老年脑卒中患者心境状态变化轨迹及核心影响因素决策树分析OA
Analysis of mental state trajectories and core influencing factors using decision tree in elderly stroke patients
目的 探讨老年脑卒中患者的心境状态变化轨迹,分析不同轨迹亚型的预测因素,为心理干预提供依据.方法 采用便利抽样法,选取2023 年1 月—2024 年 6 月南京医科大学附属脑科医院收治的120 例老年脑卒中患者作为研究对象.在基线(入院时,T0)、1 个月(T1)、3 个月(T2)、6 个月(T3)和12 个月(T4)5 个时间点,使用一般资料调查表和简明心境量表中文版(BPOMS)进行调查评估.采用增长混合模型(GMM)识别心境状态的变化轨迹的潜在分类,运用决策树模型分析其核心影响因素.结果 老年脑卒中患者在 5 个时间点的 BPOMS 总分分别为(58.42±12.36)分、(52.18±11.94)分、(47.65±10.87)分、(43.21±9.76)分、(40.33±8.45)分,各时间点评分差异具有统计学意义(P<0.001).通过 GMM 识别出 2 条心境状态变化轨迹:轨迹 1(快速改善组,63.33%)、轨迹 2(持续不良组,36.67%).决策树模型显示,希望水平、社会支持评定量表评分、医院焦虑抑郁量表—抑郁评分、脑卒中复发是心境状态轨迹亚型的重要预测因素,其中希望水平的重要性最高(100%).结论 老年脑卒中患者的心境状态总体呈改善趋势,但存在明显的群体异质性,其中希望水平是心境状态变化的核心预测因子.
Objective To explore the trajectories of mental state changes in elderly stroke patients and analyze the predic-tive factors for different trajectory subtypes,so as to provide a basis for psychological intervention.Methods A convenience sampling method was used to select 120 elderly stroke patients admitted to the Affiliated Brain Hospital of Nanjing Medical University from Janu-ary 2023 to June 2024.Assessments were conducted using a general information questionnaire and the Chinese version of the Brief pro-file of mood states(BPOMS)at five time points:baseline(at admission,T0),1 month(T1),3 months(T2),6 months(T3),and 12 months(T4).Growth mixture modeling(GMM)was employed to identify distinct latent trajectories of mental state changes,based on which patients were grouped,and a decision tree model was used to analyze the core influencing factors.Results The total BPOMS scores of the elderly stroke patients at the five time points were(58.42±12.36)scores,(52.18±11.94)scores,(47.65±10.87)scores,(43.21±9.76)scores,and(40.33±8.45)scores,respectively,showing statistically significant differences across time points(P<0.001).GMM identified two distinct mental state trajectories:trajectory 1(Rapid Improvement Group,63.33%)and traj-ectory 2(Persistent Poor Group,36.67%).The decision tree model indicated that hope level,social support rating scale score,hos-pital anxiety and depression scale-depression(HADS-D)score,and recurrent stroke was important predictive factors for the trajectory subtypes,with hope level being the most significant factor(importance:100%).Conclusion The mental state of elderly stroke pa-tients shows an overall improving trend over time,but significant group heterogeneity exists.Hope level is identified as the core predic-tor of mental state changes.
胡庆;王静;向阳
南京医科大学附属脑科医院神经内科,江苏 南京 210029南京医科大学附属脑科医院神经内科,江苏 南京 210029南京医科大学附属脑科医院神经内科,江苏 南京 210029
老年脑卒中心境状态轨迹决策树希望社会支持
Elderly strokeMental stateTrajectoryDecision treeHopeSocial support
《国际老年医学杂志》 2026 (5)
613-618,6
江苏省卫生健康委科研项目(M2022065)
评论