Data Sheet 1_Heterogeneity of benefit finding in maintenance hemodialysis patients: a decision tree-based subgroup analysis of self-efficacy and social support.pdf

BackgroundBenefit finding (BF) improves quality of life in maintenance hemodialysis (MHD) patients, yet population heterogeneity remains understudied. This study explores how self-efficacy and social support jointly influence BF patterns and identifies distinct patient subgroups.
MethodsThis multi-center cross-sectional study was conducted from April to September 2023 at five tertiary hospitals in Shanghai, China, enrolling 352 MHD patients. Data from the Benefit Finding Scale, General Self-Efficacy Scale, and Perceived Social Support Scale were used to construct a Classification and Regression Tree (CART) model employing five-fold cross-validation, with a maximum depth of 3 and a minimum leaf node size of 10%.
ResultsThe CART model (R²=0.278) identified five distinct BF subgroups (p<0.001): Low Self-Efficacy Constrained Group, Psychological Resource Deficient Group, Internally Belief Driven Group, Balanced Resource Adaptation Group, and Resource Integrated Advantage Group, each characterized by unique combinations of self-efficacy and social support. Significant differences were observed among the subgroups in terms of gender(p=0.045), education level(p=0.010), and employment status(p=0.003).
ConclusionBF levels in MHD patients demonstrated significant variations influenced by the combined effects of self-efficacy and social support. The decision tree model successfully identified patient subgroups with distinct psychological resource configurations. These findings provide a theoretical foundation for implementing stratified and personalized psychological interventions in clinical practice. Clinicians can identify and prioritize vulnerable patients who simultaneously lack self-efficacy and social support, offering them targeted positive psychological interventions that may potentially improve treatment adherence and long-term prognosis.

背景：益处发现（Benefit Finding, BF）可改善维持性血液透析（Maintenance Hemodialysis, MHD）患者的生活质量，但目前针对该人群异质性的研究仍较为匮乏。本研究旨在探讨自我效能与社会支持如何共同影响BF模式，并识别不同的患者亚组。
方法：本研究为多中心横断面研究，于2023年4月至9月在中国上海的5家三级医院开展，共纳入352例MHD患者。研究采用益处发现量表、一般自我效能感量表以及领悟社会支持量表收集数据，构建分类回归树（Classification and Regression Tree, CART）模型，采用五折交叉验证进行模型优化，设定模型最大深度为3，最小叶节点占比为10%。
结果：本次构建的CART模型（决定系数R²=0.278）成功识别出5个具有显著统计学差异的BF亚组（p<0.001），分别为低自我效能受限组、心理资源匮乏组、内在信念驱动组、资源适配均衡组以及资源整合优势组，各亚组均具有独特的自我效能与社会支持组合特征。各亚组在性别（p=0.045）、受教育程度（p=0.010）以及就业状态（p=0.003）方面均存在显著差异。
结论：MHD患者的益处发现水平存在显著差异，且该差异受自我效能与社会支持的共同影响。本研究构建的决策树模型成功识别出具有不同心理资源配置特征的患者亚组。本研究结果为临床实践中开展分层化、个性化心理干预提供了理论依据。临床医师可通过该模型识别并优先干预同时缺乏自我效能与社会支持的脆弱患者，为其提供针对性的积极心理干预手段，或可改善患者的治疗依从性与长期预后。