Skip main navigation

Real-World Adherence and Persistence to Direct Oral Anticoagulants in Patients With Atrial Fibrillation

A Systematic Review and Meta-Analysis
Originally published Cardiovascular Quality and Outcomes. 2020;13:e005969



    Stroke reduction with direct oral anticoagulants (DOACs) in atrial fibrillation (AF) is dependent on adherence and persistence in the real-world setting. Individual study estimates of DOAC adherence/persistence rates have been discordant. Our aims were to characterize real-world observational evidence for DOAC adherence/persistence and evaluate associated clinical outcomes in patients with AF.

    Methods and Results:

    PubMed, EMBASE, and CINAHL were searched from inception to June 2018. Observational studies that reported real-world DOAC adherence/persistence in patients with AF were included. Study quality was assessed using the Newcastle-Ottawa Scale. Meta-analyses for pooled estimates were performed using DerSimonian and Laird random-effects models. Outcomes included DOAC mean proportion of days covered or medication possession ratio, proportion of good adherence (proportion of days covered/medication possession ratio ≥80%), persistence, DOAC versus vitamin K antagonists persistence, and clinical outcomes associated with nonadherence/nonpersistence. Forty-eight observational studies with 594 784 unique patients with AF (59% male; mean age 71 years) were included. The overall pooled mean proportion of days covered/medication possession ratio was 77% (95% CI, 75%–80%), proportion of patients with good adherence was 66% (95% CI, 63%–70%), and proportion persistent was 69% (95% CI, 65%–72%). The pooled proportion of patients with good adherence was 71% (95% CI, 64%–78%) for apixaban, 60% (95% CI, 52%–68%) for dabigatran, and 70% (95% CI, 64%–75%) for rivaroxaban. Similar patterns were found for pooled persistence by agent. The pooled persistence was higher with DOACs than vitamin K antagonists (odds ratio, 1.44 [95% CI, 1.12–.86]). DOAC nonadherence was associated with an increased risk of stroke (hazard ratio, 1.39 [95% CI, 1.06–1.81]).


    Suboptimal adherence and persistence to DOACs was common in patients with AF, with 1 in 3 patients adhering to their DOAC <80% of the time, which was associated with poor clinical outcomes in nonadherent patients. Although it is convenient that DOACs do not require laboratory monitoring, greater effort in monitoring for and interventions to prevent nonadherence may be necessary to optimize stroke prevention. Increased clinician awareness of DOAC nonadherence may help identify at-risk patients.

    What Is Known

    • Direct oral anticoagulants (DOACs) are effective for stroke prevention in patients with atrial fibrillation.

    • Due to short elimination half-lives and short duration of anticoagulant effect of DOACs, even brief periods of DOAC nonadherence may rapidly result in subtherapeutic anticoagulant levels.

    What the Study Adds

    • Suboptimal adherence and persistence to DOACs were common: atrial fibrillation patients were not taking their DOAC 1 out of every 4 days, one-third of patients were adherent <80% of the time, and real-world persistence with DOACs was lower than in randomized control trials.

    • DOAC persistence was higher than vitamin K antagonists.

    • Suboptimal adherence to DOACs was associated with higher risk of stroke.


    Atrial fibrillation (AF) is the most common chronic cardiac arrhythmia, affecting 33.5 million individuals worldwide in 2010 and projected to double by 2030.1,2 AF is associated with a 5-fold increase in the risk of stroke and requires anticoagulant therapy for those with moderate to high-stroke risk.3 Direct oral anticoagulants (DOACs) are effective for AF stroke prevention and are increasing in popularity over traditional vitamin K antagonist (VKA) anticoagulants, due to a reduced need for frequent therapeutic monitoring, improved patient convenience, greater predictability, faster onset of anticoagulation effect, and lower potential for food and drug interactions.4–7

    Concerns have been raised regarding the heightened consequences of poor adherence with DOACs. Due to the shorter elimination half-lives of DOACs compared with warfarin, and resultant short duration of anticoagulant effect, even brief periods of DOAC nonadherence may rapidly result in subtherapeutic anticoagulant levels.8 Poor medication adherence and persistence in the real-world setting may alter the efficacy and safety estimates derived from randomized control trials (RCTs), translating into poor health outcomes and increased health care costs.9,10

    Furthermore, the precise lack of required ongoing monitoring for DOACs may, for practical purposes, predispose patients to nonadherence due to diminished clinician oversight of therapeutic regimens. Although a variety of individual real-world DOAC studies have been published, they have discordant adherence and persistence findings and have not yet been consolidated in a systematic and comprehensive manner to determine the magnitude of this problem. Therefore, given the importance of DOAC adherence and persistence for stroke prevention in AF, we undertook a comprehensive systematic review and meta-analysis to characterize the real-world observational evidence for DOAC adherence/persistence and evaluate associated clinical outcomes in the AF population.


    We conducted a systematic review and meta-analysis in accordance with the Meta-analyses of Observational Studies in Epidemiology guidelines.11 Our study is registered with PROSPERO (URL: Unique identifier: CRD42019110266). The authors declare that all supporting data are available within the article and its in the Data Supplement.

    Study Selection Criteria

    We included studies that assessed real-world DOAC adherence/persistence in patients with AF. We excluded case reports/series, review articles, studies available in languages other than English, Japanese, Korean, Vietnamese, or French, studies that used subjective adherence measures, such as, Morisky Medication Adherence Scale and studies that measured adherence using medication event monitoring systems.12,13

    Search Strategy

    We conducted a comprehensive literature search using PubMed, EMBASE, and CINAHL databases from database inception to June 25, 2018. We did not apply any language restrictions during our search, and our search terms included a combination of MeSH terms and keywords pertaining to adherence/persistence, DOACs, and AF (Methods I in the Data Supplement).

    Article Screening and Data Abstraction

    Two investigators independently and in duplicate screened and extracted study data, and disagreements were resolved by consensus with a third investigator. Studies that met the inclusion criteria based on the abstract were obtained in full-text for further screening. We also manually searched reference lists of all included articles. Extracted study data included author name, year of publication, journal, study design, country and type of data source, data source, study period, sample size, patient characteristics, conflict of interest, adherence measure and definition, persistence definition, follow-up duration, and adherence/persistence measured. We collected persistence data from studies that reported discontinuation rates. For example, for a study reporting a 15% discontinuation rate, the corresponding persistence would be documented as 85%. We collected reported study funding, author-industry financial ties, and author employment.

    Outcomes Assessed

    We extracted 3 outcomes: (1) adherence, either as proportion of days covered (PDC) or medication possession ratio (MPR); (2) good adherence, typically reported as PDC/MPR ≥80%; and (3) persistence, including the study-defined permissible gap. We included follow-up time frames as reported in the included studies, without restriction. When the included studies indicated that they calculated adherence or persistence including patients who switched between anticoagulants during the follow-up period, we excluded these results since our objective was to examine DOAC-specific rather than overall anticoagulant adherence/persistence rates. In addition, hazard ratios (HR) of clinical outcomes associated with DOAC nonadherence/nonpersistence were extracted.

    Study Quality Assessment

    Two independent investigators in duplicate assessed the risk of bias for all included studies using the Newcastle-Ottawa Scale, which consists of 3 main domains: Selection, Comparability, and Outcome.14 A study can be awarded a maximum of 1 star for each numbered item in the selection and outcome categories. A maximum of 2 stars can be given for comparability. The scores were converted to Agency for Healthcare Research and Quality standards thresholds of good, fair, and poor quality.14

    Data Analysis

    Meta-analyses for pooled estimates of mean PDC/MPR, proportion of good adherence, and persistence were performed using a DerSimonian and Laird random-effects model.15 Heterogeneity was assessed using I2 and Cochran Q, with I2>50% indicating high heterogeneity.16 Subgroup analyses were conducted by DOAC agent, region, observation time period, and persistence definition. To further explore sources of heterogeneity, meta-regression was performed by publication year, patient age, proportion of males, proportion of anticoagulant naïve, CHADS2-VASc score, and observation time period.17 Post hoc analyses comparing adherence/persistence between DOACs and VKA were performed. A pooled estimate of clinical outcomes associated with DOAC nonadherence/nonpersistence was generated. The presence of publication bias was assessed using funnel plots and Egger’s tests on meta-analyses that included 10 or more studies.18,19 Sensitivity analysis was performed using the trim and fill method.20 All analyses were performed using R Studio software (version 1.1.463, Boston, MA).


    Study Selection

    Our search strategy identified 1538 articles, and 1148 were retained after duplicates were removed. Following title and abstract review, 1012 were excluded and 136 articles underwent full-text review. Among those, 88 were excluded, leaving 48 studies included in this meta-analysis (Figure 1).

    Figure 1.

    Figure 1. PRISMA flow diagram.aOnly 4 articles had language as the sole exclusion criteria. AF indicates atrial fibrillation; DOAC, direct oral anticoagulant; MEMS, Medication Event Monitoring System; and PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-analysis.

    Study Characteristics

    The 48 included studies comprised a total of 594 784 patients with AF taking DOACs and an additional 283 182 patients taking VKA, with study sample sizes ranging from 159 to 66 090. The included studies were observational studies published between 2013 and 2018, with the majority of studies (n=40; 83%) published within the past 3 years (2016–2018; Table). The majority of studies (n=25; 52%) were from North America, and most studies (n=42; 88%) used prescription claims data as the data source.

    Table. Study Characteristics9,10,21–66

    Author, YearCountryData SourceSample Size,
    Clinical Outcomes Measure
    Anticoagulant AgentFollow-Up Duration, mo
    Mean PDC/MPRPDC/MPR ≥80%PersistenceOutcomesAPIXDABIRIVADOAC
    Annavarapu et al,21 2018USRx claims1025/1025xxxx
    Cataldo et al,22 2018ItalyRx claims765/2909xxxxxxx
    Collings et al,23 2018FranceRx claims2401/1400xxxxxxxx
    Deshpande et al,10 2018USRx claims2981xxxxx
    Deshpande et al,24 2018USRx claims3629x*xxxxx
    Gomez-Lumbreras et al,25 2018SpainRx claims1971xxxxx
    Haastrup et al,26 2018DenmarkRx claims55 116xxxx
    Harper et al,27 2018New ZealandRx claims43 339xxxxx
    Jacobs et al,28 2018Sweden/
    Rx claims5684xxxx
    Lip et al,29 2018USRx claims29 900/15 461xxxxxx
    Maura et al,30 2018FranceRx claims22 267/11 998xxxxx
    Borne et al,31 2017USRx claims2882xxxxxxxx
    Brown et al,32 2017USRx claims15 341xxxxxxxxx
    Collings et al,33 2017GermanyRx claims7013/5532xxxxxxxxx
    Douros et al,34 2017CanadaRx claims14 746/17 685xxxx
    Hellfritzsch et al,35 2017DenmarkRx claims50 623xxxxxxxx
    Hernandez et al,36 2017USRx claims8912/12 353xxxxxx
    Jackevicius et al,9 2017CanadaRx claims25 976xxxxx
    Lamberts et al,37 2017DenmarkRx claims29 821/24 230xxxxxx
    Manzoor et al,38 2017USRx claims66 090xxxxxxxxxxxx
    Maura et al,39 2017FranceRx claims22 267xxxxx
    McHorney et al,40 2017USRx claims23 309/13 366xxxxxxxxxx
    Mueller et al,41 2017ScotlandRx claims5398xxxxxxxxxx
    Paquette et al,42 2017Multiple countriesRegistry2932xxxx
    Song et al,43 2017USRx claims18 980/18 980xxxx
    Sorensen et al,44 2017DenmarkRx claims19 952/26 723xxxxxx
    Alberts et al,45 2016USRx claims36 868xxxxx
    Al-Khalili et al,46 2016SwedenMedical records350xxxxxx
    Bancroft et al,47 2016USRx claims869/869xxxxxxx
    Beshir et al,48 2016MalaysiaMedical records192xxx
    Beyer-Westendorf et al,49 2016GermanyRx claims2138/5127x*xxxxxxx
    Coleman et al,50 2016USRx claims21 756/10 878xxxxxxxx
    Coleman et al,51 2016USRx claims21 756/10 878xxxxxxxxx
    Forslund et al,52 2016SwedenRx claims6127/9969xxxxxx
    Johnson et al,53 2016UKRx claims2344/9303xxxxxxxx
    Kachroo et al,54 2016USRx claims3914/7911xxxxx
    Martinez et al,55 2016UKMedical records914/12 307xxxxxxx
    McHorney et al,56 2016USRx claims5984xxxxxx
    Simons et al,57 2016AustraliaRx claims1471/1348xxxxxxx
    Yao et al,58 2016USRx claims26471/38 190x*xxxxxxxx
    Gorst-Rasmussen et al,59 2015DenmarkRx claims2960xxxx
    Zhou et al,60 2015USRx claims2713xxxxxxx
    Cutler et al,61 2014USMedical records159xxxxxx
    Ho et al,62 2014ChinaRegistry467xxx
    Laliberté et al,63 2014USRx claims3654/14 616xxxxx
    Nelson et al,64 2014USRx claims7259/7259xxxxxx
    Shore et al,65 2014USRx claims5376xxxxxx
    Zalesak et al,66 2013USRx claims1745/1745xxxxxx
    Total594 784/283 1821421365203528172515251038454

    APIX indicates apixaban; DABI, dabigatran; DOAC, direct oral anticoagulant; MPR, medication possession ratio; PDC, proportion of days covered; RIVA, rivaroxaban; Rx, prescription; UK, United Kingdom; US, United States; and VKA, vitamin K antagonist.

    *Studies excluded from pooled estimate since SD was not reported.

    Seventeen studies (35%) reported PDC as the adherence measure, 5 (10%) reported MPR and 36 (75%) reported persistence. Fourteen studies (29%) reported mean PDC/MPR, 21 (44%) reported PDC/MPR ≥80%, and 5 (10%) reported clinical outcomes associated with nonadherence/nonpersistence. Eighteen studies (38%) reported >1 measurement of mean PDC/MPR, PDC/MPR ≥80% or persistence. Twenty studies (42%) evaluated apixaban, 35 (73%) dabigatran, 28 (58%) rivaroxaban, and 17 (35%) evaluated DOACs as a class rather than as individual agents. None of the studies reported edoxaban adherence/persistence individually. Twenty-six studies (54%) evaluated >1 agent. Among the 48 studies included, 25 studies (52%) reported a comparison between DOACs and VKA, with 1 study that reporting PDC ≥80% and 24 studies reporting persistence. Duration of observation time ranged from 3 months to 3 years, with the 12-month period being most commonly reported (Table). There was a wide variation in persistence definitions used in individual studies, with permissible gaps varying from 14 to 365 days. Among 36 studies that reported persistence, half of the studies (n=18) used a gap definition of 56 to 60 days and 33% (n=12) used a gap of 28 to 30 days.

    Half of the studies (n=24) were funded by DOAC pharmaceutical companies (9 studies for apixaban, 5 studies for dabigatran, and 10 studies for rivaroxaban), 8 studies (17%) were nonpharmaceutical company funded (ie, government or university-funded) but with author(s) receiving consultant fees from multiple pharmaceutical companies, and 16 studies (33%) were nonpharmaceutical company funded without consultant fees. Among the studies that evaluated >1 agent (n=26), 14 studies were industry-funded, and 4 out of 7 apixaban-funded studies favored apixaban, and 6 out of 7 rivaroxaban-funded studies favored rivaroxaban (Table I in the Data Supplement).

    Patient Characteristics

    The mean age of patients across the studies was 71 (range 63–79), and 59% of patients were male. The mean CHA2DS2-VASc score was 3.1, and CHADS2 score was 1.9. The mean prevalence of common comorbidities was hypertension 68%, diabetes mellitus 24%, coronary artery disease 25%, chronic heart failure 20%, stroke history 13%, and bleeding history 12% (Table II in the Data Supplement).

    Study Quality

    All 48 included studies were rated as overall good quality (Table III in the Data Supplement). The majority of studies (88%) included a fully representative AF population, and all included studies used adherence/persistence data from secure record databases, resulting in the highest level of quality for that risk of bias category. Follow-up was complete and long enough to observe for adherence/persistence patterns for all studies.

    Adherence: Mean PDC/MPR

    The overall pooled mean PDC/MPR from the random-effects meta-analysis model for all follow-up durations from the 14 studies that reported mean PDC/MPR (47 data sets) was 77% (95% CI, 75%–80%; P=0.001, I2=100%; Figure 2). Subgroup analyses by each DOAC agent showed an overall pooled mean PDC/MPR for apixaban of 81% (95% CI, 76%–87%), for dabigatran of 72% (95% CI, 69%–75%), for rivaroxaban of 79% (95% CI, 75%–84%), and for DOACs as a class of 81% (95% CI, 72%–91%), which differed between groups (P=0.003). In subgroup analysis by region, the overall pooled mean PDC/MPR for North American data sources was significantly lower than for Europe (76% [95% CI, 74%–77%] versus 93% [95% CI, 87%–99%], P<0.0001). There were no studies in the Asia/Oceania region (Figure I in the Data Supplement). The duration of observation period ranged from 3 months to 24 months. The 12-month overall pooled mean PDC/MPR was 80% (95% CI, 72%–86%; Table IV in the Data Supplement). Analysis limited to the primary follow-up duration as identified by the study authors found similar results (77% [95% CI, 72%–82%]).

    Figure 2.

    Figure 2. Forest plot; pooled mean proportion of days covered/medication possession ratio.28,31,32,38,46,51,56,59–61,65 DOAC indicates direct oral anticoagulant.

    Adherence: Proportion of Good Adherence (PDC/MPR ≥80%)

    The overall pooled proportion of good adherence for all follow-up durations from the 21 studies that reported PDC/MPR ≥80% (69 data sets) was 66% (95% CI, 63%–70%; Figure 3). Subgroup analyses by each DOAC agent showed an overall pooled proportion of good adherence for apixaban of 71% (95% CI, 64%–78%), for dabigatran of 60% (95% CI, 52%–68%), for rivaroxaban of 70% (95% CI, 64%–75%), and DOACs as a class of 67% (95% CI, 61%–72%), which did not differ between groups (P=0.17). The overall pooled proportion of good adherence from North America was significantly lower than in Europe (63% [95% CI, 60%–66%] versus 74% [95% CI, 66%–82%], P=0.01). There was only one study in Asia/Oceania, therefore, we were unable to create a pooled estimate (Figure I in the Data Supplement). The 12-month overall pooled proportion of good adherence was 68% (95% CI, 62%–74%; Table IV in the Data Supplement). Analysis limited to the primary follow-up duration as identified by the study authors found similar results (67% [95% CI, 62%–72%]). There was only one study that reported a comparison of PDC/MPR ≥80% between DOACs and VKA, therefore, we were unable to create a pooled estimate.

    Figure 3.

    Figure 3. Forest plot; pooled proportion of good adherence (proportion of days covered/medication possession ratio ≥80%).10,24,25,27,28,31,32,38–41,45,49,51,52,56,58–61,65 DOAC indicates direct oral anticoagulant.


    The overall pooled proportion of persistence for all follow-up durations from the 36 studies that reported persistence (147 data sets) was 69% (95% CI, 65%–72%; Figure 4). Subgroup analyses by each DOAC agent showed an overall pooled proportion of persistence for apixaban of 74% (95% CI, 69%–80%), for dabigatran of 62% (95% CI, 57%–67%), for rivaroxaban of 72% (95% CI, 68%–77%), and DOACs as a class of 69% (95% CI, 60%–77%), which differed between groups (P=0.006). The overall pooled proportion of persistence from North America was significantly lower than in Europe (65% [95% CI, 59%–70%] versus 74% [95% CI, 71%–76%]; P=0.004). Asia/Oceania had the lowest persistence rate of 49% (95% CI, 36%–62%). However, there were a limited number of studies (3 studies/9 data sets) in Asia/Oceania which may not represent the entire region. In contrast, 17 studies/63 data sets in North America, and 14 studies/73 data sets in Europe were reported (Figure I in the Data Supplement). The 12-month overall pooled proportion for persistence was 62% (95% CI, 56%–68%; Table IV in the Data Supplement). Analysis limited to the primary follow-up duration as identified by the study authors found similar results (63% [95% CI, 58%–69%]). The pooled proportion for 1-year persistence for studies that used a permissible gap of 14 to 15 days was 41% (95% CI, 33%–49%), and for the 56 to 60-day gap definition, it was 60% (95% CI, 53%–67%; Table V in the Data Supplement).

    Figure 4.

    Figure 4. Forest plot; pooled proportion of persistence.9,21–23,25–27,29,30,32–38,40–44,46–50,53–55,57,60,62–66 DOAC indicates direct oral anticoagulant.

    The pooled proportion showed that persistence with any DOAC was higher than for VKA (odds ratio [OR], 1.44 [95% CI, 1.12–1.86]; P=0.0046; Figure II in the Data Supplement). Both apixaban and rivaroxaban had significantly higher persistence than VKA (OR, 1.90 [95% CI, 1.22–2.97], P=0.0048; OR, 1.52 [95% CI, 1.10–2.11], P=0.011, respectively), while dabigatran did not (OR, 1.22 [95% CI, 0.93–1.61], P=0.15; Figure III in the Data Supplement).


    Random effect meta-regression was performed to explore a potential association between study or patient characteristics and DOACs adherence/persistence. Longer observation time period had lower mean PDC/MPR (β coefficient, −0.9705; P=0.001) and persistence (β coefficient, −0.0065; P=0.007), while PDC/MPR ≥80% was not significantly lower (β coefficient, −0.0076; P=0.07). In addition, anticoagulant naïve status was associated with lower PDC/MPR ≥80% (β coefficient, −0.0048; P<0.0001; Table VI in the Data Supplement).

    Clinical Outcomes Associated With Nonadherence/Nonpersistence

    Among the 5 studies that reported clinical outcomes of DOAC nonadherence/nonpersistence compared with adherence/persistence, 3 studies reported stroke, 3 reported composite outcomes of all-cause mortality and stroke±transient ischemic attack, 1 reported deep vein thrombosis and pulmonary embolism, 1 reported myocardial infarction (MI), and 2 measured bleeding events (Table VII in the Data Supplement). Of these 5 studies, 3 studies reported multiple clinical outcomes. The pooled HR demonstrated that DOAC nonadherence was associated with an increased risk of stroke (HR 1.39, [95% CI, 1.06–1.81], P=0.016]), and DOAC nonpersistence was associated with an increased risk of stroke/transient ischemic attack (HR, 4.55 [95% CI, 2.80–7.39], P<0.0001]); however, bleeding events were not significantly different between nonadherent and adherent patients (HR, 1.04 [95% CI, 0.97–1.11], P=0.29]; Figure IV in the Data Supplement). The definitions of composite outcomes differed between studies; therefore, we were unable to create a pooled estimate. Those clinical outcomes not meta-analyzed also showed that DOAC nonadherence/nonpersistence was associated with increased risk of all-cause mortality/stroke (2 studies, overlapping data source), all-cause mortality/stroke/transient ischemic attack (1 study), and deep vein thrombosis and pulmonary embolism (1 study); while there was no significant difference in MI between nonadherent and adherent DOAC patients.

    Publication Bias

    The funnel plots for mean DOAC PDC/MPR, PDC/MPR ≥80%, persistence, and DOAC versus VKA persistence were all asymmetrical on visual inspection (Figure V in the Data Supplement). However, the P values for the Egger test were 0.36, <0.0001, <0.0001, and <0.0001 for the mean DOAC PDC/MPR, PDC/MPR ≥80%, persistence, and DOAC versus VKA persistence, respectively. The trim and fill method was used for sensitivity analyses, resulting in pooled mean PDC/MPR, PDC/MPR >80%, persistence, and DOAC versus VKA persistence, summary estimates of 80% (95% CI, 77%–82%), 58% (95% CI, 54%–62%), 60% (95% CI, 56%–63%), and OR, 2.07 (95% CI, 1.60–2.69), respectively (Figure VI in the Data Supplement).


    Our systematic review and meta-analysis of 48 real-world evidence studies, involving more than 570 000 patients with AF, showed suboptimal adherence to DOACs, with patients not taking their DOAC 1 out of every 4 days, and one-third of patients showing poor adherence. The patient population of the included studies in our meta-analysis is similar to the general AF population and the landmark DOAC RCTs, making our results highly generalizable4–6 (Table VIII in the Data Supplement). Previous studies have similarly found that only 60% of patients have good persistence with warfarin.67 Although our post hoc analysis comparing DOACs and VKA showed that the persistence rate of DOACs was higher than VKA, suboptimal persistence of 69% with DOAC therapy remains a great concern for patients with AF.

    The 2019 American College of Cardiology/American Heart Association/Heart Rhythm Society Focused Update of the 2014 Guideline for the Management of Patients with Atrial Fibrillation recommends DOACs over warfarin for stroke prevention in AF, based on RCTs showing noninferiority or superiority of DOACs in preventing stroke and lowering the risk of serious bleeding.3 However, our results illustrate poor DOAC adherence in a real-world AF population, and, as such, the benefits shown in RCTs with higher levels of persistence may not be realized in clinical practice. Our study demonstrated that suboptimal DOAC adherence was associated with poor clinical outcomes, with 39% higher hazard of stroke and increased risk of thromboembolic events and all-cause mortality in nonadherent patients. This negative consequence could be due to shorter half-lives of DOACs than warfarin, resulting in more rapid elimination of anticoagulation effect in nonadherent DOAC patients.8 Therefore, poor DOAC adherence may result in potentially greater harm than nonadherence with warfarin. Although one of the advantages of DOACs over warfarin is their ease of use, the lack of monitoring may become a disadvantage that predisposes patients to DOAC nonadherence, requiring a greater effort in patient education and monitoring for adherence. Research has shown that DOAC management by pharmacists or anticoagulation clinics, and extended monitoring for 1-year after DOAC initiation was associated with higher DOAC adherence.68 Although these interventions have not been evaluated prospectively, in cases where adherence must be ensured, employing these strategies may be of benefit. Otherwise, treating with warfarin instead of a DOAC may be considered, given the ability to monitor international normalized ratio with warfarin and potential for higher time in therapeutic range for patients managed at anticoagulation clinics.69 Periodic laboratory monitoring as would be done for lipid levels or hemoglobin A1c may be another strategy in managing DOAC nonadherence. However, qualitative or quantitative measurements of blood concentration/assays for DOACs are not widely available in clinical settings or approved by US Food and Drug Administration to date. Overall, more research is needed to evaluate which strategies may effectively improve anticoagulant, and in particular, DOAC adherence/persistence.

    Among the 3 methods typically used to measure medication adherence, mean PDC/MPR, PDC/MPR ≥80%, and persistence, persistence was the most common outcome measure in the included studies. Compared with the 2-year persistence rates of 75% to 79% reported in the landmark DOAC RCTs for apixaban, dabigatran, and rivaroxaban, the pooled persistence in our study using real-world observational data at 1-year was even lower at 69%, with only 66% of patients having good adherence.4–6 As close monitoring is often required in clinical trials, the lower real-world persistence rates are not unexpected but raise concerns about the expected level of DOAC effectiveness in clinical practice.

    Among DOAC agents, dabigatran had the lowest adherence and persistence at all time periods, ≈8 to 10 percentage points lower than with apixaban and rivaroxaban, which were similar to each other. Several reasons may contribute to low dabigatran rates, including gastrointestinal adverse effects, special storage requirements unique to dabigatran, and twice-daily dosing.70,71 Of these, the dosing regimen may be a minor contributor since we found that twice-daily apixaban had similar adherence rates as once-daily rivaroxaban. These findings differ from general adherence research that find once-daily regimens have higher adherence than twice-daily regimens.72,73 Other contributing factors for higher apixaban adherence may include stronger safety and efficacy evidence for apixaban (including in those with chronic kidney disease), its cost-effectiveness, and higher amounts of direct-to-consumer advertising expenditure for apixaban than for other agents.74–77

    Our study has some limitations. First, there is heterogeneity of the included studies, possibly due to variations in definitions of adherence and persistence, as well as follow-up durations. We conducted subgroup analyses to pool the same definitions and time periods, however, residual heterogeneity persisted. Second, some studies reported adherence and persistence in graphical form, rather than specific values, which did not allow for data extraction. Third, there is no universal tool available to assess the risk of bias for systematic reviews of observational studies. We used the Newcastle-Ottawa Scale, a commonly used tool, however, due to the similar methodology for our included studies, this tool did not differentiate between study quality and rated all included studies at the level of good quality. Finally, the included studies primarily used prescription claims data, which records prescriptions dispensed, not necessarily taken. Therefore, it is possible that adherence and persistence rates are even lower than the summary estimates we report.


    Suboptimal adherence and persistence to DOACs was common in patients with AF. Although DOAC persistence rates were higher than VKA, our study found that suboptimal adherence to DOACs was associated with an increased risk of stroke, which is of great concern. Although DOACs do not require laboratory monitoring, a greater effort in patient education and monitoring for adherence and persistence may be necessary. Clinicians must be aware of potential DOAC nonadherence and nonpersistence to ensure that anticoagulation therapy for stroke prevention is optimized.


    We would like to express our gratitude to Christopher A. Lim, PharmD candidate, and Belinda Luck, PharmD candidate, at Western University of Health Sciences for collecting data for the post hoc analysis.


    Guest Editor for this article was Jonathan Piccini, MD, MHS.

    The Data Supplement is available at

    Cynthia A. Jackevicius, BScPhm, PharmD, MSc, BCPS-AQ Cardiology, BCCP, FCSHP, FAHA, FCCP, FCCS, FACC, Western University of Health Sciences, 309 E Second St, Pomona, CA 91766. Email


    • 1. Chugh SS, Havmoeller R, Narayanan K, Singh D, Rienstra M, Benjamin EJ, Gillum RF, Kim YH, McAnulty JH, Zheng ZJ, Forouzanfar MH, Naghavi M, Mensah GA, Ezzati M, Murray CJ. Worldwide epidemiology of atrial fibrillation: a Global Burden of Disease 2010 Study.Circulation. 2014; 129:837–847. doi: 10.1161/CIRCULATIONAHA.113.005119LinkGoogle Scholar
    • 2. Colilla S, Crow A, Petkun W, Singer DE, Simon T, Liu X. Estimates of current and future incidence and prevalence of atrial fibrillation in the U.S. adult population.Am J Cardiol. 2013; 112:1142–1147. doi: 10.1016/j.amjcard.2013.05.063CrossrefMedlineGoogle Scholar
    • 3. January CT, Wann LS, Calkins H, Chen LY, Cigarroa JE, Cleveland JC, Ellinor PT, Ezekowitz MD, Field ME, Furie KL, Heidenreich PA, Murray KT, Shea JB, Tracy CM, Yancy CW. 2019 AHA/ACC/HRS focused update of the 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association task force on clinical practice guidelines and the Heart Rhythm Society.J Am Coll Cardiol. 2019; 74:104–132. doi: 10.1016/j.jacc.2019.01.011CrossrefMedlineGoogle Scholar
    • 4. Granger CB, Alexander JH, McMurray JJ, Lopes RD, Hylek EM, Hanna M, Al-Khalidi HR, Ansell J, Atar D, Avezum A, Bahit MC, Diaz R, Easton JD, Ezekowitz JA, Flaker G, Garcia D, Geraldes M, Gersh BJ, Golitsyn S, Goto S, Hermosillo AG, Hohnloser SH, Horowitz J, Mohan P, Jansky P, Lewis BS, Lopez-Sendon JL, Pais P, Parkhomenko A, Verheugt FW, Zhu J, Wallentin L; ARISTOTLE Committees and Investigators. Apixaban versus warfarin in patients with atrial fibrillation.N Engl J Med. 2011; 365:981–992. doi: 10.1056/NEJMoa1107039CrossrefMedlineGoogle Scholar
    • 5. Connolly SJ, Ezekowitz MD, Yusuf S, Eikelboom J, Oldgren J, Parekh A, Pogue J, Reilly PA, Themeles E, Varrone J, Wang S, Alings M, Xavier D, Zhu J, Diaz R, Lewis BS, Darius H, Diener HC, Joyner CD, Wallentin L; RE-LY Steering Committee and Investigators. Dabigatran versus warfarin in patients with atrial fibrillation.N Engl J Med. 2009; 361:1139–1151. doi: 10.1056/NEJMoa0905561CrossrefMedlineGoogle Scholar
    • 6. Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, Breithardt G, Halperin JL, Hankey GJ, Piccini JP, Becker RC, Nessel CC, Paolini JF, Berkowitz SD, Fox KA, Califf RM; ROCKET AF Investigators. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation.N Engl J Med. 2011; 365:883–891. doi: 10.1056/NEJMoa1009638CrossrefMedlineGoogle Scholar
    • 7. Giugliano RP, Ruff CT, Braunwald E, Murphy SA, Wiviott SD, Halperin JL, Waldo AL, Ezekowitz MD, Weitz JI, Špinar J, Ruzyllo W, Ruda M, Koretsune Y, Betcher J, Shi M, Grip LT, Patel SP, Patel I, Hanyok JJ, Mercuri M, Antman EM; ENGAGE AF-TIMI 48 Investigators. Edoxaban versus warfarin in patients with atrial fibrillation.N Engl J Med. 2013; 369:2093–2104. doi: 10.1056/NEJMoa1310907CrossrefMedlineGoogle Scholar
    • 8. Adcock DM, Gosselin R. Direct Oral Anticoagulants (DOACs) in the laboratory: 2015 review.Thromb Res. 2015; 136:7–12. doi: 10.1016/j.thromres.2015.05.001CrossrefMedlineGoogle Scholar
    • 9. Jackevicius CA, Tsadok MA, Essebag V, Atzema C, Eisenberg MJ, Tu JV, Lu L, Rahme E, Ho PM, Turakhia M, Humphries KH, Behlouli H, Zhou L, Pilote L. Early non-persistence with dabigatran and rivaroxaban in patients with atrial fibrillation.Heart. 2017; 103:1331–1338. doi: 10.1136/heartjnl-2016-310672CrossrefMedlineGoogle Scholar
    • 10. Deshpande CG, Kogut S, Willey C. Real-world health care costs based on medication adherence and risk of stroke and bleeding in patients treated with novel anticoagulant therapy.J Manag Care Spec Pharm. 2018; 24:430–439. doi: 10.18553/jmcp.2018.24.5.430CrossrefMedlineGoogle Scholar
    • 11. Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D, Moher D, Becker BJ, Sipe TA, Thacker SB. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group.JAMA. 2000; 283:2008–2012. doi: 10.1001/jama.283.15.2008CrossrefMedlineGoogle Scholar
    • 12. Morisky DE, Green LW, Levine DM. Concurrent and predictive validity of a self-reported measure of medication adherence.Med Care. 1986; 24:67–74. doi: 10.1097/00005650-198601000-00007CrossrefMedlineGoogle Scholar
    • 13. Morisky DE, Ang A, Krousel-Wood M, Ward HJ. Predictive validity of a medication adherence measure in an outpatient setting.J Clin Hypertens (Greenwich). 2008; 10:348–354. doi: 10.1111/j.1751-7176.2008.07572.xCrossrefMedlineGoogle Scholar
    • 14. Wells GA, Shea B, O’Connell D, Peterson J, Welch V, Losos M, Tugwell P. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomized studies in meta-analyses. Accessed September 1, 2018.Google Scholar
    • 15. DerSimonian R, Laird N. Meta-analysis in clinical trials.Control Clin Trials. 1986; 7:177–188. doi: 10.1016/0197-2456(86)90046-2CrossrefMedlineGoogle Scholar
    • 16. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses.BMJ. 2003; 327:557–560. doi: 10.1136/bmj.327.7414.557CrossrefMedlineGoogle Scholar
    • 17. Thompson SG, Higgins JP. How should meta-regression analyses be undertaken and interpreted?Stat Med. 2002; 21:1559–1573. doi: 10.1002/sim.1187CrossrefMedlineGoogle Scholar
    • 18. Sterne JA, Egger M, Smith GD. Systematic reviews in health care: investigating and dealing with publication and other biases in meta-analysis.BMJ. 2001; 323:101–105. doi: 10.1136/bmj.323.7304.101CrossrefMedlineGoogle Scholar
    • 19. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test.BMJ. 1997; 315:629–634. doi: 10.1136/bmj.315.7109.629CrossrefMedlineGoogle Scholar
    • 20. Duval S, Tweedie R. Trim and fill: A simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis.Biometrics. 2000; 56:455–463. doi: 10.1111/j.0006-341x.2000.00455.xCrossrefMedlineGoogle Scholar
    • 21. Annavarapu S, Gandhi PK, Li Y, Arora P, Moretz C, Wang C, Sander SD, Andrews GA. Factors influencing dabigatran or warfarin medication persistence in patients with nonvalvular atrial fibrillation.J Comp Eff Res. 2018; 7:685–691. doi: 10.2217/cer-2017-0081CrossrefMedlineGoogle Scholar
    • 22. Cataldo N, Pegoraro V, Ripellino C, Bertasi V, Di Turi R, Pastorello M, Gambera M. Non-persistence risk and health care resource utilization of Italian patients with non-valvular atrial fibrillation.Recenti Prog Med. 2018; 109:113–121. doi: 10.1701/2865.28904CrossrefMedlineGoogle Scholar
    • 23. Collings SL, Vannier-Moreau V, Johnson ME, Stynes G, Lefèvre C, Maguire A, Asmar J, Bizouard G, Duhot D, Mouquet F, Fauchier L. Initiation and continuation of oral anticoagulant prescriptions for stroke prevention in non-valvular atrial fibrillation: a cohort study in primary care in France.Arch Cardiovasc Dis. 2018; 111:370–379. doi: 10.1016/j.acvd.2017.10.003CrossrefMedlineGoogle Scholar
    • 24. Deshpande CG, Kogut S, Laforge R, Willey C. Impact of medication adherence on risk of ischemic stroke, major bleeding and deep vein thrombosis in atrial fibrillation patients using novel oral anticoagulants.Curr Med Res Opin. 2018; 34:1285–1292. doi: 10.1080/03007995.2018.1428543CrossrefMedlineGoogle Scholar
    • 25. Gomez-Lumbreras A, Cortes J, Giner-Soriano M, Quijada-Manuitt MA, Morros R. Characteristics of apixaban-treated patients, evaluation of the dose prescribed, and the persistence of treatment: a Cohort Study in Catalonia.J Cardiovasc Pharmacol Ther. 2018; 23:494–501. doi: 10.1177/1074248418778544CrossrefMedlineGoogle Scholar
    • 26. Haastrup SB, Hellfritzsch M, Rasmussen L, Pottegård A, Grove EL. Use of non-vitamin K antagonist oral anticoagulants 2008-2016: a Danish Nationwide Cohort Study.Basic Clin Pharmacol Toxicol. 2018; 123:452–463. doi: 10.1111/bcpt.13024CrossrefMedlineGoogle Scholar
    • 27. Harper P, Pollock D, Stephens M. Dabigatran persistence and adherence in New Zealand: a nationwide retrospective observational study.BMJ Open. 2018; 8:e020212. doi: 10.1136/bmjopen-2017-020212CrossrefMedlineGoogle Scholar
    • 28. Jacobs MS, Schouten JF, de Boer PT, Hoffmann M, Levin LÅ, Postma MJ. Secondary adherence to non-vitamin-K antagonist oral anticoagulants in patients with atrial fibrillation in Sweden and the Netherlands.Curr Med Res Opin. 2018; 34:1839–1847. doi: 10.1080/03007995.2018.1459528CrossrefMedlineGoogle Scholar
    • 29. Lip GYH, Pan X, Kamble S, Kawabata H, Mardekian J, Masseria C, Phatak H. Discontinuation risk comparison among ‘real-world’ newly anticoagulated atrial fibrillation patients: apixaban, warfarin, dabigatran, or rivaroxaban.PLoS One. 2018; 13:e0195950. doi: 10.1371/journal.pone.0195950CrossrefMedlineGoogle Scholar
    • 30. Maura G, Billionnet C, Alla F, Gagne JJ, Pariente A. Comparison of treatment persistence with dabigatran or rivaroxaban versus vitamin K antagonist oral anticoagulants in atrial fibrillation patients: a competing risk analysis in the French national health care databases.Pharmacotherapy. 2018; 38:6–18. doi: 10.1002/phar.2046CrossrefMedlineGoogle Scholar
    • 31. Borne RT, O’Donnell C, Turakhia MP, Varosy PD, Jackevicius CA, Marzec LN, Masoudi FA, Hess PL, Maddox TM, Ho PM. Adherence and outcomes to direct oral anticoagulants among patients with atrial fibrillation: findings from the veterans health administration.BMC Cardiovasc Disord. 2017; 17:236. doi: 10.1186/s12872-017-0671-6CrossrefMedlineGoogle Scholar
    • 32. Brown JD, Shewale AR, Talbert JC. Adherence to rivaroxaban, dabigatran, and apixaban for stroke prevention for newly diagnosed and treatment-naïve atrial fibrillation patients: an update using 2013-2014 data.J Manag Care Spec Pharm. 2017; 23:958–67. doi: 10.18553/jmcp.2017.23.9.958CrossrefMedlineGoogle Scholar
    • 33. Collings SL, Lefèvre C, Johnson ME, Evans D, Hack G, Stynes G, Maguire A. Oral anticoagulant persistence in patients with non-valvular atrial fibrillation: a cohort study using primary care data in Germany.PLoS One. 2017; 12:e0185642. doi: 10.1371/journal.pone.0185642CrossrefMedlineGoogle Scholar
    • 34. Douros A, Renoux C, Coulombe J, Suissa S. Patterns of long-term use of non-vitamin K antagonist oral anticoagulants for non-valvular atrial fibrillation: Quebec observational study.Pharmacoepidemiol Drug Saf. 2017; 26:1546–1554. doi: 10.1002/pds.4333CrossrefMedlineGoogle Scholar
    • 35. Hellfritzsch M, Husted SE, Grove EL, Rasmussen L, Poulsen BK, Johnsen SP, Hallas J, Pottegård A. Treatment changes among users of non-vitamin K antagonist oral anticoagulants in atrial fibrillation.Basic Clin Pharmacol Toxicol. 2017; 120:187–194. doi: 10.1111/bcpt.12664CrossrefMedlineGoogle Scholar
    • 36. Hernandez I, Zhang Y, Saba S. Comparison of the effectiveness and safety of apixaban, dabigatran, rivaroxaban, and warfarin in newly diagnosed atrial fibrillation.Am J Cardiol. 2017; 120:1813–1819. doi: 10.1016/j.amjcard.2017.07.092CrossrefMedlineGoogle Scholar
    • 37. Lamberts M, Staerk L, Olesen JB, Fosbol EL, Hansen ML, Harboe L, Lefevre C, Evans D, Gislason GH. Major bleeding complications and persistence with oral anticoagulation in non-valvular atrial fibrillation: contemporary findings in real-life Danish patients.J Am Heart Assoc. 2017; 6:e004517. doi: 10.1161/JAHA.116.004517.LinkGoogle Scholar
    • 38. Manzoor BS, Lee TA, Sharp LK, Walton SM, Galanter WL, Nutescu EA. Real-world adherence and persistence with direct oral anticoagulants in adults with atrial fibrillation.Pharmacotherapy. 2017; 37:1221–1230. doi: 10.1002/phar.1989CrossrefMedlineGoogle Scholar
    • 39. Maura G, Pariente A, Alla F, Billionnet C. Adherence with direct oral anticoagulants in nonvalvular atrial fibrillation new users and associated factors: a French nationwide cohort study.Pharmacoepidemiol Drug Saf. 2017; 26:1367–1377. doi: 10.1002/pds.4268CrossrefMedlineGoogle Scholar
    • 40. McHorney CA, Ashton V, Laliberté F, Germain G, Wynant W, Crivera C, Schein JR, Lefebvre P, Peterson ED. Adherence to rivaroxaban compared with other oral anticoagulant agents among patients with nonvalvular atrial fibrillation.J Manag Care Spec Pharm. 2017; 23:980–988. doi: 10.18553/jmcp.2017.23.9.980CrossrefMedlineGoogle Scholar
    • 41. Mueller T, Alvarez-Madrazo S, Robertson C, Bennie M. Use of direct oral anticoagulants in patients with atrial fibrillation in Scotland: applying a coherent framework to drug utilisation studies.Pharmacoepidemiol Drug Saf. 2017; 26:1378–1386. doi: 10.1002/pds.4272CrossrefMedlineGoogle Scholar
    • 42. Paquette M, Riou França L, Teutsch C, Diener HC, Lu S, Dubner SJ, Ma CS, Rothman KJ, Zint K, Halperin JL, Huisman MV, Lip GYH, Nieuwlaat R. Persistence with dabigatran therapy at 2 years in patients with atrial fibrillation.J Am Coll Cardiol. 2017; 70:1573–1583. doi: 10.1016/j.jacc.2017.07.793CrossrefMedlineGoogle Scholar
    • 43. Song X, Gandhi P, Gilligan AM, Arora P, Wang C, Henriques C, Sander S, Smith DM. Comparison of all-cause, stroke, and bleed-specific healthcare resource utilization among patients with non-valvular atrial fibrillation (NVAF) and newly treated with dabigatran or warfarin.Expert Rev Pharmacoecon Outcomes Res. 2017; 3:213-222. doi: 10.1080/14737167.2017.1347041Google Scholar
    • 44. Sørensen R, Jamie Nielsen B, Langtved Pallisgaard J, Ji-Young Lee C, Torp-Pedersen C. Adherence with oral anticoagulation in non-valvular atrial fibrillation: a comparison of vitamin K antagonists and non-vitamin K antagonists.Eur Heart J Cardiovasc Pharmacother. 2017; 3:151–156. doi: 10.1093/ehjcvp/pvw048CrossrefMedlineGoogle Scholar
    • 45. Alberts MJ, Peacock WF, Fields LE, Bunz TJ, Nguyen E, Milentijevic D, Schein JR, Coleman CI. Association between once- and twice-daily direct oral anticoagulant adherence in nonvalvular atrial fibrillation patients and rates of ischemic stroke.Int J Cardiol. 2016; 215:11–13. doi: 10.1016/j.ijcard.2016.03.212CrossrefMedlineGoogle Scholar
    • 46. Al-Khalili F, Lindström C, Benson L. Adherence to anticoagulant treatment with apixaban and rivaroxaban in a real-world setting.Clin Trials and Regul Sci Cardiol. 2016; 18:1–4.CrossrefGoogle Scholar
    • 47. Bancroft T, Lim J, Wang C, Sander SD, Swindle JP. Health care resource utilization, costs, and persistence in patients newly diagnosed as having nonvalvular atrial fibrillation and newly treated with dabigatran versus warfarin in the United States.Clin Ther. 2016; 38:545–56.e1. doi: 10.1016/j.clinthera.2016.01.008CrossrefMedlineGoogle Scholar
    • 48. Beshir SA, Chee KH, Lo YL. Factors associated with abrupt discontinuation of dabigatran therapy in patients with atrial fibrillation in Malaysia.Int J Clin Pharm. 2016; 38:1182–1190. doi: 10.1007/s11096-016-0350-1CrossrefMedlineGoogle Scholar
    • 49. Beyer-Westendorf J, Ehlken B, Evers T. Real-world persistence and adherence to oral anticoagulation for stroke risk reduction in patients with atrial fibrillation.Europace. 2016; 18:1150–1157. doi: 10.1093/europace/euv421CrossrefMedlineGoogle Scholar
    • 50. Coleman CI, Tangirala M, Evers T. Treatment persistence and discontinuation with rivaroxaban, dabigatran, and warfarin for stroke prevention in patients with non-valvular atrial fibrillation in the United States.PLoS One. 2016; 11:e0157769. doi: 10.1371/journal.pone.0157769CrossrefMedlineGoogle Scholar
    • 51. Coleman CI, Tangirala M, Evers T. Medication adherence to rivaroxaban and dabigatran for stroke prevention in patients with non-valvular atrial fibrillation in the United States.Int J Cardiol. 2016; 212:171–173. doi: 10.1016/j.ijcard.2016.03.051CrossrefMedlineGoogle Scholar
    • 52. Forslund T, Wettermark B, Hjemdahl P. Comparison of treatment persistence with different oral anticoagulants in patients with atrial fibrillation.Eur J Clin Pharmacol. 2016; 72:329–338. doi: 10.1007/s00228-015-1983-zCrossrefMedlineGoogle Scholar
    • 53. Johnson ME, Lefèvre C, Collings SL, Evans D, Kloss S, Ridha E, Maguire A. Early real-world evidence of persistence on oral anticoagulants for stroke prevention in non-valvular atrial fibrillation: a cohort study in UK primary care.BMJ Open. 2016; 6:e011471. doi: 10.1136/bmjopen-2016-011471CrossrefMedlineGoogle Scholar
    • 54. Kachroo S, Hamilton M, Liu X, Pan X, Brixner D, Marrouche N, Biskupiak J. Oral anticoagulant discontinuation in patients with nonvalvular atrial fibrillation.Am J Manag Care. 2016; 22:e1–e8.MedlineGoogle Scholar
    • 55. Martinez C, Katholing A, Wallenhorst C, Freedman SB. Therapy persistence in newly diagnosed non-valvular atrial fibrillation treated with warfarin or NOAC. A cohort study.Thromb Haemost. 2016; 115:31–39. doi: 10.1160/TH15-04-0350CrossrefMedlineGoogle Scholar
    • 56. McHorney CA, Peterson ED, Laliberté F, Germain G, Nelson WW, Crivera C, Schein J, Lefebvre P. Comparison of adherence to rivaroxaban versus apixaban among patients with atrial fibrillation.Clin Ther. 2016; 38:2477–2488. doi: 10.1016/j.clinthera.2016.09.014CrossrefMedlineGoogle Scholar
    • 57. Simons LA, Ortiz M, Freedman SB, Waterhouse BJ, Colquhoun D, Thomas G. Improved persistence with non-vitamin-K oral anticoagulants compared with warfarin in patients with atrial fibrillation: recent Australian experience.Curr Med Res Opin. 2016; 32:1857–1861. doi: 10.1080/03007995.2016.1218325CrossrefMedlineGoogle Scholar
    • 58. Yao X, Abraham NS, Alexander GC, Crown W, Montori VM, Sangaralingham LR, Gersh BJ, Shah ND, Noseworthy PA. Effect of adherence to oral anticoagulants on risk of stroke and major bleeding among patients with atrial fibrillation.J Am Heart Assoc. 2016; 5:e003074. doi: 10.1161/JAHA.115.003074LinkGoogle Scholar
    • 59. Gorst-Rasmussen A, Skjøth F, Larsen TB, Rasmussen LH, Lip GY, Lane DA. Dabigatran adherence in atrial fibrillation patients during the first year after diagnosis: a nationwide cohort study.J Thromb Haemost. 2015; 13:495–504. doi: 10.1111/jth.12845CrossrefMedlineGoogle Scholar
    • 60. Zhou M, Chang HY, Segal JB, Alexander GC, Singh S. Adherence to a novel oral anticoagulant among patients with atrial fibrillation.J Manag Care Spec Pharm. 2015; 21:1054–1062. doi: 10.18553/jmcp.2015.21.11.1054CrossrefMedlineGoogle Scholar
    • 61. Cutler TW, Chuang A, Huynh TD, Witt RG, Branch J, Pon T, White R. A retrospective descriptive analysis of patient adherence to dabigatran at a large academic medical center.J Manag Care Spec Pharm. 2014; 20:1028–1034. doi: 10.18553/jmcp.2014.20.10.1028CrossrefMedlineGoogle Scholar
    • 62. Ho MH, Ho CW, Cheung E, Chan PH, Hai JJ, Chan KH, Chan EW, Leung GK, Tse HF, Siu CW. Continuation of dabigatran therapy in “real-world” practice in Hong Kong.PLoS One. 2014; 9:e101245. doi: 10.1371/journal.pone.0101245CrossrefMedlineGoogle Scholar
    • 63. Laliberté F, Cloutier M, Nelson WW, Coleman CI, Pilon D, Olson WH, Damaraju CV, Schein JR, Lefebvre P. Real-world comparative effectiveness and safety of rivaroxaban and warfarin in nonvalvular atrial fibrillation patients.Curr Med Res Opin. 2014; 30:1317–1325. doi: 10.1185/03007995.2014.907140CrossrefMedlineGoogle Scholar
    • 64. Nelson WW, Song X, Coleman CI, Thomson E, Smith DM, Damaraju CV, Schein JR. Medication persistence and discontinuation of rivaroxaban versus warfarin among patients with non-valvular atrial fibrillation.Curr Med Res Opin. 2014; 30:2461–2469. doi: 10.1185/03007995.2014.933577CrossrefMedlineGoogle Scholar
    • 65. Shore S, Carey EP, Turakhia MP, Jackevicius CA, Cunningham F, Pilote L, Bradley SM, Maddox TM, Grunwald GK, Barón AE, Rumsfeld JS, Varosy PD, Schneider PM, Marzec LN, Ho PM. Adherence to dabigatran therapy and longitudinal patient outcomes: insights from the veterans health administration.Am Heart J. 2014; 167:810–817. doi: 10.1016/j.ahj.2014.03.023CrossrefMedlineGoogle Scholar
    • 66. Zalesak M, Siu K, Francis K, Yu C, Alvrtsyan H, Rao Y, Walker D, Sander S, Miyasato G, Matchar D, Sanchez H. Higher persistence in newly diagnosed nonvalvular atrial fibrillation patients treated with dabigatran versus warfarin.Circ Cardiovasc Qual Outcomes. 2013; 6:567–574. doi: 10.1161/CIRCOUTCOMES.113.000192LinkGoogle Scholar
    • 67. Kimmel SE, Chen Z, Price M, Parker CS, Metlay JP, Christie JD, Brensinger CM, Newcomb CW, Samaha FF, Gross R. The influence of patient adherence on anticoagulation control with warfarin: results from the International Normalized Ratio Adherence and Genetics (IN-RANGE) Study.Arch Intern Med. 2007; 167:229–235. doi: 10.1001/archinte.167.3.229CrossrefMedlineGoogle Scholar
    • 68. Shore S, Ho PM, Lambert-Kerzner A, Glorioso TJ, Carey EP, Cunningham F, Longo L, Jackevicius C, Rose A, Turakhia MP. Site-level variation in and practices associated with dabigatran adherence.JAMA. 2015; 313:1443–1450. doi: 10.1001/jama.2015.2761CrossrefMedlineGoogle Scholar
    • 69. Baker WL, Cios DA, Sander SD, Coleman CI. Meta-analysis to assess the quality of warfarin control in atrial fibrillation patients in the United States.J Manag Care Pharm. 2009; 15:244–252. doi: 10.18553/jmcp.2009.15.3.244CrossrefMedlineGoogle Scholar
    • 70. Pradaxa® product information. Ridgefield, CT: Boehringer Ingelheim Pharmaceuticals Inc; 2018.Google Scholar
    • 71. FDA drug safety communication: Special storage and handling requirements must be followed for Pradaxa (dabigaran etexilate mesylate) capsules. Accessed March 20, 2019.Google Scholar
    • 72. Claxton AJ, Cramer J, Pierce C. A systematic review of the associations between dose regimens and medication compliance.Clin Ther. 2001; 23:1296–1310. doi: 10.1016/s0149-2918(01)80109-0CrossrefMedlineGoogle Scholar
    • 73. Coleman CI, Roberts MS, Sobieraj DM, Lee S, Alam T, Kaur R. Effect of dosing frequency on chronic cardiovascular disease medication adherence.Curr Med Res Opin. 2012; 28:669–680. doi: 10.1185/03007995.2012.677419CrossrefMedlineGoogle Scholar
    • 74. Shah A, Shewale A, Hayes CJ, Martin BC. Cost-effectiveness of oral anticoagulants for ischemic stroke prophylaxis among nonvalvular atrial fibrillation patients.Stroke. 2016; 47:1555–1561. doi: 10.1161/STROKEAHA.115.012325LinkGoogle Scholar
    • 75. Siontis KC, Zhang X, Eckard A, Bhave N, Schaubel DE, He K, Tilea A, Stack AG, Balkrishnan R, Yao X, Noseworthy PA, Shah ND, Saran R, Nallamothu BK. Outcomes associated with apixaban use in patients with end-stage kidney disease and atrial fibrillation in the United States.Circulation. 2018; 138:1519–1529. doi: 10.1161/CIRCULATIONAHA.118.035418LinkGoogle Scholar
    • 76. Sanders GD, Lowenstern A, Borre E, Chatterjee R, Goode A, Sharan L, Allen LaPointe NM, Raitz G, Shah B, Yapa R, Davis JK, Lallinger K, Schmidt R, Kosinski A, Al-Khatib S. Stroke prevention in patients with atrial fibrillation: a systematic review update.In: Comparative Effectiveness Review No. 214. Rockville MD: Agency for Healthcare Research and Quality (US); 2018.Google Scholar
    • 77. Millman J. It’s true: Drug companies are bombarding your TV with more ads than ever. The Washington Post. 2015. Accessed March 20, 2019.Google Scholar