.png){kind=icon}
.png){kind=icon}
Return
Prediction of depressive symptoms in middle-aged and older adult hospitalized patients with chronic diseases: a multicenter study in Anhui; China using machine learning methods
.png){kind=icon}
OA
AI
S
H
Q
X
X
G
X
S
R
L
M
M
DOI：10.3389/fpubh.2026.1865279.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
ObjectiveThis cross-sectional study has dual objectives: to investigate the predictive value of machine learning (ML) for the prevalence of depressive symptoms in middle-aged and older adult hospitalized patients with chronic diseases; and to identify significant factors influencing depressive symptoms in this population.MethodsA cross-sectional study was conducted among 618 hospitalized middle-aged and older adult patients with chronic diseases. Participants completed questionnaires assessing depression; chronic illness stigma; oral frailty; social isolation; family health; and demographic characteristics. The XG Boost model algorithm was employed for feature selection and variable importance ranking. A predictive model was constructed to assess the risk of depressive symptoms; and the feature importance honeycomb plot was utilized to illustrate the relationships between variables and prediction outcomes.ResultsThe study found multiple risk factors significantly associated with depression; including gender; place of residence; number of surgeries in the past year; hospitalization in the past 2 years; social isolation; level of education; comorbidity; age; malignant disease; oral frailty; stigma associated with illness; and family health. The XG Boost model demonstrated optimal predictive performance; achieving an AUC value of 0.931. Key predictive factors included stigma scale for chronic; family health; oral frailty; social isolation; malignant disease; and hospitalization in the past 2 years.ConclusionThis study constructed a risk prediction model for depressive symptoms in middle-aged and older adult hospitalized patients with chronic diseases; providing an important reference basis for mental health interventions in this population.
Keywords:
depressive symptoms
machine learning
chronic diseases
China
hospitalized
Journal
IF:
3.4
Papers:
2.5W
Citations:
5.7W
Organization
Cited Papers