口腔癌患者失志綜合征風(fēng)險(xiǎn)預(yù)測(cè)模型構(gòu)建及驗(yàn)證

打開文本圖片集

[中圖分類號(hào)] R739.8[文獻(xiàn)標(biāo)志碼] A [doi] 10.7518/hxkq.2025.2024340

Atotalof 486patients withoralcancerin WestChina HospitalofStomatologyof Sichuan Universityand Sun Yat-sen Memorial Hospital of Sun Yat-sen University from 2O24 March to July were selected by convenience sampling.We integrated clinical dataand evidence fromprevious studies to identifythekeyvariablesaffectingthe demoralization syndrome in patients with oralcancer.The 486patients were divided intoatrainingsetandavalidation set in an 8：2 ratio.Aclinical risk prediction model was established based on the individual dataof 365 patients inthedevelopmentcohort.Through leastabsoluteshrinkageand selectionoperator （LASSO）regresson，a moderate to severe risk prediction modelof demoralizationsyndrome inoralcancer was constructed，andaclinical machine-learning nomogram was constructed.Bootstrapresampling wasused for internalvalidation.The dataof121 patients inthe validationcohort were externallvalidated.ResultsThe incidenceof the demoralizationsyndrome inpatients withoral cancer was405 cases （83.3%） ，ofwhich279cases （57.4%） ）weremild，176cases （36.2%） ）weremoderate，and31 cases （6.4%） were severe.Thecore model，includingpatienteducation level，disease understanding，and MDASI-HNscore， was used to predict theriskof outcome.Internal validation of the model yielded C statisticof0.7836（ 95% CI： 0.78- （204號(hào) 0.87），beta of 0.8434，andcalibration interceptof -0.040 6 .Through externalvalidation，thevalidation setC statistic was0.80 （ 95%CI ：0.71-0.87），beta was 0.80，and calibration intercept was -0.08 .Conclusion Our risk prediction model of the demoralizationsyndrome in patients withoral cancer performedrobustly in validationcohorts of diferent nursing environments.The model has good correction and good discrimination and canbeused as an evaluation and prediction item at admission.

[Key words]oral cancer；demoralization syndrome； machine learning； prediction model

研究[1-2]表明，1990—2019年中國(guó)口腔癌年齡標(biāo)準(zhǔn)化發(fā)病率（age-standardized incidencerate，ASIR）和年齡標(biāo)準(zhǔn)化死亡率（age-standardizedmortalityrate，ASMR）分別以平均每年 1.6% 和0.7% 的速度上升，而中國(guó)的ASIR漲幅程度明顯高于全球平均水平。（剩余13243字）