Citations

Clear

4093 citation(s) found for "Yu, Jonathan".

Abstract 2597: Centralized and Study-specific Phenotyping in a Large-scale Stroke Genetics Study: An Analysis from the NINDS Stroke Genetics Network (SiGN)
Worrall, Bradford B. , Gamble, Dale M., Barrett, Kevin M. , Malik, Shaneela , Ossi, Raid G., Biffi, Alessandro , Helenius, Joanna, Huq, Mohammed Saiful, Katsnelson, Michael , Kittner, Steven J. , Furie, Karen L., Cortellini, Lynelle, Woo, Daniel , Markus, Hugh S., Sacco, Ralph L., McArdle, Patrick F. , Ay, Hakan , Brown, Robert D. , Rabinstein, Alejandro A. , Southerland, Andrew M. , Arsava, Ethem Murat , Lorenzano, Svetlana , Huq, Mohammed, Meschia, James F. , Rosand, Jonathan M. , Rundek, Tatjana

Mayo Clinic

Stroke | July 3, 2021 | Volume: 43 | Issue: suppl_1 | Pages: No page available

10.1161/str.43.suppl_1.a2597 Get It

Background: The Stroke Genetics Network (SiGN) funded by the NINDS aims to identify genetic risk factors in ischemic stroke using whole-genome association studies (GWAS). High quality phenotyping is crucial to successful application of GWAS. As a heterogenous disorder, stroke poses specific challenges. The Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification is a broadly used, but its validity is challenged especially when performed by multiple investigators with differing interpretations of the system. The Causative Classification System for Ischemic Stroke (CCS) system is a new, web-based, and computerized algorithm that integrates clinical, diagnostic, and etiologic stroke characteristics in an evidence-based manner ( ccs.mgh.harvard.edu ) to generate subtypes. Methods: In planning the SiGN proposal, a sample of 20 coded charts were collected from a subset of participating studies to assess feasibility of central adjudication and comparability to study-specific TOAST. Two central adjudicators reviewed all records and generated TOAST and CCS subtypes. These were compared to study-specific TOAST subtype and the CCS phenotype generated for SiGN by local trained adjudicators. CCS data is now available for 7134 included cases using both a 5 and a 7 category system as defined in the table . Results: All 4 phenotypes were available for 115 ischemic stroke cases from 6 studies in SiGN. Basic demographics were 54% women, 63% white, and median age between 65-74. Table 1 provides the agreement between the various subtypes. Table 2 describes the types of disagreement. Conclusions: Central adjudication with only two adjudicators and curated medical records yielded more consistent subtyping independent of phenotyping system. The agreement for TOAST was higher than published rates by independent groups (∼0.50). In contrast, the agreement for CCS was lower than previously published (0.85-0.95). Site adjudicators' familiarity with TOAST and inexperience with CCS may contribute. Although CCS is an automated algorithm and has a number of user friendly features, our findings suggest that formal training and certification process before starting to use CCS may be worthwhile to achieve optimal benefit from the system.
Abstract W MP67: Etiology-Specific Clinical and Imaging Features in Ischemic Stroke
Brown, Robert D. , Kittner, Steven J. , Woo, Daniel , Worrall, Bradford B. , Ay, Hakan , Arsava, Ethem Murat , Meschia, James F. , Mitchell, Braxton D. , Rexrode, Kathryn M. , Rosand, Jonathan M. , Rundek, Tatjana

Mayo Clinic

Stroke | July 3, 2021 | Volume: 45 | Issue: suppl_1 | Pages: No page available

10.1161/str.45.suppl_1.wmp67 Get It

Background: Identification of clinical and imaging characteristics of ischemic stroke that can predict the underlying etiology is critical for accurate stroke subtyping. We sought to investigate the relationship between stroke characteristics and stroke etiology in 7,591 patients in the NINDS Stroke Genetics Network Study. Methods: We determined the probability of having a major etiology in the presence of certain stroke characteristics that were previously linked to specific stroke mechanisms. We excluded patients with multiple competing major etiologies. All patients had brain imaging, intra- and extra-cranial vascular imaging, and complete cardiac assessment. We defined major etiologies according to the Causative Classification of Stroke System Criteria without reference to the stroke characteristics under study. Results: History of 1 ipsilateral ischemic events within the last month, the presence of internal watershed infarcts or multiple, ipsilateral, unilateral, and temporally separate infarcts were associated with large artery atherosclerosis (Table). Multiple acute infarcts in different cerebral circulations were more common in patients with major cardiac source of embolism as well as in those with uncommon/cryptogenic causes. Stereotypic lacunar TIAs within the last week and lacunar syndrome were associated with lacunar infarction Discussion: Clinical and imaging stroke characteristics appear to predict stroke etiologies. This information may find utility in developing algorithms for a guided stroke evaluation as well as help identify the most likely causative etiology in the presence of multiple competing etiologies.
Abstract 130: Clinically Important Difference in the Atrial Fibrillation Effect on QualiTy-of-Life (AFEQT) Score: Results from the ORBIT-AF Registry
Simon, DaJuanicia N., Thomas, Laine , O’Brien, Emily C. , Kowey, Peter R. , Reiffel, James A. , Piccini, Jonathan P. , Fonarow, Gregg C. , Gersh, Bernard J. , Naccarelli, Gerald V. , Spertus, John A. , Peterson, Eric D.

Mayo Clinic

Circulation: Cardiovascular Quality and Outcomes | July 3, 2021 | Volume: 9 | Issue: suppl_2 | Pages: No page available

10.1161/circoutcomes.9.suppl_2.130 Get It

Background: The Atrial Fibrillation Effect on QualiTy-of-Life (AFEQT) survey has recently been validated to measure the impact of atrial fibrillation on patients’ quality of life, but a clinically important difference (CID) in AFEQT score has not been defined. Knowing the CID is needed to interpret the meaningfulness of differences between treatments in clinical trials; or patient populations for quality assessment. Objectives: To calculate CID values in AFEQT in the ORBIT registry. Methods: ORBIT-AF is a US-based outpatient AF registry that measured disease-specific QoL with the AFEQT tool (score range= 0 (worst) to 100) at baseline and at 1 year follow-up. Two anchor-based methods were used to relate changes in AFEQT to clinically important differences in the more established European Heart Rhythm Association (EHRA) measure of functional status. Ranging from 1 (no symptoms) to 4 (disabling), a change of 1 EHRA class was defined as an important change in the anchor. Both the mean change and receiver operating characteristics (ROC) methods were then used to identify CIDs in AFEQT at 1 year follow-up. This was done for both improvement and worsening on the anchor. The mean change method defines a CID as the mean change in AFEQT score among patients with a 1 EHRA class change. The ROC method identifies a CID as the point on the ROC curve that best discriminates patients who experienced an important change in the anchor (≥ 1 EHRA class change) from those who experienced no change. Results: AFEQT was assessed in 2008 AF patients at baseline and 1347 patients at 1 year from 99 US sites participating in ORBIT-AF. CIDs and 95% confidence intervals (CI) corresponding to an improvement in EHRA for the mean change method were 5.4 (3.6, 7.2) AFEQT points and 1.9 (0.4, 9.3) AFEQT points for the ROC method. CIDs corresponding to worsening in EHRA for the mean change method were -4.2 (-6.9,-1.5) AFEQT points and -7.4 (-13.9,-4.6) AFEQT points for the ROC method. Conclusions: Changes in AFEQT as small as 2 points may be clinically relevant, although CIDs vary depending on the method of calculation. The variability suggests identifying a single universal CID to assess improvement in quality of life in AF patients may not be ideal and improvement may relate to the nature of a patient’s symptoms and their baseline level of activity.
Abstract 152: Persistence of Warfarin versus Dabigatran Therapy in Patients with Atrial Fibrillation: Results from the ORBIT AF Registry
Kim, Sunghee , Shrader, Peter, Blanco, Rosalia , Thomas, Laine , Ansell, Jack , Go, Alan S. , Kowey, Peter R. , Burton, Paul , Piccini, Jonathan P. , Jackson, Larry R. , Fonarow, Gregg C. , Gersh, Bernard J. , Mahaffey, Kenneth W. , Hylek, Elaine M. , Peterson, Eric D.

Mayo Clinic

Circulation: Cardiovascular Quality and Outcomes | July 3, 2021 | Volume: 8 | Issue: suppl_2 | Pages: No page available

10.1161/circoutcomes.8.suppl_2.152 Get It

Introduction: Warfarin and non-vitamin K antagonist oral anticoagulants (NOACs) such as dabigatran are effective for the prevention of stroke and systemic embolism in patients with atrial fibrillation (AF). However, few analyses have compared persistence rates of warfarin vs. dabigatran-treated patients. Methods: The ORBIT-AF registry enrolled patients with AF from 173 clinical practices across the US. One-year persistence of warfarin vs. dabigatran was defined as treatment at baseline visit and at 1-year follow-up. Multivariable logistic regression analysis was used to identify characteristics associated with warfarin or NOAC persistent use. Results: At baseline, 6.4% (N=459/7,150) were treated with dabigatran and 93.6% (N=6,691/7,150) with warfarin. Patients treated with warfarin at baseline were older (74 vs. 71, p<.0001), had more NYHA class III/IV heart failure (7 vs. 5, p=.0007), lower creatinine clearance (77 vs. 88, p<.0001), higher CHA2DS2-VASc risk scores, and more prior CVA/TIA events (16 vs. 11, p=.003). Patients treated with dabigatran had more severe symptoms (EHRA class III: 20 vs. 14, p<.0001), higher rates of treatment with a rhythm control strategy (42.7 vs. 28.2, p<.0001), more attempts at cardioversion (38.3 vs. 32.1, p<.006), and catheter ablation of AF (9.8 vs. 5.2, p<.0001). At 12 months, adjusted persistence rates for warfarin were higher than dabigatran [82% (80-84) vs. 67% (61-73) p<.0001]. Factors independently associated with one year persistence were: African American race (OR 1.53, 95% C.I. 1.07-2.19, p=0.02), Hispanic race (OR 1.66, 95% C.I. 1.06-2.60), paroxysmal AF (vs. new onset) (OR 1.68, 95% C.I. 1.21-2.33, p=.002), LVH (OR 1.40, 95% C.I. 1.08-1.81, p=.01), persistent AF (vs. new onset) (OR 1.91, 95% C.I. 1.35-2.69, p=.0002), and CHA2DS2-VASc risk scores ≥ 2 (OR 1.94, 95% C.I. 1.18-3.19, p=.009). Conclusion: Persistence rates for warfarin were higher at one year than those on dabigatran. In addition, factors associated with persistence of warfarin include: African American and Hispanic race, type of AF including permanent and persistent, LVH, and CHA2DS2-VASc risk scores ≥ 2.
Abstract 14: Clinical Characteristics and Treatment Patterns of Medicaid Patients with Atrial Fibrillation: Insights From the ORBIT-AF I Registry
O’Brien, Emily C. , Kim, Sunghee , Thomas, Laine , Kowey, Peter R. , Burton, Paul S, Piccini, Jonathan P. , Fonarow, Gregg C. , Mahaffey, Kenneth W. , Gersh, Bernard J. , Peterson, Eric D.

Mayo Clinic

Circulation: Cardiovascular Quality and Outcomes | July 3, 2021 | Volume: 8 | Issue: suppl_2 | Pages: No page available

10.1161/circoutcomes.8.suppl_2.14 Get It

Background: Little is known about whether insurance status affects the presentation and treatment of patients with atrial fibrillation (AF). Methods: We used data from the ORBIT-AF Registry (2010 - 2011), a national, large outpatient registry, to evaluate clinical characteristics and oral anticoagulation (OAC) use. We examined differences in comorbidities and receipt of OAC among patients enrolled in the Medicaid program at baseline compared with those not enrolled in Medicaid. After restricting to patients who were taking OAC, we compared receipt of novel oral anticoagulants (dabigatran or rivaroxaban) versus warfarin by Medicaid status during 2 years of followup. We used logistic regression models adjusting for demographic and clinical covariates to evaluate baseline OAC receipt by Medicaid status. Results: Of 10,133 patients, N=470 (4.6%) were classified as having Medicaid insurance. Compared with those with other insurance, Medicaid patients were younger (70.0 years; IQR=61.0 - 79.0 vs. 75.0 years; IQR=67.0 - 82.0), more likely to be female (53.0% vs. 41.8%), and less likely to be white (58.7% vs. 90.7%; all p<0.001). Medicaid patients had higher rates of smoking, prior stroke/TIA, diabetes, hypertension, and HF. Medicaid patients were less likely to be taking OAC at baseline, a difference that was particularly pronounced for high stroke risk patients (CHADS2>=2, p<0.001; Table). Among those on anticoagulation, Medicaid patients were less likely to receive NOAC over 2 years of followup. In adjusted analyses, Medicaid patients were less likely to receive OAC at baseline, but this difference was not statistically significant (HR = 0.82; 95% CI = 0.61, 1.09). Among untreated patients with CHADS2>=2, Medicaid patients were more likely than patients with other insurance to have “unable to adhere/monitor” listed as a contraindication to OAC (p<0.001). Conclusions: In a contemporary, community-based AF cohort, Medicaid patients had a greater comorbidity burden and higher stroke risk, yet were less likely to receive OAC compared with those with other forms of insurance.
Abstract P20: One-Year Hospitalization Costs Associated with Carotid Stenting Versus Endarterectomy: Results from the CREST Trial
Vilain, Katherine , Magnuson, Elizabeth A. , Sternbergh, Charles, Clark, Wayne M. , Begg, Richard J., Sam, Albert D., Weaver, Fred A., Brott, Thomas G. , Li, Haiyan , Cohen, David Jonathan

Mayo Clinic

Circulation: Cardiovascular Quality and Outcomes | July 3, 2021 | Volume: 4 | Issue: suppl_2 | Pages: No page available

10.1161/circoutcomes.4.suppl_2.ap20 Get It

Background: The CREST trial demonstrated that for patients at standard risk of surgical complications, there was no significant difference in the primary composite outcome of periprocedural death, MI, or stroke, or late ipsilateral stroke between carotid artery stenting (CAS) and carotid endarterectomy (CEA), although CAS had a higher rate of stroke, and CEA had a higher rate of MI. The economic implications of these two strategies are unknown. Methods: We performed a prospective health economic study alongside the CREST trial. Costs were assessed from the perspective of the US health care system in 2008 dollars using a combination of resource-based and event-based methods. Costs for carotid revascularization procedures were based on measured resource use and unit costs derived from a sample of study hospitals. Non-procedural costs for these hospitalizations were estimated using hospital billing data (charges) and cost-center-specific cost-to-charge-ratios. Costs for follow-up events were estimated using national average DRG reimbursements. The primary analysis was based on a modified intention-to-treat population for which the assigned procedure was attempted (n=1212 CAS; 1193 CEA). Results: CAS was associated with higher total procedural costs (Δ=$675, see Table), driven mainly by higher costs for disposable supplies. Length of stay was shorter for CAS, with associated reductions in non-procedural hospital costs (Δ = -$436). Total cost for the index hospitalization remained slightly higher for CAS (Δ=$239) with similar differences at 1 year. Conclusions: For patients at standard risk of surgical complications, total 1-year costs are slightly higher for CAS vs. CEA, driven largely by higher initial procedural costs. Cost-utility analysis will be performed to determine whether differences in quality of life observed in CREST render CAS an economically attractive strategy. CAS (n= 1212) CEA (n=1193) Δ CAS - CEA (95% CI via bootstrap) Index Procedure Costs, $ Excluding MD Fees 6782 ± 1412 5743 ± 1370 1039 (926, 1148) MD Fees 1114 ± 240 1478 ± 108 -364 (-379, -349) Total 7896 ± 1551 7221 ± 1450 675 (555, 800) Index Hospital Stay 7159 ± 5108 7595 ± 7482 -436 (-951, -59) TOTAL Index Hosp Costs 15055 ± 5539 14816 ± 7709 239 (-302, 778) Index Hosp LOS (days) ICU LOS 0.7 ± 1.1 0.8 ± 1.4 -0.1 (-0.24, -0.04) Non-ICU LOS 1.9 ± 3.2 2.2 ± 4.1 -0.2 (-0.55, 0.06) TOTAL LOS 2.6 ± 3.3 3.0 ± 4.5 -0.4 (-0.7, -0.06) 1-year Rehosp Costs: Repeat Revasc - CAS 295 ± 2097 273 ± 2126 -21 (-142, 201) Repeat Revasc - CEA 287 ± 2088 477 ± 2510 -190 (-371, -11) Ipsilateral Stroke 556 ± 3273 384 ± 2580 172 (-75, 402) TIA 48 ± 520 33 ± 487 16 (-25, 56) Death 63 ± 846 45 ± 779 18 (-49, 85) TOTAL Rehosp costs 1321 ± 4827 1293± 4502 28 (-334, 396) TOTAL 1-year costs 16375 ± 7730 16108 ± 9030 267 (-366, 961)
Abstract 336: Real-World Experiences with Novel Anticoagulants for AF: Initial Insights from ORBIT-AF Phase II
Milford–Beland, Sarah, Ollis, Donna M, Blanco, Rosalia , Ansell, Jack , Go, Alan S. , Kowey, Peter R. , Thomas, Laine , Chang, Paul, Piccini, Jonathan P. , Steinberg, Benjamin A. , Fonarow, Gregg C. , Gersh, Bernard J. , Hylek, Elaine M. , Mahaffey, Kenneth W. , Peterson, Eric D.

Mayo Clinic

Circulation: Cardiovascular Quality and Outcomes | July 3, 2021 | Volume: 7 | Issue: suppl_1 | Pages: No page available

10.1161/circoutcomes.7.suppl_1.336 Get It

Background: Novel therapies for stroke prevention in atrial fibrillation (AF) are emerging. However, data on use and outcomes of these agents in the community are limited, particularly in patients newly diagnosed with AF or recently transitioned to a novel anticoagulant. Methods: Phase II of the Outcomes Registry for Better Informed Treatment of Atrial Fibrillation (ORBIT-AF II) is a multicenter, national, AF registry. We are enrolling 15,000 patients with new-onset AF, or those with known AF recently-transitioned to a novel anticoagulant. The primary objective is to evaluate the safety of novel anticoagulants, with specific focus on use, transitions, and management around invasive procedures and bleeding events. An adaptive design is used to ensure diversity of enrollment across geographic, provider, and treatment groups. The primary outcomes are major bleeding and thromboembolic events. Results: Of 1,011 patients in ORBIT-AF II, the median age was 73, 44% were female, 23% had a history of heart failure, 34% had a prior MI, and 11% had a prior stroke/TIA. The mean CHADS2 score was 2. Overall, 76% had AF new-onset, and a range of anticoagulant strategies was used in these patients (Figure). Among the 24% with known AF transitioned to a novel agent, 66% received rivaroxaban, versus 8.1% dabigatran, and 26% apixaban. Compared with patients having new-onset AF, those with known AF recently-transitioned to a novel anticoagulant were slightly older (median age 74 vs. 72), had higher CHADS2 scores (>2 in 70% vs. 62%), but less severe kidney disease (0 vs. 1.7%). Conclusions: Patients with new-onset AF are frequently treated with novel anticoagulants and these patients have lower risk scores than those with known AF transitioned to a novel drug. These data will provide insights into clinical management and patient centered outcomes among those treated with novel anticoagulants in real world practice settings.
Abstract 325: Paradoxical Relationship Between Bleeding Risk, Stroke Risk, and Time in Therapeutic Range for Atrial Fibrillation Patients on Warfarin: Data from ORBIT-AF
Thomas, Laine , Kowey, Peter R. , Chang, Paul, Singer, Daniel E. , Ansell, Jack , Blanco, Rosalia , Go, Alan S. , Piccini, Jonathan P. , Pokorney, Sean D. , Holmes, DaJuanicia N. , Fonarow, Gregg C. , Gersh, Bernard J. , Mahaffey, Kenneth W. , Hylek, Elaine M. , Peterson, Eric D.

Mayo Clinic

Circulation: Cardiovascular Quality and Outcomes | July 3, 2021 | Volume: 7 | Issue: suppl_1 | Pages: No page available

10.1161/circoutcomes.7.suppl_1.325 Get It

Objective: Time in the therapeutic range (TTR) of INR 2.0-3.0 is an important measurement of warfarin anticoagulation quality. There are few reports on TTR in United States (US) clinical practice. Methods: Using the US ORBIT AF registry, we analyzed TTR for patients on warfarin with at least 5 recorded INR measurements. The Rosendaal method of linear interpolation was used between each pair of measured INR values. Patients were stratified into quartiles based on TTR. Results: We studied 5,451 patients treated at 169 US sites with 48% of patients being followed at an anticoagulation clinic. Patients had a median age of 76 years [interquartile range (IQR) 68-82 years] and median follow-up was 18.9 months [IQR 12.7-23.3 months], and patients were 42% women. The median number of INR draws was 21 [IQR 12-30]. Patients had an overall median TTR of 68% [IQR 52-80%]. The median TTR by quartile was 40%, 60%, 73%, and 87%. Patients were more likely to be subtherapeutic (median time of 17%) than supratherapeutic (median time of 10%). Patients age 61-80 years and those with a college education had the highest TTRs. Patients with their INR followed at an anticoagulation clinic were more likely to be in the highest TTR quartile than in the lowest (27% vs 23%, p=0.0002). Patients with worsening renal function and higher risk for bleeding (ATRIA) or stroke (CHADS2 score) had lower TTR (p<0.0001 for all) (Figure). Conclusions: In US clinical practices, warfarin patients have therapeutic INR levels about 2/3 of the time. Those at highest risk for stroke and bleeding paradoxically were the least likely to be in the therapeutic range.
Abstract 77: Balancing Risks and Benefits of Anticoagulation Therapy in Atrial Fibrillation in Community Practice: Results from the ORBIT-AF Registry
Cullen, Michael W. , Kim, Sunghee , Piccini, Jonathan P. , Go, Alan S. , Ansell, Jack , Kowey, Peter R. , Fonarow, Gregg C. , Mahaffey, Kenneth W. , Lópes, Renato D. , Peterson, Eric D. , Gersh, Bernard J.

Mayo Clinic

Circulation: Cardiovascular Quality and Outcomes | July 3, 2021 | Volume: 5 | Issue: suppl_1 | Pages: No page available

10.1161/circoutcomes.5.suppl_1.a77 Get It

Background Oral anticoagulation (OAC) can reduce stroke risk at the cost of increased bleeding risk in those with atrial fibrillation (AF). Observational data have shown that higher-risk patients with AF most likely to benefit from OAC are less likely to receive OAC at hospital discharge. Methods We used data from ORBIT-AF Registry, a cohort of 9,589 AF patients enrolled among 173 participating outpatient practices. OAC was defined as warfarin or dabigatran use at study enrollment. Stroke and bleeding risk were calculated using the CHADS2 and ATRIA scores, respectively. Results The study population had a mean age of 73.5 years; 57.8% were men. Overall, 76.4% of patients received OAC. Use of OAC rose with increasing CHADS2 stroke risk, from 67% for CHADS2 <1 to 80% for CHADS2 ≥2 (p<0.0001). OAC use fell slightly with increasing ATRIA bleeding risk, from 77% for ATRIA score ≤3 to 74% with ≥5 (p=0.002 for trend). Among patients with low bleeding risk, rates of OAC increased commensurate with stroke risk (p<0.0001 for interaction; see figure). Higher bleeding risk tended to decrease rates of OAC among patients with a CHADS2 score ≥2 (p=0.13 for interaction). Conclusions In community-based outpatients with AF, use of OAC rose with increasing thromboembolic risk and declined with higher bleeding risk. These findings suggest that the risk-treatment paradox may be less that previously reported. Provision of OAC in community practice appears to appropriately consider patients' stroke and bleeding risks. Further research is required to understand how quality improvement initiatives can further improve stroke prevention.
Abstract 17182: Patterns of Discontinuation for Non-vitamin K Oral Anticoagulants in Atrial Fibrillation: Results From the ORBIT-AF II Registry
Simon, DaJuanicia N., Thomas, Laine , Ansell, Jack , Kowey, Peter R. , Piccini, Jonathan P. , Steinberg, Benjamin A. , Gersh, Bernard J. , Fonarow, Gregg C. , Hylek, Elaine M. , Mahaffey, Kenneth W. , Peterson, Eric D.

Mayo Clinic

Circulation | July 3, 2021 | Volume: 132 | Issue: suppl_3 | Pages: No page available

10.1161/circ.132.suppl_3.17182 Get It

Background: Oral anticoagulation (OAC) is effective at preventing stroke in patients with atrial fibrillation (AF), yet warfarin is often poorly tolerated. Non-vitamin K oral anticoagulants (NOACs) are as or more effective as warfarin, yet their tolerance and persistence in clinical practice is not known. Methods: We assessed patterns of persistent OAC use among 2,345 AF patients starting on therapy in the ORBIT-AF II registry (71% starting on a NOAC, and 29% on warfarin). Results: By 6 months, 364 (22%) patients started on a NOAC had discontinued or changed initial therapy versus 143 (21%) started on warfarin initially (p=0.5). Among warfarin users, patients who switched or discontinued therapy were of similar age (median ages 72 and 74 vs. 74 for stable users, p=0.7) and CHA2DS2-VASc scores (mean 98 and 3.66 vs. 3.84, p=0.4). Among NOAC users, those who discontinued treatment were younger (median age 68 vs. 73 for those who switched and 72 for stable users; p=0.0004), and lower CHA2DS2-VASc scores (3.02 vs. 3.58 and 3.47, respectively; p=0.0008). The median time to change or discontinuation was more rapid in those started on a NOAC vs warfarin (97 days vs. 122 days, p=0.003). Among those on warfarin at baseline, 7.6% (n=52) were switched to a NOAC within 6 months, whereas transitions from NOAC to warfarin was 2.5% (n=42).Transitions among NOACs occurred in 9.8%, 3.2%, and 5.5% of patients on baseline dabigatran, rivaroxaban, and apixaban, respectively. Physician preference was the most common reason for both OAC and warfarin changes (Table). Drug cost was the primary reason for change of therapy in 15% of NOAC users (vs. 0 for warfarin). Conclusions: At 6-month follow-up, one in five newly started on OAC had discontinued or changed. These rates of change were similar among warfarin and NOAC treated patients. Cost concerns drove discontinuation in a modest number of patients, however, cost concerns were more prevalent in NOAC-treated patients.