Mårdby Ann-Charlotte, Schiöler Linus, Sundell Karolina Andersson, Bjerkeli Pernilla, Lesén Eva, Jönsson Anna K
Research and Development, Sahlgrenska University Hospital, Röda Stråket 8, 413 45, Gothenburg, Sweden.
Section of Epidemiology and Social Medicine, Department of Public Health and Community Medicine at Institute of Medicine, University of Gothenburg, Box 453, 405 30, Gothenburg, Sweden.
Eur J Clin Pharmacol. 2016 Nov;72(11):1381-1389. doi: 10.1007/s00228-016-2106-1. Epub 2016 Aug 3.
The purpose of this study are to analyse adherence to antidepressant treatment over 2 years in Sweden among women and men who initiated treatment with citalopram and to identify groups at risk of non-adherence using trajectory models.
The study population, including individuals 18-85 years who initiated citalopram use between 1 July 2006 and 30 June 2007, was identified in the Swedish Prescribed Drug Register and followed for 2 years. Adherence was estimated with continuous measure of medication acquisition (CMA) and group-based trajectory modelling, a method which describes adherence patterns over time by estimating trajectories of adherence and the individual's probability of belonging to a specific trajectory.
The study population included 54,248 individuals, 64 % women. Mean CMA was 52 % among women and 50 % among men (p < 0.001). Five different adherence patterns (Trajectories) were identified. Similar proportion of women and men belonged to each Trajectory. Around 29 % of the women and 27 % of the men belonged to the Trajectory which showed full adherence throughout the 2-year study period. The other four Trajectories showed adherence that declined to different degrees and at different stages in time. Having low socioeconomic status was more common among individuals in Trajectories showing declining adherence than in the adherent Trajectory.
Using trajectory modelling, five Trajectories describing different patterns of adherence to citalopram treatment over time were identified. A large proportion discontinued treatment early and having low socioeconomic status increased the risk of being non-adherent.
本研究旨在分析瑞典使用西酞普兰开始治疗的女性和男性在两年内对抗抑郁药物治疗的依从性,并使用轨迹模型识别存在不依从风险的群体。
在瑞典处方药登记册中确定了研究人群,包括2006年7月1日至2007年6月30日期间开始使用西酞普兰的18 - 85岁个体,并对其进行了两年的随访。通过药物获取连续测量(CMA)和基于群体的轨迹建模来估计依从性,该方法通过估计依从性轨迹和个体属于特定轨迹的概率来描述随时间的依从模式。
研究人群包括54248名个体,其中64%为女性。女性的平均CMA为52%,男性为50%(p < 0.001)。识别出五种不同的依从模式(轨迹)。女性和男性属于每个轨迹的比例相似。约29%的女性和27%的男性属于在两年研究期间表现出完全依从的轨迹。其他四个轨迹显示出在不同时间阶段不同程度下降的依从性。在依从性下降的轨迹中的个体比在依从轨迹中的个体社会经济地位低更为常见。
使用轨迹建模,识别出了五种描述随时间对西酞普兰治疗不同依从模式的轨迹。很大一部分人早期停止治疗,社会经济地位低会增加不依从的风险。