Document Type


Publication Date



Maine Medical Center, Medical Education, Maine Medical Center Research Institute, Pen Bay Medical Center

MeSH Headings

Blood Coagulation, Health Services, Anticoagulants


Background: There is a 4.5% annual risk of a thromboembolic stroke in patients with atrial fibrillation (AF) and a CHADS2-VASC score 2 or greater. Among these patients, anticoagulation using Warfarin or a direct oral anticoagulant has been shown to reduce the risk of events by 45%. However, there has been some provider concern about the increased risk of bleeding while on anticoagulation in patients who are older, frailer, and have multiple comorbidities. Previous studies have suggested that anticoagulation is recommended in nearly all patients with a CHADS2-VASC score at 2 or greater. Understanding factors that make patients poor candidates for anticoagulation may help providers with their decision to start patients on therapy. This study attempted to identify characteristics or risk factors that affect anticoagulation treatment success among older patients at Pen Bay Medical Center (PBMC).

Methods: Patients aged 70 and older treated with warfarin and managed by the Anticoagulation Services at PBMC were followed for 24 months. Data collected at enrollment and at every 6 months included patient demographics, clinical characteristics, and time in therapeutic range (TTR), a percentage of time that a patient’s International Normalized Ratio (INR) values were in range over a 6-month period, and the primary outcome of interest. TTR was dichotomized at the end of the study as in range (>=60%) or out of range (<60%). Chi-square tests of association were performed to compare characteristics between those in range at least 60% of the time vs. who were not at study end. A regression model is currently being developed in order to identify predictors of TTR in range/out of range.

Results: This study enrolled 132 patients aged 70 and older seen at the Anticoagulation Services at PBMC over a 24-month period. Males were significantly more likely than females to be in range at least 60% of the time at study end (92% vs. 72%, respectively, p=0.004). There was no statistically significant difference in age category (p=0.74), mini Cognitive scores (p=0.26), Frailty scores (p=0.09), mobility scale (p=0.51), CHADS2-VASC (p=0.78), living arrangement (p=0.09), or alcohol use (p=0.06) between the two groups.

Conclusions: Preliminary results from this analysis indicate that age and other factors among elderly patients, including frailty, cognitive ability, or mobility do not affect whether anticoagulation is successful at the end of a 24-month period. Further work is underway to develop a model to understand how these factors may interact with each other to predict anticoagulation treatment outcomes in this patient population.


2020 Costas T. Lambrew Research Retreat