Australian Institute of Health Innovation, Macquarie University, North Ryde, New South Wales, Australia.
Br J Clin Pharmacol. 2024 Dec;90(12):3308-3319. doi: 10.1111/bcp.16220. Epub 2024 Aug 25.
Polypharmacy serves as a quality indicator in residential aged care facilities (RACFs) due to concerns about inappropriate medication use. However, aggregated polypharmacy rates at a single time offer limited value. Longitudinal analysis of polypharmacy patterns provides valuable insights into identifying potential overuse of medicines. We aimed to determine long-term trajectories of polypharmacy (≥9 medicines) and factors associated with each polypharmacy trajectory group.
This was a longitudinal cohort study using electronic data from 30 RACFs in New South Wales, Australia. We conducted group-based trajectory modelling to identify and characterize polypharmacy trajectories over 3 years. We evaluated the model fitness using the Bayesian Information Criterion, entropy (with a value of ≥0.8 considered ideal) and several other metrics.
The study included 2837 permanent residents (median age = 86 years, 61.7% female and 47.4% had dementia). We identified five polypharmacy trajectory groups: group 1 (no polypharmacy, 46.0%); group 2 (increasing polypharmacy, 9.4%); group 3 (decreasing polypharmacy, 9.2%); group 4 (increasing-then decreasing polypharmacy, 10.0%), and group 5 (persistent polypharmacy, 25.4%). The model showed excellent performance (e.g., entropy = 0.9). Multinomial logistic regressions revealed the profile of each trajectory group (e.g., group 5 residents had higher odds of chronic respiratory disease compared with group 1).
Our study identified five polypharmacy trajectory groups, including one with over a quarter of residents following a persistently high trajectory, signalling concerning medication overuse. Quality indicator programs should adopt tailored metrics to monitor diverse polypharmacy trajectory groups, moving beyond the current one-size-fits-all approach and better capturing the evolving dynamics of residents' medication regimens.
由于担心药物使用不当，在养老院（RACF）中，多种药物治疗被用作质量指标。然而，在单一时间点汇总的多种药物治疗率提供的价值有限。对多种药物治疗模式的纵向分析为识别潜在的药物过度使用提供了有价值的见解。我们旨在确定长期多种药物治疗（≥9 种药物）的轨迹以及与每种多种药物治疗轨迹组相关的因素。
这是一项使用澳大利亚新南威尔士州 30 个养老院的电子数据进行的纵向队列研究。我们使用基于群组的轨迹建模来确定和描述 3 年内的多种药物治疗轨迹。我们使用贝叶斯信息准则、熵（值≥0.8 被认为理想）和其他几个指标来评估模型拟合度。
该研究纳入了 2837 名常住居民（中位数年龄为 86 岁，61.7%为女性，47.4%患有痴呆症）。我们确定了五种多种药物治疗轨迹组：组 1（无多种药物治疗，46.0%）；组 2（多种药物治疗逐渐增加，9.4%）；组 3（多种药物治疗逐渐减少，9.2%）；组 4（多种药物治疗先增加后减少，10.0%）和组 5（持续多种药物治疗，25.4%）。该模型表现出优异的性能（例如，熵=0.9）。多项逻辑回归揭示了每个轨迹组的特征（例如，与组 1 相比，组 5 的居民患有慢性呼吸道疾病的可能性更高）。
我们的研究确定了五种多种药物治疗轨迹组，其中包括四分之一以上的居民持续处于较高的药物治疗轨迹，这表明存在药物过度使用的问题。质量指标计划应采用定制的指标来监测不同的多种药物治疗轨迹组，避免一刀切的方法，更好地捕捉居民药物治疗方案的动态变化。
