Center for Pharmaceutical Policy and Prescribing, University of Pittsburgh, Pittsburgh, Pennsylvania; Department of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania.
Department of Pharmacy and Therapeutics, University of Pittsburgh, Pittsburgh, Pennsylvania.
Am J Cardiol. 2020 Sep 1;130:64-69. doi: 10.1016/j.amjcard.2020.06.015. Epub 2020 Jun 17.
We aimed to assess the association between urban/rural residence and the risk of ischemic stroke in individuals with newly diagnosed atrial fibrillation (AF), and to quantify the role of oral anticoagulation (OAC) initiation in the variation in stroke risk between urban and rural residents with AF. Using 5% random samples of Medicare claims, we identified fee-for-service beneficiaries who were diagnosed with AF between January 2014 and December 2015. Beneficiaries were followed for 1 year since their AF diagnosis, and were categorized according to their initiation of OAC within the year. We used the Rural-Urban Continuum Codes to define urban (levels 1 to 3) and rural (levels 4 to 9) areas. We applied marginal structural models to examine to what extent the difference in stroke risk between rural and urban areas were attributable to the difference in OAC initiation. In the year of AF diagnosis, 52% of those residing in urban areas and 56% residing in rural areas initiated an OAC (p <0.001). Urban residence, compared with rural residence, was associated with a 22% (hazard ratio and 95% confidence interval: 1.22 [1.13, 1.31]) increased risk of stroke. The hazard ratio attributed to urban residence decreased to 1.14 (1.01, 1.30) after accounting for the mediating role of lack of OAC initiation. Lack of OAC initiation explained 34% of the increased stroke risk observed in urban areas. In conclusion, urban residents with newly diagnosed AF were less likely to initiate OAC than rural counterparts, which explained one third of the increased risk of stroke observed in urban areas.
我们旨在评估城乡居住与新发心房颤动(AF)个体缺血性卒中风险的关联,并定量评估口服抗凝剂(OAC)的启动在 AF 城乡居民卒中风险差异中的作用。我们使用医疗保险索赔的 5%随机样本,确定了在 2014 年 1 月至 2015 年 12 月期间被诊断为 AF 的按服务收费受益人的数据。自 AF 诊断后,对受益人进行了 1 年的随访,并根据其在一年内启动 OAC 的情况进行了分类。我们使用农村-城市连续体代码来定义城市(1 级至 3 级)和农村(4 级至 9 级)地区。我们应用边缘结构模型来检查城乡地区卒中风险差异在多大程度上归因于 OAC 启动的差异。在 AF 诊断的当年,52%居住在城市地区的人和 56%居住在农村地区的人启动了 OAC(p<0.001)。与农村地区相比,城市居住与卒中风险增加 22%相关(风险比和 95%置信区间:1.22[1.13,1.31])。在考虑到缺乏 OAC 启动的中介作用后,城乡居住相关的风险比降低至 1.14(1.01,1.30)。缺乏 OAC 启动解释了在城市地区观察到的卒中风险增加的 34%。总之,与农村居民相比,新发 AF 的城市居民更不可能启动 OAC,这解释了城市地区卒中风险增加的三分之一。