Abstract:

Objective To explore the development trajectory of rehabilitation exercise adherence and its potential classes predictors in stroke patients. Methods By the convenience sampling method，252 stroke patients who were hospitalized in 2 tertiary hospitals in Daqing from September 2020 to May 2021 were selected.The general situation questionnaire，questionnaire of exercise adherence for stroke patients，the brief illness perception questionnaire，Herth hope index，and hospital anxiety and depression scale were used to collect the baseline data of patients，and the rehabilitation exercise adherence was assessed at 6，14 and 24 weeks after stroke. Latent growth mixed model was used to identify the development trajectory of rehabilitation exercise adherence among stroke patients，and multivariate logistic regression analysis was used to explore the predictors of different latent trajectories classes. Results A total of 4 developmental trajectories of rehabilitation exercise adherence among stroke patients were identified：low adherence-improvement group（7.5%），sustained moderate adherence group（21.4%），moderate adhere-nce-decline group（46.5%）and low adherence-decline group（24.6%）. Multivariate Logistic regression analysis showed that stroke type，number of stroke attacks，baseline hope，anxiety and depression levels were independent predictors of different developmental trajectory of rehabilitation exercise adherence among stroke patients（P<0.05）. Conclusion There is significant group heterogeneity in the developmental trajectory of rehabilitation exercise adherence in stroke patients within 24 weeks of onset. Only a small number of patients’ adherence continued to be at a high level or improved from a low level，and about 1/4 of the patients continued to be at a low level and tending to decline. Medical staff should identify patients in the low adherence-decline group early on the basis of predictive factors，and implement targeted，specific and efficient adherence interventions to improve the development trend of adherence.

Key words: Stroke, Rehabilitation Exercise, Adherence, Developmental Trajectory, Predictors, Potential Classes, Nursing Care

摘要：

目的 探讨脑卒中患者康复锻炼依从性发展轨迹及其潜在类别的预测因素。方法 采用便利抽样法，选取2020年9月—2021年5月大庆市2所三级甲等医院住院的252例脑卒中患者为调查对象。采用一般情况调查表、脑卒中患者功能锻炼依从性量表、疾病感知问卷简版、Herth希望指数、医院焦虑抑郁量表收集患者的基线资料，于脑卒中后第6、14、24周评估患者的康复锻炼依从性水平，应用潜变量增长混合模型识别依从性发展轨迹，并采用多元Logistic回归分析不同发展轨迹类别的预测因素。结果 共识别出脑卒中患者4种康复锻炼依从性发展轨迹：低依从性-改善组（7.5%）、持续中依从性组（21.4%）、中依从性-下降组（46.5%）与低依从性-下降组（24.6%）。多元Logistic回归分析显示，脑卒中类型、脑卒中发作次数以及基线希望、焦虑、抑郁水平是脑卒中患者康复锻炼依从性不同发展轨迹类别的独立预测因素（P<0.05）。结论 脑卒中患者发病24周内康复锻炼依从性发展轨迹存在显著群体异质性，仅少部分患者依从性持续处于中等水平或由低水平获得改善，约1/4患者持续处于低水平并有下降趋势。医护人员应依据预测因素早期识别低依从性-下降组患者，实施针对性、特异性与高效的康复锻炼依从性干预，以改善依从性的发展趋势。

关键词: 脑卒中, 康复锻炼, 依从性, 发展轨迹, 预测因素, 潜在类别, 护理