帕金森病患者跌倒风险调查及基于LASSO回归的列线图预测模型建立与验证

打开文本图片集

【中图分类号】R741

【文献标志码】A

Investigation of fall risk in patients with Parkinson's disease and establishment and validation of a nomogram prediction model based on LASSO regression

HuHaiqiong，LiLixia，Shao Yu，HuangYuanyuan，Xiao Fajun，XiaKe （Department of Neurology，Sichuan Nanchong Mental Health Center）

【AbstractObectiveTinvestigatetheiskofalsinpatientswithParkinson’sdiseasendstablishandvalidateapredictionodel. Methods：Weselected372patientswith Parkinson'sdiseaseatSichuanNanchong Mental HealthCentrefromJanuary2O toSeptember2023.Thepatientswere dividedina7：3ratiointo model group（260cases）andvalidationgroup（112cases）.According topreviousliteratureadspedactdinicalpctieltdgralftisegtc.）ndasadcto （thedurationofarkison'sease，tetpefmedicatiostaken，etc.）atayessociatedwitllsipatietswithakso's disease.In themodelgroup，betweenpatients withandthosewithoutfalswithin1yearasreportedbythepatientsortheirfamily members，potentialpredictorsforfalseredeterinedthroughomparisonof gneralinformationanddisease-relatedfctors，east absoluteshrinkageandselectionoperator（LASSO）regresson，ndmultivarablelogisticegressnBasedontesignificantfactos， nomogram model was established and validated.Results：In the model group，81（ 31.15% ）of the 260 patients experienced falls. According totheLASSOregresionandmultivariablelogisticregressonresults，loholconsumption，thetypeofmedicatios，the scoreofthe UnifiedParkinson’sDiseaseRatingScalepartI（UPDR-II），theBergBalanceScalescor，tepresenceofartris， andthe presenceof osteoporosiswere independent factorsinfluencingfalls inpatientswith Parkinson’s disease.The areaundercurve （AUC）of the receiver operator characteristic curve（ROC）for predictingthe risk offallsin patientswith Parkinson’sdiseasewas 0.896（95%CI=0.856-0.935） in the model group and 0.883（95%CI=

0.840-0.926）inthevalidationgroup.Thecalibrationcurveanalysisresultsshowedthatthepreditioncurvesofthemodelandvalidationgroupscloselyftedthestandardcurves.ThedecisioncurveanalysisresultsidicatedthatwhentheprobabitythresholdforpredictingthefallriskinParkinson’sdseaseusingthenomogramwasO.1-.90，thenetbenefitrteofthepatientswasgreaterhanO. Conclusion ∵ Therisk offals in patients with Parkinson’sdisease are mainlyinfluenced byfactors such asalcoholconsumptionand thetypeofmedications.Thenomogrammodelestablishedinthisstudycanbeusedtopredictthefallriskinpatientswitharkison’s disease.

【Keywords】Parkinson’sdisease;fall;multivariable analysis;nomogram model

帕金森病为中枢神经系统变性疾病的一种，据统计国内65岁人群中帕金森病发病率高达 1.7% ，该病多发于中老年人群，患者可出现静正性震颤、肌强直、运动迟缓等表现[-2。（剩余9594字）