摘要
目的 探讨心力衰竭患者非计划再入院率及影响因素。方法 选取2019年3月~2021年3月如皋市人民医院收治的228例经治疗后病情稳定出院的心力衰竭患者作为研究对象,统计患者1年非计划再入院情况,并依据是否发生非计划再入院分为再入院组和未再入院组,比较两组临床资料,分析非计划再入院相关影响因素,采用R语言构建非计划再入院的列线图预测模型,发现该模型的预测心力衰竭患者非计划再入院的价值。结果 完成随访的223例患者中,有96例(43.05%)因再发心力衰竭住院;再入院组慢性肾病占比(19.79%)、贫血占比(29.17%)、营养风险占比(36.46%)、心房颤动(房颤)占比(40.63%)、血清NT-proBNP水平高于未再入院组,遵医用药占比(64.58%)低于未再入院组(P<0.05);慢性肾病、贫血、营养风险、房颤、遵医用药、血清NT-proBNP水平均为心力衰竭患者非计划再入院的影响因素(P<0.05);Logit(P)=-13.259+3.199×慢性肾病+2.267×贫血+2.749×营养风险+3.227×房颤-0.799×遵医用药+3.416×NT-proBNP,据此构建列线图预测模型,其C-index为0.962,具有良好的区分度;预测模型ROC曲线的曲线下面积(AUC)为0.962(95%CI 0.939~0.984),预测效能较高。结论 如皋地区心力衰竭患者非计划再入院率为43.05%,处于较高水平,慢性肾病、贫血、营养风险、房颤、遵医用药、血清NT-proBNP水平均为非计划再入院的影响因素,根据上述影响因素构建列线图预测模型有助于预测心力衰竭患者1年内非计划再入院风险,具有较高预测效能。
Objective To investigate the unplanned readmission rate and influencing factors of heart failure patients in Rugao area.Methods A totle of 228 heart failure patients who were discharged from our hospital area with stable conditions after treatment from March 2019 to March 2021 were selected as the study subjects,and the 1-year unplanned readmissions were counted,and the patients were divided into readmission and non-readmission groups according to whether unplanned readmissions occurred,and the clinical data of the two groups were compared,and the influencing factors related to unplanned readmissions were analyzed using a multifactor logistic regression equation,and a column line graph prediction model for unplanned readmissions was constructed using R language,and the value of the model in predicting unplanned readmissions in heart failure patients was evaluated using subject work characteristic(ROC) curves.Results Among 223 patients who completed follow-up,96(43.05%) were hospitalized for re-entry heart failure;the percentages of chronic kidney disease(19.79%),anemia(29.17%),nutritional risk(36.46%),atrial fibrillation(40.63%),and serum NT-proBNP levels were higher in the readmission group than in the non-readmission group,and the percentages of medication compliance(64.58%) were lower than those in the non-readmission group(P<0.05);chronic kidney disease,anemia,nutritional risk,atrial fibrillation,medication compliance,and serum NT-proBNP levels were all factors influencing unplanned readmission in patients with heart failure(P<0.05);Logit(P) =-13.259 + 3.199 × chronic kidney disease + 2.267 × anemia + 2.749 × nutritional risk + 3.227 × atrial fibrillation-0.799 × medication compliance +3.416×NT-pro BNP,according to which the column line graph prediction model was constructed and validated by Bootstrap self-sampling method,which showed that the predicted values were basically consistent with the actual observed values,and the C-index of this prediction model was 0.962,with good discrimination;the curve of the ROC curve of the prediction model The area under the curve(AUC) of the prediction model was 0.962(95%CI0.939-0.984),with high predictive efficacy.Conclusion The unplanned readmission rate of heart failure patients in Rugao city area is 43.05%,which is at a high level.Chronic kidney disease,anemia,nutritional risk,atrial fibrillation,medication compliance,and serum NT-pro BNP level are all influencing factors for unplanned readmission.The prediction model based on the above-mentioned factors can help predict the risk of unplanned readmission in heart failure patients within 1 year,with high predictive power.
作者
黄荔荔
徐琴
沈建宏
李雯
HUANG Li-li;XU Qin;SHEN Jian-hong;LI Wen(Department of Cardiology Rugao People’s Hospital,Rugao 226500,China)
出处
《中国心血管病研究》
CAS
2022年第9期811-816,共6页
Chinese Journal of Cardiovascular Research
基金
2021年如皋市科技攻关计划项目(RG2021SZL002)。
关键词
心力衰竭
非计划再入院
影响因素
列线图模型
预测
Heart failure
Unplanned readmission
Influencing factors
Line graph model
Prediction
