American Journal of Cardiovascular Drugs (2020) 20:61–71 https://doi.org/10.1007/s40256-019-00356-2 ORIGINAL RESEARCH ARTICLE Educational Impact on Apixaban Adherence in Atrial Fibrillation (the AEGEAN STUDY): A Randomized Clinical Trial Gilles Montalescot1 · Carlos Brotons2 · Bernard Cosyns3 · Harry J. Crijns4 · Armando D’Angelo5 · Ludovic Drouet6 · Franz Eberli7 · Deirdre A. Lane8 · Bruno Besse9 · Anthony Chan10 · Eric Vicaut11 · Harald Darius12 on behalf of the AEGEAN Study Investigators Published online: 26 June 2019 © The Author(s) 2019 Abstract Introduction Adherence to non-vitamin-K oral anticoagulants (NOACs) may be lower than to vitamin K antagonists because NOACs do not require routine monitoring. Objective We assessed the impact of an educational program on adherence and persistence with apixaban in patients with non-valvular atrial fibrillation (NVAF). Methods Patients with NVAF eligible for NOACs with one or more stroke risk factor (prior stroke/transient ischemic attack, age ≥ 75 years, hypertension, diabetes, or symptomatic heart failure) were randomized (1:1) to standard of care (SOC) or SOC with additional educational (information booklet, reminder tools, virtual clinic access). The primary outcome was adherence to apixaban (2.5 or 5 mg twice daily) at 24 weeks. Patients receiving the educational program were re-randomized (1:1) to continue the program for 24 further weeks or to switch to secondary SOC. Implementation adherence and persistence were reassessed at 48 weeks. Results In total, 1162 patients were randomized (SOC, 583; educational program, 579). Mean implementation adherence ± standard deviation (SD) at 24 weeks was 91.6% ± 17.1 for SOC and 91.9% ± 16.1 for the educational program arm; results did not differ significantly between groups at any time-point. At 48 weeks, implementation adherence was 90.4% ± 18.0, 90.1% ± 18.6, and 89.3% ± 18.1 for continued educational program, SOC, and secondary SOC, respectively; and corre- sponding persistence was 86.1% (95% confidence interval [CI] 81.3–89.7), 85.2% (95% CI 81.5–88.2), and 87.8% (95% CI 83.4–91.1). Serious adverse events were similar across groups. Conclusion High implementation adherence and persistence with apixaban were observed in patients with NVAF receiving apixaban. The educational program did not show additional benefits. Clinical trial registration This study is registered at ClinicalTrials.gov [NCT01884350]. 1 Introduction reducing the risk by about two-thirds [1]. However, VKAs have drawbacks limiting their use: extensive drug and food Vitamin K antagonists (VKAs) are highly effective for interactions, narrow therapeutic range, and need for frequent stroke prevention in non-valvular atrial fibrillation (NVAF), international normalized ratio (INR) monitoring and dose adjustments [1]. Accordingly, ongoing patient education is Bruno Besse: Position at the time of manuscript preparation. No necessary with VKA treatment. longer an employee of Bristol-Myers Squibb. In patients with NVAF, non-VKA oral anticoagulants (NOACs) are at least as effective as warfarin for preventing The full list of AEGEAN Investigators is given in the Electronic stroke or systemic embolism (SE), while reducing the risk of Supplementary Material. intracranial hemorrhage by half [2–6]. In particular, apixa- Electronic supplementary material The online version of this ban has shown superiority over warfarin in preventing stroke article (https​://doi.org/10.1007/s4025​6-019-00356​-2) contains or SE and is associated with a lower risk of major bleeding supplementary material, which is available to authorized users. complications and mortality [2]. NOACs have fewer food * Gilles Montalescot and drug interactions than VKAs and do not require routine gilles.montalescot@aphp.fr coagulation monitoring, overcoming many of the limitations Extended author information available on the last page of the article of VKAs [7]. Vol.:(0123456789) 62 G. Montalescot et al. completion of follow-up) and persistence, using an elec- Key Points tronic monitoring device (EMD) that recorded detailed information on the timing of medication access [20]. Patients with non-valvular atrial fibrillation who do not receive adequate anticoagulation are at increased risk of embolic events compared with those continuously 2 Methods receiving the recommended treatment; poor adherence to anticoagulation can impact stroke protection. 2.1 Study Population This was the first prospective, randomized study investi- gating the impact of an education program on total days This phase IV randomized study was conducted at 171 sites of adherence and persistence to a non-vitamin-K oral in seven European countries (Belgium, France, Germany, anticoagulant (apixaban). Italy, Spain, Switzerland, and the UK). Patients with newly diagnosed or existing documented NVAF or atrial flutter, Implementation adherence and persistence rates were and a C­ HADS2 score ≥ 1 were eligible. Patients were either high over the first 48 weeks of administration, with VKA experienced (≥ 3 months’ treatment; target was one- attrition over time, with or without additional patient third of randomized population) or VKA naïve (≤ 30 days’ education; lack of impact of the educational interven- VKA treatment in the last 12 months). All patients were tion may be related to the experience of the investigators NOAC naïve. Patients with NVAF or atrial flutter who were (high level of standard care) and limited uptake of digital receiving aspirin ≤ 165 mg once daily for an indication other support strategies. than atrial fibrillation were eligible. The inclusion of patients receiving concomitant aspirin was in line with those of the pivotal ARISTOTLE trial [2]. In the absence of routine anticoagulation monitoring, it is Patients were excluded if they had other conditions important that patients treated with NOACs take their medi- requiring anticoagulation; had conditions associated with cation as prescribed during all three phases of adherence: increased risk of hemorrhage; were at high risk of bleeding initiation (when the first dose is taken); implementation (the (e.g., due to other concomitant antiplatelet treatment or as extent to which a patient’s actual dosing corresponds to the a result of elective surgery or invasive procedures); were prescribed dosing); and persistence (absence of discontinu- unable to self-administer study medication; were hospital- ation, defined as the time when the next dose is omitted and ized; or were in long-term residential care (full details in the no more doses are taken) [8]. Poor adherence may decrease Electronic Supplementary Material [ESM]). protection from stroke [9]. Patients consented to, and initiated therapy with, apixa- Information on adherence to NOACs during the imple- ban ­(Eliquis®, Bristol-Myers Squibb/Pfizer EEIG, Uxbridge, mentation phase is limited, with most studies instead UK) and were subsequently asked to participate in the moni- describing persistence. Compared with warfarin, persistence toring phase of the study. After patients provided consent to (at 2.0, 1.9, 2.8, and 1.8 years) with NOACs in pivotal trials participate, they were centrally randomized (via electronic has been shown to be lower for dabigatran, similar for rivar- case report form) in a 1:1 ratio to receive standard-of-care oxaban or edoxaban, and higher for apixaban, respectively (SOC) patient information or an additional educational pro- [2–6]. gram. After 24 weeks, patients receiving the educational Dedicated educational interventions aim to improve program were re-randomized 1:1 to continue receiving the adherence by improving patients’ understanding of NVAF, educational program or switch to SOC for an additional goals of treatment, and risks associated with non-adherence. 24 weeks (Fig. 1). The trial design and the protocol were Many observational studies, but only a few small rand- approved by the relevant national regulatory authorities and omized clinical studies, have specifically investigated the national and/or local ethics committees. All patients pro- impact of educational programs on adherence to warfarin vided written informed consent. The trial is registered at [10, 11]. While several observational studies have assessed http://www.clini​caltr​ials.gov (NCT01884350). adherence to NOACs [12–19], no studies have investigated the impact of education on NOAC adherence in patients with 2.2 Study Procedures NVAF. This study, the AEGEAN (Assessment of an Education Most patients received standard apixaban 5 mg tablets twice and Guidance program for Eliquis Adherence in non-val- daily for 48 weeks. Apixaban 2.5 mg tablets twice daily were vular Atrial fibrillatioN), investigated the impact of a pro- allocated to patients with creatinine clearance of 15–29 ml/ active educational program on implementation adherence min or two or more of the following: age ≥ 80 years, (from first administration of apixaban to discontinuation/ body weight ≤ 60 kg, or serum creatinine ≥ 133 µmol/l. Impact on Apixaban Adherence in Atrial Fibrillation 63 Commercial packages (over-labelled for clinical trial use application; and (3) access to a virtual clinic organized at the only, according to country regulations) were provided. All country level, utilizing staff from an existing anticoagulant sites were provided with 5 mg tablets and 2.5 mg tablets in clinic (virtual clinic locations listed in ESM 3). Patients were separately labelled, numbered packs. During the telephone called by the virtual clinic within 1 week of randomization randomisation procedure, the dose-reduction criteria were to ensure they had correctly understood the disease, rationale entered, and the site staff were advised which subject pack for anticoagulant treatment, and importance of adherence. they would dispense (5 or 2.5 mg packs), according to the They were also given general information and advice regard- dose-reduction criteria for apixaban in the approved sum- ing anticoagulant treatment, information on symptoms of mary of product characteristics. Sites were allocated a new bleeding or thromboembolic events, and a toll-free num- patient pack with the appropriate dose for any subsequent ber for further questions. Follow-up calls to reinforce the dose reduction required (at a subsequent visit). importance of adherence and provide additional advice (if Patients were asked not to break the 5 mg tablets if they needed) occurred 10–20 days after the first call and monthly were allocated to the 2.5 mg dose twice daily and not to take thereafter. Patients could call the virtual clinic as often as two 2.5 mg tablets of apixaban if they were allocated to 5 mg needed. All information was provided using the official local twice daily. The EMD could accommodate blister packs of language(s). After randomization, five visits were scheduled either dosage. during the study (weeks 4, 12, 24, 36, and 48; Table 1 in All patients received SOC information about apixaban the ESM). treatment from the investigator and the standard patient information leaflet and information card, according to the 2.3 Outcome Measures usual practice at each study center. Patients assigned to the additional educational program also received (1) a patient The primary outcome was implementation adherence of education booklet explaining NVAF and anticoagulant treat- the dosing regimen (the proportion of days adhering to the ment for stroke prevention; (2) one or more of the reminder prescribed regimen) at 24 weeks, assessed using an EMD, tools (chosen at the patient’s discretion): key ring, short Helping ­Hand® (WestRock Switzerland Ltd., Sion, Switzer- message service alert on mobile phone, or smartphone land), an established monitoring device previously utilized Fig. 1  Study design. Patients consented to and initiated therapy with tinuations see Table 3 in the Electronic Supplementary Material. ASA apixaban and were subsequently asked to participate in the imple- acetylsalicylic acid, NVAF non-valvular atrial fibrillation, OAC oral mentation phase of the study. Following consent, patients were cen- anticoagulant, R randomization, SOC standard of care, VKA vitamin- trally randomized in a 1:1 ratio to receive standard-of-care patient K antagonist information or an additional educational program. For subject discon- 64 G. Montalescot et al. in adherence studies [21–24]. This study used the basic 2.4 Statistical Analysis model (model number: 1110-01), which had no display or reminder function to enhance implementation (see Fig. 1 The study had 85% power to detect a between-group differ- in the ESM). The EMD was operated by inserting com- ence of 4% in the percentage of days with a correct execution mercial blister packs into the device, which then electroni- of the dosing regimen during 24 weeks, assuming an intra- cally recorded every time the blister was removed (date and group SD of 21.1%. Anticipating a dropout rate of 10%, time). It was assumed that a single dose of study medication 1112 patients, 556 per group, were needed to obtain a sam- was administered every time the blister pack was removed. ple size of approximately 1000 evaluable patients (primary Sites uploaded data from the device at every study visit, but outcome analysis). investigators and patients remained blinded to the data. The The primary outcome, calculated as the percentage of secondary outcome of the study was implementation adher- days of implementation adherence during the period for each ence to the dosing regimen at 48 weeks. Days of adherent study group, was tested by the two-sample independent t and non-adherent implementation were defined as follows: tests at the 5% significance level (two-sided). Only the time period for which patients were persistent was taken into con- • Day of adherent implementation: 24-h window during sideration for the analysis. Homogeneity of variances was which checked graphically and with the F test. The Cochran t test was used for unequal variances between the two groups. – the treatment was taken twice daily as prescribed Within each study group, the percentage of days with (one tablet two times daily) a correct execution of apixaban dosing during the first 12, – only one isolated tablet was missed in 24 h, and no 12- to 24, and 24- to 48-week periods was compared using tablet was missed the previous day or the day after) a paired t test, at a two-sided α level of significance of 0.05. (see exception below). The percentage of patients within each group who were • Day of non-adherent implementation: 24-h window dur- “persistent” over time was described using Kaplan–Meier ing which estimates. If treatment with apixaban was withheld tempo- rarily by the investigator during the study period, the time – two consecutive tablets within 24 h were missed window during which apixaban was withheld was not con- – one tablet was missed for several consecutive days, sidered for adherence measurement purposes. Study treat- with the first day considered a day of adherent imple- ment that was withheld for more than 30 consecutive days mentation was considered a permanent discontinuation. – one tablet was missed on alternate days, every sec- Descriptive statistics were used to summarize clinical ond day with a missed tablet constituting a day of outcomes per group to evaluate the effect of the educational non-adherent implementation. program on efficacy and safety outcomes. The study was not powered to detect outcome differences between the groups. The details on how adherence units were calculated can be found in the submitted statistical analysis plan (ESM, Sect. 8.4.2). Thus, the primary outcome is a measure of the 3 Results optimality of medication intake and reflects both imple- mentation and persistence, with any non-persistent period 3.1 Patients and Procedures corresponding to non-adherent days. Other study objectives included percentage of “persistent” patients (i.e., patients Between 21 November 2013 and 19 December 2014, a who temporarily, but not permanently, discontinued their total of 1228 patients consented and were enrolled in the treatment regimen at the discretion of the investigator) AEGEAN study. Among them, 1162 were randomized to and evaluation of the effect of the educational program on either SOC (n = 583) or the additional educational pro- clinical outcomes (death, myocardial infarction, stroke, gram (n = 579). Of 524 patients remaining in the additional transient ischemic attack or SE, venous thromboembolism, educational program at week 24, a total of 263 were re- major bleeding [non-fatal] or clinically relevant non-major randomized to SOC (secondary SOC) and 261 continued bleeding, and fatal bleeding). Bleeding events were defined the additional educational program to week 48 (patient flow according to the International Society on Thrombosis and shown in Fig. 2 in the ESM). Baseline patient characteris- Haemostasis criteria (ESM 4). All ischemic, hemorrhagic tics were well-balanced between the two initial randomized events and deaths were adjudicated by an independent com- groups (Table 1). mittee (ESM 2) who were blinded to patients’ study group. The proportions of patients communicating with the vir- tual clinic and using reminders at 24 weeks are presented Impact on Apixaban Adherence in Atrial Fibrillation 65 in Table 2 in the ESM. Overall, > 95% of patients in the still considered persistent if they temporarily, but not per- additional education program group had communicated with manently, discontinued their treatment regimen. The mean the virtual clinic during the first 24 weeks. Most patients total interruption of study drug up to 24 weeks was 1.2 (± used the home phone (> 80%) as the preferred method of 3.6) days in the continued additional educational program communication with the clinic and dose reminders (> 40%) group, 1.8 (± 8.3) days in the primary SOC group and 1.9 as the preferred type of reminder. Few (< 1%) patients (± 5.5) days in the secondary SOC group. Mean total inter- used email or a smartphone app. Nearly half (45.3%) of the ruption of study drug up to 48 weeks was 2.2 (± 5.9) days in patients in the additional education program group chose to the continued additional educational program group, 3.8 (± receive one reminder; few patients (< 1%) chose to receive 19.6) days in the primary SOC group and 4.8 (± 18.0) days two reminders. in the secondary SOC group. Comparisons of implementation adherence between the 3.2 Implementation Adherence and Persistence SOC and the additional educational program for different subgroups of patients demonstrated no significant differ- In the additional educational program group, the mean ± SD ences in the likelihood of having < 80% implementation (median [minimum, maximum]) percentage of days of adherence (Fig. 3 in the ESM; p > 0.1 for all subgroup implementation adherence (i.e., the primary outcome) was comparisons). 91.9 ± 16.1% (98.2% [3%, 100%]) (Fig. 2). In the SOC group, the mean implementation adherence was 91.6 ± 3.3 Clinical Outcomes 17.1% (98.2% [1%, 100%]) (Fig. 2). By 48 weeks, imple- mentation adherence had declined to 90.4 ± 18.0% (97.3% Adjudicated clinical events are summarized in Table 2. [2%, 100%]) in the continued additional educational pro- Overall, 4.8% of patients had at least one adverse event gram group; 90.1 ± 18.6% (97.9% [1%, 100%]) in the SOC leading to permanent discontinuation (4.5% in the additional group; and 89.3 ± 18.1% (97.9% [2%, 100%]) in the sec- educational group; 5.1% in the SOC group). Between 24 ondary SOC group. There were no significant differences and 48 weeks, 2.2% of patients had an adverse event lead- between groups at 24 or 48 weeks (Fig. 3; p > 0.7 for all ing to permanent discontinuation (1.1, 2.2, and 3.0% in the between-group comparisons). Up to 48 weeks, the average additional educational, SOC, and secondary SOC groups, percentage of missed doses was low in each of the additional respectively). Bleeding events were low and evenly distrib- educational program, SOC, and secondary SOC groups uted across groups (Table 2). (24 weeks: 4.4 ± 8.2%, 4.9 ± 8.5%, and 4.7 ± 7.2%, respec- tively; 48 weeks: 4.9 ± 8.6%, 5.9 ± 9.3%, and 5.7 ± 7.8%). At 24 weeks, the percentage of fully adherent subjects was 4 Discussion 28.1% in the primary SOC group and 26.5% in the additional educational program group, with no significant difference This was the first prospective, randomized study investigat- between groups. The median duration of full adherence was ing the impact of an education program on adherence to 58.0 days in the primary SOC group and 59.0 days in the an NOAC. In daily practice, many patients do not adhere additional educational program group. to lifesaving treatments or to treatments that can prevent Persistence declined gradually across 48 weeks in all serious complications and/or improve quality of life [8, groups (Fig. 4). However, the percentage of patients per- 25–27]. However, strict implementation of NOAC intake is sistent at 48 weeks was high: 86.1% (95% confidence inter- crucial because its anticoagulant effect wanes within 12–24 h val [CI] 81.3–89.7), 85.2% (95% CI 81.5–88.2), and 87.8% after the last intake [7]. The AEGEAN study showed that (95% CI 83.4–91.1) in the continued additional educational implementation adherence and persistence rates with apixa- program, SOC, and secondary SOC groups, respectively ban therapy for NVAF are high over the first 48 weeks of (p > 0.5 for all between-group comparisons). At week administration (~ 90% for the implementation phase), with 48, only 9.2% (n = 100) of patients had missed three con- attrition of implementation adherence and persistence over secutive doses at any time during the study period. At this time. While most studies present rates of adherent patients, timepoint, the mean ± standard error percentage of days rather than the total days of adherence as per the primary of implementation adherence was 90.8 ± 18.0% (median outcome in this study, rates of persistence compare favora- 98.1% [1%, 100%]) in previously VKA-treated patients and bly with previously reported rates in other NOAC studies 89.6 ± 18.5% (97.6% [2%, 100%]) in VKA-naïve patients (persistence of 56–74% over a median follow-up duration of (p = 0.32). In addition, the likelihood of having < 80% up to 36 weeks) [17, 28, 29]. However, these studies were implementation adherence in previously VKA-treated and retrospective with significant measurement bias [28, 29]. VKA-naïve patients was similar in all groups (p > 0.05 for Although the deployment of a 24- or 48-week proactive edu- all between-group treatment comparisons). Patients were cational program in the AEGEAN study did not substantially 66 G. Montalescot et al. Table 1  Baseline patient characteristics Characteristics Additional educational program (Na = 579) Primary SOC (Na = 583) Age (years) 73.1 ± 9.1 72.6 ± 8.9 Age ≥ 75 years 282 (48.7) 276 (47.3) Female 234 (40.4) 232 (39.8) Weight (kg) 81.7 ± 16.7 82.8 ± 17.7 Heart rate (bpm) 74.9 ± 16.7 74.3 ± 15.9 SBP (mmHg) 132.7 ± 16.4 135.0 ± 17.1 DBP (mmHg) 78.0 ± 10.7 79.5 ± 10.8 MMSE 28.2 ± 1.7 28.3 ± 1.6 Type of diagnosis AF 557 (96.4) 562 (96.4) Atrial flutter 19 (3.3) 18 (3.1) AF + atrial flutter 2 (0.3) 3 (0.5) Type of NVAF Paroxysmal 301 (52.4) 280 (48.4) Persistent 137 (23.9) 148 (25.6) Permanent 136 (23.7) 150 (26.0) Stroke and bleeding risk factors Congestive heart failure 93 (16.1) 83 (14.2) Hypertension 497 (85.8) 516 (88.5) Diabetes mellitus 151 (26.1) 154 (26.4) History of stroke or TIA 107 (18.5) 94 (16.1) History of coronary heart disease/stent or PVD 142 (24.5) 138 (23.7) Renal ­failureb 11 (1.9) 7 (1.2) CHADS2 2.1 ± 1.0 2.1 ± 1.1 CHA2DS2-VASc 3.6 ± 1.3 3.6 ± 1.4 HAS-BLED 1.7 ± 0.8 1.7 ± 0.8 High risk of bleeding (HAS-BLED ≥ 3) 87 (15.2) 87 (15.1) Baseline VKA status VKA naïvec 370 (63.9) 371 (63.6) VKA ­treatedd 207 (35.8) 210 (36.0) Concomitant medications at baseline 5.1 ± 2.9 5.0 ± 2.7 Previous treatment with ASA 114 (19.7) 116 (19.9) Concomitant treatment with ASA 100 (17.9) 89 (14.8) Dose of study drug assigned at baseline (mg) 5 499 (86.2) 512 (88.1) 2.5 80 (13.8) 69 (11.9) ASA acetylsalicylic acid, AF atrial fibrillation, DBP diastolic blood pressure, MMSE Mini Mental State Examination, NVAF non-valvular atrial fibrillation, PVD peripheral vascular disease, SBP systolic blood pressure, SOC standard of care, TIA transient ischemic attack, VKA vitamin K antagonist Values are mean ± standard deviation or n (%) a N represents the number of patients contributing to summary statistics. All percentages are based on the number of patients with available data for each characteristic b Creatinine clearance < 30 ml/min c VKA-naïve patients must not have received VKA treatment for > 30 days within the last 12 months d VKA-treated patients must have received the VKA treatment for > 3 months increase the implementation adherence and persistence rates, adherence and persistence findings may reflect the impact the very high SOC rates are encouraging, and these data are of continued follow-up of patients in the SOC group, with consistent with the reported data for apixaban [28, 29]. The five follow-up visits in the first year, the importance of Impact on Apixaban Adherence in Atrial Fibrillation 67 study meant that adherence data feedback was not possible. However, a recently published telemonitoring study using daily patient feedback and employing an EMD showed very high adherence (89.6% over a 3-month observation period) to NOAC therapy despite the small sample size, probably largely attributable to being monitored [18]. Implementation adherence and persistence rates reported with VKAs are generally lower than those with NOACs, with historical cohort studies and clinical trials reporting rates of 42–78% (duration of follow-up of 3–84 months) [9]. This may be attributed to burdensome therapeutic monitor- ing and dose-adjustment requirements associated with these agents [32], as well as drug–drug and drug–food interactions Fig. 2  Implementation of the correct apixaban dose by study group at that need to be explained in detail at treatment initiation week 24 (mean; primary outcome). Implementation adherence (treat- and require ongoing education. One-third of the patients in ment taken as prescribed with one or less dose missed within 24 h AEGEAN were VKA experienced and switched to apixa- and no tablet missed on the previous 2 consecutive days) was moni- ban at randomization. In these patients, implementation tored by an electronic monitoring device. Not all patients handed in the device at week 24 adherence in the SOC group was similar to that observed in anticoagulant-naïve patients. The lack of a significant benefit of our proactive educa- tional program on apixaban implementation adherence and persistence was unexpected. In addition to the high imple- mentation adherence rate with SOC (~ 90%), several fac- tors may have offset the effect of the educational program: (1) investigators in the AEGEAN trial are likely to be more experienced and well-trained in prescribing NOACs and pro- viding more NOAC-specific information than the general population of physicians, narrowing the gap between SOC patient information and the additional educational program; (2) limited patient uptake of dose reminder strategies, espe- cially those associated with email and cell phones, possibly because of the advanced age of the patients; and (3) the nature of the virtual clinic, as recent findings suggest that face-to-face engagement can maximize effectiveness [17]. Nonetheless, the virtual clinic contacted patients in the Fig. 3  Overall implementation of the correct apixaban dose (mean) intervention arm regularly and proactively, albeit remotely, by study group. P > 0.7 for all between-group comparisons at 24 and which would act as an intermittent reminder for medica- 48 weeks. Apixaban with standard of care = patients randomized to receive only standard-of-care information following apixaban initia- tion adherence, and most virtual clinics reported positive tion; Apixaban with additional educational program = patients rand- patient interactions. It is also possible that a major differ- omized to receive an additional educational program following apixa- ential effect would be expected on the initiation phase of ban initiation; Apixaban with secondary standard of care = patients adherence (patients who start/do not start treatment), which with educational program switched to standard-of-care information only following 24 weeks of apixaban treatment. Additional data from was not measured in this study as most patients had initiated all patients handing in the device at the end of the trial were included treatment in hospital. Finally, a lack of observed effects from for calculating implementation adherence by week, up to week 48 similar educational programs has been reported in clinical situations other than NVAF [33, 34], reinforcing the diffi- culties of such initiatives. In contrast, the recent FACILITA providing appropriate NOAC-specific information at ini- study showed that a mixed intervention, consisting of patient tiation (as recommended [30]), as well as the use of the education and a simple calendar reminder of drug intake, EMD. While the EMD used in this study did not incorpo- was an effective strategy to improve dabigatran adher- rate a reminder function, the monitoring of medication use ence (to 91% and 89% at 6 and 12 months, respectively, could itself act as a continuous reminder (the Hawthorne compared with 65% and 63% for the control group) [19]. effect) [31] and act as a motivator for implementation adher- Also, in the IMPACT-AF study, the use of oral anticoagula- ence. In the AEGEAN study, the double-blind aspect of the tion in patients with atrial fibrillation was improved by a 68 G. Montalescot et al. Fig. 4  Persistence by study group over 48 weeks (second- ary outcome). Apixaban with standard of care = patients randomized to receive only standard-of-care information following apixaban initiation; Apixaban with additional educational program = patients randomized to receive an additional educational program following apixaban initiation; Apixaban with secondary standard of care = patients with educational program switched to standard-of-care information only following 24 weeks of apixaban treatment Table 2  Adjudicated clinical endpoints at 24 and 48 weeks of follow-up Endpoint 24 weeks 24–48 weeks Apixaban, AEP Apixaban, pSOC Apixaban, AEP Apixaban, pSOC Apixaban, sSOC (N = 583) (N = 579) (N = 261) (N = 583) (N = 263) Patients Events Patients Events Patients Events Patients Events Patients Events Death 4 (0.7) 4 4 (0.7) 4 1 (0.4) 1 1 (0.2) 1 3 (1.1) 3 Cardiovascular death 2 (0.3) 2 4 (0.7) 4 0 0 1 (0.2) 1 2 (0.8) 2 Stroke, TIA, SE 5 (0.9) 6 1 (0.2) 1 1 (0.4) 1 1 (0.2) 1 3 (1.1) 3 Ischemic stroke 0 0 1 (0.2) 1 1 (0.4) 1 0 0 0 0 Hemorrhagic stroke 2 (0.3) 3 0 0 0 0 0 0 1 (0.4) 1 TIA 0 0 0 0 0 0 1 (0.2) 1 1 (0.4) 1 SE 3 (0.5) 3 0 0 0 0 0 0 1 (0.4) 1 Myocardial infarction 4 (0.7) 4 2 (0.3) 2 0 0 1 (0.2) 1 0 0 Venous thromboembolism 1 (0.2)a 1 0 0 1 (0.4) 1 0 0 0 0 Pulmonary embolism 1 (0.2) 1 0 0 0 0 0 0 0 0 Deep vein thrombosis 1 (0.2) 1 0 0 1 (0.4) 1 0 0 0 0 MB (non-fatal) or CRNMB 9 (1.6)b 10 7 (1.2) 7 0 0 2 (0.3) 2 2 (0.8) 2 MB (non-fatal) 4 (0.7) 4 2 (0.3) 2 0 0 2 (0.3) 2 1 (0.4) 1 CRNMB 5 (0.9) 5 5 (0.9) 5 0 0 0 0 0 0 Fatal bleeding 1 (0.2) 1 0 0 0 0 0 0 1 (0.4) 1 AEP additional educational program, CRNMB clinically relevant non-major bleeding, MB major bleeding, pSOC primary standard of care, SE systemic embolism, sSOC secondary standard of care, TIA transient ischemic attack Data are presented as N or N (%) a One patient, who had one event of deep vein thrombosis + pulmonary embolism b One patient had a major and fatal bleeding multifaceted and multilevel educational program, where the a significant improvement after intervention. Moreover, provision of physician education targeted the initiation phase education was provided to not only patients but also their of adherence as well as the implementation and persistence families and healthcare providers, an important aspect of phases [35]. However, these studies had some notable differ- VKA treatment. There was also regular monitoring with a ences from the present study. Most patients in IMPACT-AF feedback mechanism providing relevant information to coor- (> 80%) were anticoagulated with VKA, with poor imple- dinating centers, such as notifications when patients were mentation of treatment at baseline (66%), leaving room for not taking their drugs and allowing physician intervention Impact on Apixaban Adherence in Atrial Fibrillation 69 to prevent discontinuation of anticoagulation and improve which could have impacted adherence and reduced the effect persistence [35]. Despite these efforts, the persistence rate of the educational program on this outcome. observed in IMPACT-AF remained lower than seen in the In conclusion, our study demonstrated high implementa- current study. tion adherence and persistence rates during the first 48 weeks Our study also has limitations. The selection of participat- in patients started on apixaban for NVAF and highlighted the ing patients may not fully represent patients in real life, as importance of regular patient follow-up and education in patients willing to participate in an adherence study are more maintaining favorable adherence to NOAC therapy in both likely to be better adherers. Furthermore, patient consent to SOC and intervention arms of the study. participate in the implementation phase was only obtained once patients had consented to initiating apixaban therapy. Acknowledgements The authors acknowledge editorial support from Lucid Group (Stuart Wakelin, Claire Price, Angela Pozo Ramajo, Thus, patients who did not consent to the study, possibly due Susan Cheer, Lisa Tatler, Jessica Stevens, Vatsala Malik, Eman Zaman, to lack of motivation (and who might have exhibited poorer and Emer O’Connor), funded by Bristol-Myers Squibb and Pfizer. The implementation adherence or persistence), were not included Bristol-Myers Squibb and Pfizer Alliance thank the patients who volun- in the trial. While the follow-up schedule was selected to be teered to participate and their caregivers, physicians, and clinical study teams for their contribution to the advancement of patient care. The close to the SOC for most countries, in some centers this image of the EMD was provided courtesy of Aardex group. would constitute additional visits that may have acted as a reminder and improved implementation adherence. Our Compliance with Ethical Standards results apply to the duration of the study, and adherence may further decrease over time beyond 48 weeks—we do not Funding This work was supported by Bristol-Myers Squibb (Paris, know the impact of education beyond this period. Measuring France) and Pfizer Limited (Tadworth, UK). The sponsor was involved implementation adherence improves it per se, as the patient with the design and conduct of the study and the collection, manage- ment, analysis, and interpretation of the data. is aware they are being monitored, resulting in overestima- tion. However, this effect was minimized because neither the Author contributions Trial conduct and data analysis were overseen patient nor the investigator was aware of the implementation by the executive steering committee (ESM 3) in collaboration with adherence data during the trial. AEGEAN used a low-bias, the study sponsors (Bristol-Myers Squibb and Pfizer Ltd). The coor- dinating center was the ACTION Study Group at Pitié-Salpêtrière rich sampling method (EMD Helping ­Hand®) that precisely hospital. Data were collected, managed, and analyzed by two Clinical recorded the date and time of medication access, providing Research Organizations: Lincoln (data management, Boulogne-Billan- detailed information on implementation adherence (although court, France) and Everest (data analysis, Ontario, Canada). Statistical pill disposal after extraction from the packet cannot be ruled analyses were validated by an independent statistician from ACTION Academic Research Organization (http://www.actio​n-coeur​.org). The out) [20]. The type of EMD used may also have influenced study chairman had unrestricted access to data after database lock. The results across the study, as some models with displays pro- first author (GM) prepared the first draft of the manuscript, which was vide instant feedback. Instant positive feedback following revised by all authors, who made the decision to submit the manuscript intake of medication may encourage future implementation for publication; they take responsibility for data accuracy. adherence and persistence; thus, it is possible that overall adherence across all cohorts would increase if EMDs with Conflicts of interest GM reports research grants to his institution from ICAN, Fédération Française de Cardiologie, Medtronic, MSD, Pfizer, displays were used. We believe our definition of a unit of Sanofi-Aventis, Servier, and INSERM; consultancy fees from Beth non-adherence in implementation (missing two consecu- Israel Deaconess Medical, Brigham Women’s Hospital, Menarini, tive pills) is conservative. It has been argued that missing TIMI Study Group, and ACTELION; and educational support from three consecutive doses of an oral anticoagulant with a 12-h Cardiovascular Research Foundation, CME Resources, Europa, Else- vier, Lead-Up, WebMD CCC. CB, AD’A, and HD have received con- half-life would have the same impact on its pharmacokinetic sultancy fees from Bristol-Myers Squibb/Pfizer. BC reports personal profile as missing one (double) dose in a once-daily regi- fees from St Jude Medical and research grants from Daiichi Sankyo. men [36]. Thus, it is likely that the lower limit of apixaban’s HJC and LD report research grants to their institutions from Bristol- therapeutic window, where the risk of thrombosis begins to Myers Squibb/Pfizer. FE has received speaker honoraria from Bristol- Myers Squibb. DAL reports research grants to her institution from increase, is reached only after missing three or more doses. Bristol-Myers Squibb and Boehringer Ingelheim, and speaker and While the cost of medication may also impact implemen- consultancy fees from Boehringer Ingelheim, Bayer, and Bristol My- tation adherence and persistence, medications were fully ers Squibb/Pfizer. BB is a former employee of Bristol-Myers Squibb. reimbursed in all countries that participated in this study AC is an employee of Pfizer Inc. EV reports grants to his institution from Boehringer and Sanofi; has received consultancy fees from Ab- and therefore medication acquisition costs would not have bott, Bristol-Myers Squibb, Celgene, Daiichi Sankyo, European Car- impacted the outcome. Finally, the study enrolled patients diovascular Research Center, Fresenius, LFB, Lilly, Medtronic, Pfizer, with AF who were relatively stable compared with an unse- and Sorin Group, and has received lecture fees from Novartis. lected AF population, as hospitalized patients and those with an increased risk of bleeding were excluded. This may have Open Access This article is distributed under the terms of the Crea- tive Commons Attribution-NonCommercial 4.0 International License reduced the occurrence of clinical events such as bleeding, 70 G. Montalescot et al. (http://creativecommons.org/licenses/by-nc/4.0/), which permits any anticoagulant medications based on the Pharmacy Quality Alli- noncommercial use, distribution, and reproduction in any medium, ance measure. Curr Med Res Opin. 2015;31:2167–73. provided you give appropriate credit to the original author(s) and the 16. Hanemaaijer S, Sodihardjo F, Horikx A, Wensing M, De Smet source, provide a link to the Creative Commons license, and indicate PA, Bouvy ML, Teichert M. Trends in antithrombotic drug use if changes were made. and adherence to non-vitamin K oral anticoagulants in The Neth- erlands. Int J Clin Pharm. 2015;37:1128–35. 17. Shore S, Carey EP, Turakhia MP, Jackevicius CA, Cunningham F, Pilote L, et al. Adherence to dabigatran therapy and longitudinal References patient outcomes: insights from the veterans health administration. Am Heart J. 2014;167:810–7. 1. Hart RG, Pearce LA, Aguilar MI. Meta-analysis: antithrombotic 18. Desteghe L, Vijgen J, Koopman P, Dilling-Boer D, Schurmans therapy to prevent stroke in patients who have nonvalvular atrial J, Dendale P, Heidbuchel H. Telemonitoring-based feedback fibrillation. Ann Intern Med. 2007;146:857–67. improves adherence to non-vitamin K antagonist oral antico- 2. Granger CB, Alexander JH, McMurray JJ, Lopes RD, Hylek EM, agulants intake in patients with atrial fibrillation. Eur Heart J. Hanna M, ARISTOTLE Committees and Investigators, et al. 2018;39:1394–403. Apixaban versus warfarin in patients with atrial fibrillation. N 19. Márquez-Contreras E, Martell-Claros N, Márquez-Rivero S, Her- Engl J Med. 2011;365:981–92. mida-Campa E, Gracia-Diez C, Sanchez-López E, Gil-Guillén 3. Connolly SJ, Ezekowitz M, Yusuf S, Eikelboom J, Oldgren J, V, Compliance and Inertia Working Group, Spanish Society of Parekh A, RE-LY Steering Committee and Investigators, et al. Hypertension (SEH-LELHA). Strategies for improving dabigatran Dabigatran versus warfarin in patients with atrial fibrillation. N adherence for stroke prevention in patients with non-valvular atrial Engl J Med. 2009;361:1139–51. fibrillation: education and drug intake reminders (FACILITA 4. Connolly SJ, Ezekowitz MD, Yusuf S, Reilly PA, Wallentin L, study). Curr Med Res Opin. 2018;34:1301–8. Randomized Evaluation of Long-Term Anticoagulation Therapy 20. Osterberg L, Blaschke T. Adherence to medication. N Engl J Med. Investigators. Newly identified events in the RE-LY trial. N Engl 2005;353:487–97. J Med. 2010;363:1875–6. 21. Christensen A, Christrup LL, Fabricius PE, Chrostowska M, 5. Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, Wronka M, Narkiewicz K, Hansen EH. The impact of an elec- Breithardt G, ROCKET AF Investigators, et al. Rivaroxaban tronic monitoring and reminder device on patient compliance versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. with antihypertensive therapy: a randomized controlled trial. J 2011;365:883–91. Hypertens. 2010;28:194–200. 6. Giugliano RP, Ruff CT, Braunwald E, Murphy SA, Wiviott SD, 22. Kuypers DR, Peeters PC, Sennesael JJ, Kianda MN, Vrijens B, Halperin JL, ENGAGE AF-TIMI 48 Investigators, et al. Edoxaban Kristanto P, ADMIRAD Study Team, et al. Improved adherence versus warfarin in patients with atrial fibrillation. N Engl J Med. to tacrolimus once-daily formulation in renal recipients: a rand- 2013;369:2093–104. omized controlled trial using electronic monitoring. Transplanta- 7. Steffel J, Verhamme P, Potpara TS, Albaladejo P, Antz M, tion. 2013;95:333–40. Desteghe L, ESC Scientific Document Group, et al. The 2018 23. Dobbels F, De Bleser L, Berben L, Kristanto P, Dupont L, Nevens European Heart Rhythm Association Practical Guide on the use F, et al. Efficacy of a medication adherence enhancing intervention of non-vitamin K antagonist oral anticoagulants in patients with in transplantation: the MAESTRO-Tx trial. J Heart Lung Trans- atrial fibrillation. Eur Heart J. 2018;39:1330–93. plant. 2017;36:499–508. 8. Vrijens B, De Geest S, Hughes DA, Przemyslaw K, Demonceau J, 24. De Bleser L, De Geest S, Vandenbroeck S, Vanhaecke J, Dobbels Ruppar T, ABC Project Team, et al. A new taxonomy for describ- F. How accurate are electronic monitoring devices? A laboratory ing and defining adherence to medications. Br J Clin Pharmacol. study testing two devices to measure medication adherence. Sen- 2012;73:691–705. sors (Basel). 2010;10:1652–60. 9. Rodriguez RA, Carrier M, Wells PS. Non-adherence to new oral 25. Ruddy K, Mayer E, Partridge A. Patient adherence and persistence anticoagulants: a reason for concern during long-term anticoagu- with oral anticancer treatment. CA Cancer J Clin. 2009;59:56–66. lation? J Thromb Haemost. 2013;11:390–4. 26. Wu JR, Moser DK, Lennie TA, Burkhart PV. Medication adher- 10. Khan TI, Kamali F, Kesteven P, Avery P, Wynne H. The value ence in patients who have heart failure: a review of the literature. of education and self-monitoring in the management of warfarin Nurs Clin N Am. 2008;43:133–53. therapy in older patients with unstable control of anticoagulation. 27. van der Wal MH, Jaarsma T. Adherence in heart failure in Br J Haematol. 2004;126:557–64. the elderly: problem and possible solutions. Int J Cardiol. 11. Clarkesmith DE, Pattison HM, Lane DA. Educational and behav- 2008;125:203–8. ioural interventions for anticoagulant therapy in patients with atrial 28. Pan X, Kachroo S, Liu X, Kawabata H, Phatak H. Real world fibrillation. Cochrane Database Syst Rev. 2013;2013:CD008600. discontinuation rates with apixaban versus warfarin, dabigatran, 12. Shore S, Ho PM, Lambert-Kerzner A, Glorioso TJ, Carey EP, or rivaroxaban among atrial fibrillation patients newly initiated Cunningham F, et al. Site-level variation in and practices associ- on anticoagulation therapy: early findings. J Am Coll Cardiol. ated with dabigatran adherence. JAMA. 2015;313:1443–50. 2014;63:Suppl:A415. Abstract. 13. Diener HC, Aisenberg J, Ansell J, Atar D, Breithardt G, Eikel- 29. Yao X, Abraham NS, Alexander GC, Crown W, Montori VM, boom J, et al. Choosing a particular oral anticoagulant and dose Sangaralingham LR, et al. Effect of adherence to oral anticoagu- for stroke prevention in individual patients with non-valvular atrial lants on risk of stroke and major bleeding among patients with fibrillation: part 2. Eur Heart J. 2017;38:860–8. atrial fibrillation. J Am Heart Assoc. 2016;5:e003074. 14. Raparelli V, Proietti M, Cangemi R, Lip GYH, Lane DA, Basili 30. Lane DA, Aguinaga L, Blomström-Lundqvist C, Boriani G, Dan S. Adherence to oral anticoagulant therapy in patients with atrial GA, Hills MT, et al. Cardiac tachyarrhythmias and patient val- fibrillation: focus on non-vitamin K antagonist oral anticoagu- ues and preferences for their management: the European Heart lants. Thromb Haemost. 2016;117:209–18. Rhythm Association (EHRA) consensus document endorsed 15. McHorney CA, Crivera C, Laliberté F, Nelson WW, Germain G, by the Heart Rhythm Society (HRS), Asia Pacific Heart Bookhart B, et al. Adherence to non-vitamin-K-antagonist oral Rhythm Society (APHRS), and Sociedad Latinoamericana de Impact on Apixaban Adherence in Atrial Fibrillation 71 Estimulación Cardíaca y Electrofisiología (SOLEACE). Europace. 34. van Dalem J, Krass I, Aslani P. Interventions promoting adherence 2015;17:1747–69. to cardiovascular medicines. Int J Clin Pharm. 2012;34:295–311. 31. Braunholtz DA, Edwards SJL, Lilford RJ. Are randomized clinical 35. Vinereanu D, Lopes RD, Bahit MC, Xavier D, Jiang J, Al-Khalidi trials good for us (in the short term)? Evidence for a “trial effect”. HR, IMPACT-AF investigators, et al. A multifaceted intervention J Clin Epidemiol. 2001;54:217–24. to improve treatment with oral anticoagulants in atrial fibrillation 32. Kneeland PP, Fang MC. Current issues in patient adherence (IMPACT-AF): an international, cluster-randomised trial. The and persistence: focus on anticoagulants for the treatment and Lancet. 2017;390:1737–46. prevention of thromboembolism. Patient Prefer Adherence. 36. Vrijens B, Heidbuchel H. Non-vitamin K antagonist oral anti- 2010;4:51–60. coagulants: considerations on once- vs. twice-daily regimens 33. Cohen A, Assyag P, Boyer-Chatenet L, Cohen-Solal A, Perdrix and their potential impact on medication adherence. Europace. C, Dalichampt M, Réseau Insuffisance Cardiaque (RESICARD) 2015;17:514–23. PREVENTION Investigators, et al. An education program for risk factor management after an acute coronary syndrome: a rand- omized clinical trial. JAMA Intern Med. 2014;174:40–8. Affiliations Gilles Montalescot1 · Carlos Brotons2 · Bernard Cosyns3 · Harry J. Crijns4 · Armando D’Angelo5 · Ludovic Drouet6 · Franz Eberli7 · Deirdre A. Lane8 · Bruno Besse9 · Anthony Chan10 · Eric Vicaut11 · Harald Darius12 on behalf of the AEGEAN Study Investigators 1 7 Sorbonne University, ACTION Study Group, Institut de Department of Cardiology, Stadtspital Triemli, Zurich, Cardiologie (AP-HP), Centre Hospitalier Universitaire Pitié- Switzerland Salpêtriėre, 47 Boulevard de l’Hôpital, 75013 Paris, France 8 Liverpool Centre for Cardiovascular Science, University 2 Sardenya Primary Health Care Centre-Biomedical Research of Liverpool and Liverpool Heart & Chest Hospital, Institute Sant Pau (IIB Sant Pau), Barcelona, Spain Liverpool L7 8TX, United Kingdom 3 9 Department of Cardiology, Universitair Ziekenhuis Brussel, Global Clinical Research, Bristol-Myers Squibb, Paris, Brussels, Belgium France 4 10 Cardiovascular Research Institute Maastricht, Maastricht Internal Medicine, Pfizer Healthcare Ireland, Dublin, Ireland University Medical Centre, Maastricht, The Netherlands 11 Université Paris 7, the ACTION Study Group, Methodology 5 Coagulation Service and Thrombosis Research Unit, and Statistical Unit, Centre Hospitalier Universitaire Scientific Institute San Raffaele, Milan, Italy Lariboisière (AP-HP), Paris, France 6 12 AP-HP, Paris Diderot University, Paris, France Vivantes Neukoelln Medical Center, Berlin, Germany