儿童肺炎住院时间延长预测模型的构建及临床应用价值

打开文本图片集

中图分类号：R725.631 文献标志码：A DOI：10.11958/20251021

Abstract：ObjectiveToconstructa prediction model forthe length hospitalization inchildren with pneumonia basedonclinical characteristics.MethodsAretrospective analysis theclinical data1255childrenwith pneumonia was conducted.The patients were divided into two groups based on the median length hospitalization： the ⩽7 days group （628 cases） the >7 days group （627 cases）. The diferences between the two groups in demographic characteristics，past medical history，linicalmanifestations，laboratorytestresults，imaging findings，treatmentplansotherclinicaldatawere compared.A multivariatestepwiseLogisticregresionanalysiswasperformed toidentifythe factorsinfluencing hospitalization for >7 days to construct a prediction model. The model was evaluated using the receiver operating characteristic （ROC） curve the clinical decision curve.ResultsCompared to the ⩽7 days group， children in the >7 days group wereyoungeringae，hadhigher height，ahigherproportion preterminfants，ahigherproportion previous pneumonia historyahigherbodytemperaturesatadmision.Furthermore，inthe >7days group，white bloodcellount， neutrophilcount，plateletcount，C-reactiveprotein（RP）procalcitoninlevels were elevated.Theproportionbilateral lesions，oxygentherapyrespiratorysupportpleural efusionwere higher，whilelymphocytecounthemoglobinlevels were lower （ P<0.05 ）.The results the multivariate Logistic regression analysis showed that age （OR=O.979，95% CI： 0.972-0.987），historyprematurity （OR=1.751，95%CI：1.216-2.521），previoushistory pneumonia （OR=1.52，95%Cl： 1.037-2.228），admission temperature （OR=1.290，95%CI： 1.097-1.518），serum CRP （OR=1.019，95% CI： 1.013-1.025）， pleural efusion（OR=1.980，95% CI： 1.309-2.994） oxygen therapy （OR=2.849，95% CI： 1.851-4.385）were independent risk factors for a hospital stay > 7 days in children with pneumonia.The model had an accuracy 79.2%， the area under thecurve （AUC）was 0.919 （95%CI：0.854-0.961）.ConclusionTheregression model constructed basedonclinical characteristics caneffectivelypredictthelength hospitalizationinchildrenwith pneumonia.Itprovidessientific evidence for the early identification high-riskchildren，optimization treatment plans shortening hospital stays.

Keywords：child; pneumonia;length stay;Logistic models;prediction model

肺炎是全球5岁以下儿童发病和死亡的主要原因，严重威胁着儿童健康。（剩余9283字）