors for body image disturbance in head and neck cancer patients included a total of 3 aspects:sociodemographic factors,disease-and treatment-related factors,and psychosocial factors.2.Finally,332 patients with head and neck cancer were included,including 123cases(37.0%)with body image disturbance.There were 235 cases in the modeling group with 89 cases(37.9%)of body image disturbance and 97 cases in the validation group with 34 cases(35.1%)of body image disturbance.Differential analysis of all variables between patients in the modeling and validation groups showed that there was no statistical difference between the two groups in terms of general information,disease-and treatment-related information,and psychosocial-related information(P>0.05).Univariate analysis of the modeling group showed statistically significant differences(P<0.05)when comparing the primary chaperone,tumor stage,treatment modality,presence or absence of cervical lymph node dissection,presence or absence of tracheal cannula,presence or absence of flap reconstruction,functional score,anxiety score,depression score,comprehension social support score,social anxiety score,positive coping score,and negative coping score.The results of multi-factor logistic analysis showed that the differences were statistically significant(P<0.05)for six factors:tumor stage,presence of tracheal cannula,functional score,anxiety score,comprehension social support score,and negative coping score.The area under the ROC curve was 0.897(95%CI 0.858 to 0.936,P<0.001)for the nomogram graph prediction model constructed based on the above risk factors.The Jorden index with a maximum value of 0.638 corresponded to a cut-off value of 0.377,a sensitivity of 0.843,and a specificity of 0.795;the Hosmer-Lemeshow(H-L)X~2 test showed thatX~2=7.828,P=0.348.The results of the validation group test showed that The area under the ROC curve was 0.850(95%CI 0.774 to 0.927,P<0.001),the sensitivity was 0.912,the specificity was 0.698,and the accuracy was 77.3%;the results of the Hosmer-Lemeshow(H-L)X~2 test showed thatX~2=5.409,P=0.610.Conclusion:1.Tumor stage>stage I,presence of tracheal cannula,functional status score<20,anxiety score≥8,comprehension social support score<69,and negative coping score≥13 were independent risk factors for the development of body image disturbance in patients with head and neck cancer.2.The study constructed a nomogram graph prediction model of body image disturbance in head and neck cancer patients with high discrimination and calibration,good prediction effect,and simplicity,which can provide reference for clinical nurses to assess the risk of body image disturbance in head and neck cancer patients and make early intervention.