基于随机森林算法的2型糖尿病共病病人焦虑和抑郁影响因素研究OA

Objective:To investigate the status quo of anxiety and depression in patients with type 2 diabetes mellitus(T2DM)and comorbidities based on random forest algorithm and analyze their influencing factors.Methods:A total of 1 652 patients with T2DM and comorbidities were selected by convenience sampling from a tertiary hospital and 3 community hospitals in Pudong New District,Shanghai city from November 2023 to April 2024.A general information questionnaire,the Generalized Anxiety Disorder Scale(GAD-7),and the Patient Health Questionnaire-9(PHQ-9)were used for the survey.Random forest was employed to screen for significant influencing factors.Logistic regression analysis was conducted to explore the factors influencing anxiety and depression in these patients.Results:Among the 1 652 patients,382(23.1%)had anxiety,and 565(34.2%)had depression.Six factors influenced anxiety,ranked in descending order of importance was marital status,employment status,age,number of comorbidities,family history of diabetes,and alcohol consumption history.Nine factors influenced depression,ranked in descending order of importance was age,family history of diabetes,employment status,education level,marital status,monthly household income,number of comorbidities,disease duration,and body mass index(BMI).The result of Logistic regression analysis showed that age,marital status,employment status,family history of diabetes,and number of comorbidities were common influencing factors for both anxiety and depression in patients with T2DM and comorbidities(P<0.05).Alcohol consumption history was a factor for anxiety(P<0.05).While BMI,education level,monthly household income,and disease duration were factors for depression(P<0.05).Conclusions:The incidence of anxiety and depression was relatively high among patients with T2DM and comorbidities,with multiple influencing factors.Healthcare providers should identify these factors early and implement targeted interventions to improve patients'mental health and outcomes.