Abstract:

Objective To explore the latent classes of medication deviation trajectories and their influencing factors in elderly patients with coronary heart disease（CHD），and to provide a theoretical basis for targeted clinical interventions.Methods A total of 285 elderly patients with CHD，hospitalized in a tertiary hospital in Guilin from March 2024 to March 2025，were selected using convenience sampling. A general information questionnaire，the Frailty Scale，and the Drug Literacy Questionnaire were used for the baseline survey. The Medication Deviation Assessment Tool was used to collect data at 4 time points：1 week（T1），1 month（T2），3 months（T3），and 6 months（T4） after discharge. Latent Class Growth Modeling（LCGM） was used to identify the latent classes of medication deviation trajectories，and multivariate logistic regression analysis was employed to explore the influencing factors.Results A total of 253 patients were ultimately included. The incidence of medication deviation at the 4 time points was 37.5%，63.2%，70.7%，and 79.8%，respectively. Totally 3 distinct latent classes were identified：a rapid deterioration group（34.0%），a stable maintenance group（29.3%），and a persistent deviation group（36.8%）. Drug literacy，frailty，educational level，duration of illness，number of medications，and whether percutaneous coronary inter-vention（PCI） was received were influencing factors for the latent classes of medication deviation trajectories（P<0.05）.Conclusion Medication deviation in elderly patients with CHD shows a general upward trend with heterogeneous developmental trajectories among subgroups. Healthcare professionals can carry out targeted interventions based on the influencing factors of these trajectories.

Key words: Coronary Heart Disease, Elderly, Medication Deviation, Latent Class Growth Model, Longitudinal Studies, Nursing Care

摘要：

目的 探究老年冠心病患者用药偏差变化轨迹的潜在类别及其影响因素，为临床开展针对性干预提供理论依据。方法 采用便利抽样法，选取2024年3月至2025年3月在桂林市某三级甲等综合医院住院的285例老年冠心病患者作为调查对象。采用一般资料调查表、衰弱筛查量表、药物素养调查问卷进行基线调查，使用用药偏差评估工具于患者出院后1周（T1）、出院后1个月（T2）、出院后3个月（T3）、出院后6个月（T4）4个时间点进行调查。运用潜类别增长模型识别老年冠心病患者用药偏差变化轨迹的潜在类别，使用多元Logistic回归分析探究其影响因素。结果 最终共纳入患者253例，老年冠心病患者4个时间点用药偏差发生率分别为37.5%、63.2%、70.7%、79.8%。共识别出3个潜在类别，分别为快速恶化组（34.0%）、稳定维持组（29.3%）、持续偏差组（36.8%）。药物素养、衰弱情况、文化程度、患病年限、服药种类、是否接受经皮冠状动脉介入治疗是老年冠心病患者用药偏差变化轨迹潜在类别的影响因素（均P<0.05）。结论 老年冠心病患者用药偏差总体呈逐渐上升趋势，且存在群体异质性发展轨迹，医护人员可根据变化轨迹的影响因素进行针对性干预。

关键词: 冠心病, 老年人, 用药偏差, 潜类别增长模型, 纵向研究, 护理