全飞秒激光近视手术治疗病人术前固视训练依从性预测模型的建立与验证  

Establishment and verification of a prediction model for preoperative fixation training compliance in patients treated with small incision lenticule extraction surgery

作　　者：卢海燕[1] 何坤夏 韩云 LU Haiyan;HE Kunxia;HAN Yun(The People′s Hospital of Guangxi Zhuang Autonomous Region,Guangxi 530021 China)

机构地区：[1]广西壮族自治区人民医院,广西530021

出　　处：《循证护理》2024年第1期90-94,共5页Chinese Evidence-Based Nursing

摘　　要：目的:探讨全飞秒激光近视手术治疗病人术前固视训练依从性的影响因素,构建列线图预测模型并验证。方法:采用便利抽样法,选取2022年1月-2023年3月收治的全飞秒激光近视手术治疗病人345例,按入组先后顺序将前70%的病人设为建模组(242例),后30%的病人设为验证组(103例)。收集病人一般资料,采用自制的固视训练知信行问卷、中文版一般自我效能感量表、自制的病人固视训练依从性量表对病人进行调查。采用Logistic回归分析筛选固视训练依从性危险因素,通过R软件构建列线图模型并验证。结果:345例全飞秒激光近视手术治疗病人中,108例(31.30%)术前固视训练依从性差。建模组多因素Logistic回归分析显示,年龄、工作状态、知信行评分、自我效能评分是全飞秒激光近视手术治疗病人术前固视训练依从性的影响因素(均P<0.05)。列线图模型验证结果显示,AUC为0.793,95%CI(0.734,0.852);Calibration校准曲线中理想曲线与实际曲线走势基本一致,斜率接近1。结论:构建的全飞秒激光近视手术治疗病人术前固视训练依从性风险预测列线图模型具有良好的区分度、预测精准度,可用于评估全飞秒激光近视手术治疗病人术前固视训练依从性差的风险,利于及早采取干预措施提高病人的训练依从性。Objective:To explore the influencing factors of preoperative fixation training compliance in patients undergoing Small Incision Lenticule Extraction(SMILE)surgery and to construct and verify a nomogram prediction model.Methods:A total of 345 patients treated with SMILE surgery from January 2022 to March 2023 were enrolled as subjects by the convenience sampling method.The first 70%of admitted patients were set as the modeling group(242 cases)in order of enrollment,and the last 30%were the validation group(103 cases).The patients′general information were collected,and the patients were investigated using the self-made Fixation Training Knowledge Scale,Belief and Action Questionnaire,the Chinese version of the General Self-Efficacy Scale,and the self-made Patient Fixation Training Compliance Scale.Logistic regression analysis was used to screen risk factors for fixation training compliance,and a nomogram model was constructed and verified using R software.Results:Among 345 patients treated with SMILE surgery,108(31.30%)had poor compliance with preoperative fixation training.Multi-factor Logistic regression analysis of the modeling group showed that age,working status,knowledge,belief and behavior,and self-efficacy affect the compliance of preoperative fixation training in patients treated with SMILE surgery(all P<0.05).The nomogram model verification results showed that the AUC was 0.793,95%CI(0.734-0.852);the ideal curve in the Calibration curve was consistent with the actual curve,and the slope was close to 1.Conclusions:The constructed risk prediction nomogram model for preoperative fixation training compliance in patients treated with SMILE surgery has good discrimination and prediction accuracy,which could be used to evaluate preoperative fixation in patients treated with SMILE surgery.The risk of poor training compliance is conducive to taking early intervention measures to improve patients′training compliance.

关 键 词：全飞秒激光近视手术 固视训练 依从性 影响因素 预测模型 

分 类 号：R473.77[医药卫生—护理学]