Construction and validation of a predictive model for depressive symptoms in migrant elderly
YANG Yong
SONG Qingtao
GE Zhonghao
XUAN Na
YIN Xiuli
Abstract：Objective To develop and validate a risk prediction model for incident depressive symptoms among mi‐grant elderly and verify its predictive efficacy.Methods By adopting the convenient cluster sampling method,613 mi‐grant elderly without depressive symptoms were enrolled from 2 community health service centers in Jinan City to establish a cohort and conduct a 2-year follow-up.The research data were randomly divided into a training set and a testing set at a ratio of 3∶1,with the training set used for model fitting and the testing set for internal validation.Univariate analysis and multivariate logistic regression analysis were applied to screen the influencing factors of new-onset depressive symptoms in migrant elderly.A prediction model was constructed on the training set using the three-time five-fold cross-validation.The receiver operating characteristic(ROC)curve and the area under the ROC curve(AUC)were used to verify the model effi‐cacy.Results A total of 78 migrant elderly developed new-onset depressive symptoms during the follow-up period.Mul‐tivariate logistic regression analysis showed that compared with the highest quartile(Q4)of upward intergenerational sup‐port,those in Q(1 OR=1.23,95%CI:1.04-1.45,P<0.01)and Q(2 OR=1.12,95%CI:1.01-1.24,P=0.03)had a higher risk of new-onset depressive symptoms;compared with Q4 of bidirectional emotional support,those in Q1(OR=1.15,95%CI:1.06-1.27,P=0.02)and Q2(OR=1.10,95%CI:1.01-1.23,P=0.03)had an elevated risk of new-onset depres‐sive symptoms.The AUC of the constructed risk prediction model in the testing set was 0.801 1(95%CI:0.767 4-0.834 9),with an accuracy of 76.19%,a sensitivity of 75.96% and a specificity of 71.78%.The efficacy of the training set was com‐parable to that of the testing set,suggesting a stable predictive performance of the model.Conclusions Upward intergen‐erational support and bidirectional emotional support are relatively important predictors of depressive symptoms in migrant elderly.The risk prediction model constructed in this study has good predictive efficacy,which can provide a reliable tool and scientific evidence for communities to accurately identify migrant elderly at high risk of depression and formulate multi-target intervention strategies.
Keywords：migrant elderlyintergenerational emotional supportdepressive symptompredictive model
Publication Date：2026-02-25
Online Publishing Date：2026-03-19(First online date of this platform, not the publication date of the document)
Pages：5( 40-44 )