Accepted Article

Received Date : 03-Jun-2014 Accepted Date : 08-Jan-2015 Article type

: Original Article - Cardiovascular Medicine

Dabigatran adherence in atrial fibrillation patients during the first year after diagnosis: a nationwide cohort study

Running title: Dabigatran adherence during the first year after AF diagnosis

Anders Gorst-Rasmussen, Senior Biostatistician1, 2, Flemming Skjøth , Senior Biostatistician1, 2, Torben

Bjerregaard Larsen , Associate Professor1, 2, Lars Hvilsted Rasmussen, Professor of Cardiovascular Medicine2, Gregory Y.H. Lip, Professor of Cardiovascular Medicine2, 3, Deirdre A. Lane, Senior Lecturer in

Cardiovascular Health2, 3

1

Department of Cardiology, Aalborg AF Study Group, Aalborg University Hospital, Sdr. Skovvej 15, 9100

Aalborg, Denmark 2

Aalborg Thrombosis Research Unit, Department of Clinical Medicine, Faculty of Health, Aalborg

University, Sdr. Skovvej 15, 9100 Aalborg, Denmark. 3

University of Birmingham, Centre for Cardiovascular Sciences, City Hospital, Dudley Road, Birmingham

B18 7QH, United Kingdom

This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process which may lead to differences between this version and the Version of Record. Please cite this article as an 'Accepted Article', doi: 10.1111/jth.12845 This article is protected by copyright. All rights reserved.

Accepted Article

Correspondence to: Dr Deirdre A Lane, University of Birmingham Centre for Cardiovascular Sciences, City Hospital, Dudley Road, Birmingham B18 7QH, United Kingdom Tel: +44 121 507 5080

E-mail: [email protected]

ABSTRACT Background: There is a perception among physicians that lack of routine monitoring with non-vitamin K antagonist oral anticoagulants (NOACs) may lead to poor medication adherence. We studied adherence during the first year of usage in a cohort of patients with newly diagnosed non-valvular atrial fibrillation (AF) started on the NOAC, dabigatran etexilate. Methods and results: Nationwide Danish patient and prescription purchase registries were used to identify newly diagnosed AF patients initiating dabigatran, co-morbidities, and refill patterns under a twice-daily, one pill regimen. Adherence was characterised among remaining users (N=2,960) after 1 year using the proportion of days covered (PDC), gap rates, and restart rates. The overall 1-year PDC was 83.9%, with 76.8% patients having a 1-year PDC in excess of 80%. Patients with CHA2DS2-VASc score ≥2 were more adherent than patients with CHA2DS2-VASc score of 1 (PDC

ratio: 1.12; 95% confidence interval [CI]: 1.08-1.17) and generally patients with higher morbidity were more adherent. In particular, patients with prior bleeding were more adherent than patients with no prior bleeding (PDC ratio: 1.02; 95% CI: 0.98-1.06). The overall gap rate was 1.4 gaps per year. There were no clear tendencies in gap rates among subgroups, although patients with higher morbidity tended to have slightly more, but shorter gap periods. Conclusion: More than 75% of patients were adherent >80% during the first year. Patients with higher

morbidity, including patients with a higher risk of stroke or bleeding, exhibited better adherence. This improvement may be attributable to more regular contact with the health care system.

This article is protected by copyright. All rights reserved.

Accepted Article

23

Andrade SE, Kahler KH, Frech F, Chan KA. Methods for evaluation of medication adherence and persistence using automated databases. Pharmacoepidemiol Drug Saf 2006; 15: 565–74; discussion 575–7.

24

Ho PM, Bryson CL, Rumsfeld JS. Medication adherence: its importance in cardiovascular outcomes. Circulation 2009; 119: 3028–35.

25

Roughead EE, Ramsay E, Priess K, Barratt J, Ryan P, Gilbert AL. Medication adherence, first episode duration, overall duration and time without therapy: the example of bisphosphonates. Pharmacoepidemiol Drug Saf 2009; 18: 69–75.

26

Tsai K, Erickson SC, Yang J, Harada AS, Solow BK, Lew HC. Adherence, persistence, and switching patterns of dabigatran etexilate. Am J Manag Care 2013; 19: e325–32.

27

Schulman S, Kearon C, Kakkar AK, Mismetti P, Schellong S, Eriksson H, Baanstra D, Schnee J, Goldhaber SZ. Dabigatran versus warfarin in the treatment of acute venous thromboembolism. N Engl J Med 2009; 361: 2342–52.

28

Schulman S. Advantages and limitations of the new anticoagulants. J Intern Med 2014; 275: 1–11.

29

January CT, Wann LS, Alpert JS, Calkins H, Cleveland JC, Cigarroa JE, Conti JB, Ellinor PT, Ezekowitz MD, Field ME, Murray KT, Sacco RL, Stevenson WG, Tchou PJ, Tracy CM, Yancy CW. 2014 AHA/ACC/HRS Guideline for the Management of Patients With Atrial Fibrillation: Executive Summary: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol 2014; .

30

Deedwania PC. New oral anticoagulants in elderly patients with atrial fibrillation. Am J Med 2013; 126: 289–96.

31

Vaughan Sarrazin MS, Cram P, Mazur A, Ward M, Reisinger HS. Patient perspectives of dabigatran: analysis of online discussion forums. Patient 2014; 7: 47–54.

32

Miller NH. Compliance with treatment regimens in chronic asymptomatic diseases. Am J Med 1997; 102: 43–9.

33

Horne R, Weinman J, Barber N. Concordance, adherence and compliance in medicine taking. Report for the National Coordinating Centre for NHS Service Delivery and Organisation (NCCSDO); 2005.

34

Clarkesmith DE, Pattison HM, Lip GYH, Lane DA. Educational intervention improves anticoagulation control in atrial fibrillation patients: the TREAT randomised trial. PLoS One 2013; 8: e74037.

35

Pamboukian SV, Nisar I, Patel S, Gu L, McLeod M, Costanzo MR, Heroux A. Factors associated with non-adherence to therapy with warfarin in a population of chronic heart failure patients. Clin Cardiol 2008; 31:30.

This article is protected by copyright. All rights reserved.

Accepted Article This article is protected by copyright. All rights reserved.

Accepted Article

Given the less stringent requirements for physician monitoring with dabigatran, it is of special interest to assess longer-term adherence in the population of AF patients who apparently remain on dabigatran therapy and do not experience bleeding issues. This population of relatively unproblematic patients is presumably less closely monitored, and adherence may suffer as a result[16] In the present study, we used nationwide Danish registries to describe dabigatran adherence (in the sense of regular dose taking according to recommendations[17] in the population of patients who initiated dabigatran therapy shortly after diagnosis of AF, and remained bleeding-free and plausible dabigatran users throughout the first year of therapy. Rather than trying to correlate adherence with outcomes, a complex and bias-prone exercise in the context of observational data[18], the aim of the present study was to provide a detailed, descriptive characterization of the extent and form of non-adherence in this particular patient population.

METHODS Data sources and study population Study data were obtained by linking three nationwide databases: (i) the Danish National Prescription Registry which records purchase date, Anatomical Therapeutic Chemical [ATC] classification code, and package size for every prescription purchase in Denmark[19]; (ii) the Danish National Patient Register which contains discharge diagnoses for all hospital admissions in Denmark[20]; and (iii) the Danish Civil Registration System[21] which holds information on sex, date of birth, vital and emigration status. All registry data were available up to 31 June 2014. We prioritized specificity over sensitivity for indirectly identifying a preliminary population of treatmentnaïve patients initiating dabigatran therapy for non-valvular AF. First, we identified patients who had received a hospital diagnosis of non-valvular AF during the period 1 August 2011 to 30 June 2013. Both inpatient and outpatient diagnoses were included. We then restricted the cohort to those purchasing dabigatran (110mg or 150mg) within 30 days after diagnosis. We finally excluded those who had purchased

This article is protected by copyright. All rights reserved.

Accepted Article

other anticoagulants within 2 years of the first dabigatran purchase date (switchers); and patients with knee or hip arthroplasty in a 10-day window around the date of the first dabigatran purchase. The final study population was that of plausible dabigatran users at 1 year after the first purchase. To obtain this population, we excluded from the preliminary population patients who within the first year experienced a switch in treatment (warfarin, rivaroxaban, or heparin purchase), bleeding event, death, emigration, or end of follow-up (30 June 2014). Note that exclusion due to emigration or end of follow-up is unlikely to bias results, since it can be reasonably regarded as “independent censoring”, i.e. the non-excluded study population is representative of the hypothetical complete population. The remaining patients were included in the final study population, irrespective of their actual purchase history in the year following the first dabigatran purchase. This entailed a ‘presumption of persistence’, in accordance with the recommendation of lifelong anticoagulant therapy for the majority of our study population[2]. Note that this final patient population is a selected population by definition and does not reflect all dabigatran users. Instead, it reflects the large subgroup of patients who are relatively ‘unproblematic’, in the sense that they neither switch therapy nor experience bleeding issues during the first year of therapy. Using the registries, we addressed the following clinical question: what was the extent of medication adherence among AF patients who ought to have been receiving dabigatran therapy for 1 year and if nonadherence was evident, what pattern did this follow? Specifically, we analysed non-adherence and treatment gap durations during the first year of therapy in a combined fashion, in order to identify subgroups with decreased adherence and to structurally characterise non-adherence, e.g. assess whether it resulted from a few long or many short gaps.

Patient baseline characteristics Information on co-morbidities and concomitant medications at date of first dabigatran purchase was obtained from the registries. Past-year cardiovascular prescriptions were determined from prescription purchases in the code C section of the Anatomical Therapeutic Chemical Classification System. History of bleeding was

This article is protected by copyright. All rights reserved.

Accepted Article

defined in terms of International Classification of Diseases [ICD] revision 10 (ICD-10) codes as a composite of: intracranial (I60, I61, I62, S063C, S064, S065, S066), major (D62, J942, H113, H356, H431, N02, N95, R04, R31, R58), or gastrointestinal bleeding (K250, K260, K270, K280, K290). History of stroke was defined as the composite of stroke (I63, I649), systemic embolic event (SE) (I74), or transient ischemic attach (TIA) (G45). Hospitalization was defined as any-cause hospitalization for at least one night. Lastly, we calculated a registry-based CHA2DS2-VASc score following the approach previously described[22].

Definition of outcome measures Adherence to dabigatran was defined as one-pill twice daily. Adequate adherence to dabigatran was categorized as adherence >80%[23]. A gap day was defined as any day during follow-up where dabigatran would be unavailable under a twice-daily, one pill regimen. A gap period was a sequence of consecutive gap days. We adopted a conservative default grace period of zero days, reflecting the short half-life of dabigatran[9]. More importantly, the short grace period, alongside the ‘presumption of persistence’ implicit in the definition of the 1-year study population, implied that all gap periods, long and short, contributed towards increased non-possession of medication. These conventions ensured a high degree of sensitivity to gaps, a preferable starting point when gap periods are an independent target of analysis. To account for medication provided during hospitalization, days with hospitalization were excluded from calculations of dabigatran availability. For each patient, we used the proportion of days covered (PDC) to describe 1-year adherence, i.e. the relative number of non-gap days during the first year. The gap count was defined as the number of gaps occurring during the first year of therapy, as exemplified in Figure 1. Within each gap period, we measured time from the first gap day and until the next dabigatran purchase (restart time). Restart rates were classified as: rapid - restart less than 1 week after stop; intermediate - restart 1-8 weeks after stop; and late - more than 8 weeks after stop. To characterise the distribution of restart times among patients at risk of reacquisition, we used the hazard rate of restarting.

This article is protected by copyright. All rights reserved.

Accepted Article

Statistical analysis Patient characteristics at baseline (time of first dabigatran purchase) were summarized descriptively. We calculated 1-year PDCs according to clinical subgroups and used Poisson regression with robust variance estimation to obtain PDC ratios and confidence intervals according to subgroups. We also calculated the average number of gaps (1-year gap rate) within subgroups. To characterise variation in gap patterns between subgroups (i.e. whether non-adherence was attributable to many small versus few large gap periods), we estimated 1-year gap rate ratios by using Poisson regression, adjusting for the 1-year PDC. The resulting effect measure could be interpreted as the relative gap count when comparing a typical subgroup patient with a patient with identical PDC from a reference group. Accordingly, a gap rate ratio less than 1 (respectively, greater than 1) implied that the non-reference patient on average arrived at their PDC through comparatively less numerous but longer (respectively, more but shorter) gaps. To analyse gap periods, we used weekly running averages to smooth the crude restart rates, while restart probabilities were estimated using the Kaplan-Meier method. Cox proportional hazards regression was used to calculate restart rate ratios according to subgroups, stratified according to the following three time periods: week 1 after gap start (rapid restart), week 1-8 after gap start (moderate restart) and more than 8 weeks after gap start (slow restart). Confidence intervals (CI) were estimated using cluster-robust methods to account for patients contributing multiple gap periods. Statistical analyses were performed using R version 3.2.0. A two-sided P-value less than 0.05 was considered statistically significant.

RESULTS The final study population of plausible dabigatran users at 1 year (N=2,960) was obtained from the preliminary study population of patients started on dabigatran due to AF (N=3,761) by excluding all those

This article is protected by copyright. All rights reserved.

Accepted Article

who, during the first year: died (N=232; 6.2%); experienced a bleeding event (N=99; 2.6%); or switched to warfarin, rivaroxaban, or heparins (N=470; 12.5%), see Figure 2. Baseline patient characteristics are shown in Table 1. Patients started on the 110 mg dose were substantially older (mean age 81.1 years) than patients started on the 150 mg dose (mean age 66.4 years). Baseline characteristics for the excluded patients in the preliminary study population were similar (data not shown). Females were generally older than men (mean age 75.3 years vs. 69.4 years) and had more comorbidities (data not shown).

Adherence 1 year after dabigatran initiation, PDC The 1-year mean PDC was 83.9% (standard deviation [SD] 27.7%), which increased to 84.9% when incorporating a 3-day grace period. Figure 3 shows a graph of the proportion of patients above a given PDC when the PDC varies from 0% to 100%. The proportion of patients with a 1-year PDC above 80% was 76.8%, and 37.2% of patients had a PDC of 100%. 1-year PDCs according to subgroups are shown in Figure 4. Females were significantly more adherent than males (PDC ratio: 1.06, 95% CI: 1.03-1.08). Patients who were regular users of cardiovascular drugs were significantly more adherent than non-regular medication users (PDC ratio: 1.12, 95% CI: 1.09-1.15), as were patients with a history of stroke/SE/TIA (PDC ratio: 1.09, 95% CI: 1.06-1.12). Patients with CHA2DS2-

VASc score ≥2 were also more adherent than the reference group with 1 CHA2DS2-VASc risk factor (PDC ratio: 1.12, 95% CI: 1.08-1.17).

Adherence among patients with a history of bleeding was not decreased compared to patients with no history of bleeding (PDC ratio: 1.02, 95% CI 0.98-1.06). Younger patients (aged 80 % of the time during the first year of usage. Second, patients at high risk of stroke (CHA2DS2-VASc score ≥2) were significantly more adherent than those with a CHA2DS2-VASc score of 1;

similar observations applied for other patient groups with relatively higher morbidity (e.g. patients with recent hospitalizations, patients who took several cardiovascular medications), including patients with prior bleeding. Third, women were significantly more adherent than men. The overall PDC of 83.9% in the present study was higher than in a recent registry-based study, which reported a PDC of 67.4%[26] in a warfarin-naïve cohort. The present PDC is still substantially lower than the estimates reported in the RE-COVER trial where adherence was 98%[27]; however, trial-based medication adherence may not be directly comparable with adherence estimates from clinical practice because trial patients are likely to be more motivated[15]. In our study, 76.8% of patients newly prescribed dabigatran demonstrated adequate adherence, a PDC>80%, which is well above the 50%-65% adequate

This article is protected by copyright. All rights reserved.

Accepted Article

adherence reported with other one-pill, twice-daily medications (metformin, glipizide, and metoprolol)[28]. In addition, our PDC estimates may be regarded as conservative, since we elected to let all and any gap periods count towards increased non-adherence, rather than incorporating default grace periods or adjusting for non-persistence. In our study, the high stroke risk subgroup of guideline-recommended[2,3,29] definite candidates for oral anticoagulation (CHA2DS2-VASc score ≥2) were more adherent than probable candidates (CHA2DS2-VASc

score =1). Similar findings applied when comparing patients started on a 110 mg dose to those started on a 150 mg dose. Patients with a CHA2DS2-VASc score ≥2 and those started on the 110 mg dose generally represented the elderly and frail AF subpopulation that would be expected to be in more frequent contact with the health system in general. This may in part explain the improved adherence. Indeed, we also saw higher adherence in other subgroups with relatively higher morbidity where more frequent health system contacts would be expected.

Another factor which could contribute to the improved adherence in the higher-morbidity subgroups is a better understanding of treatment necessity in patients with higher morbidity who are used to medical treatment. Patients may intentionally choose not to adhere or persist with chronic treatments if they hold inaccurate perceptions about their condition or recommended therapy; therefore, it is important that AF patients understand the importance of OAC for stroke prevention and that long-term adherence to the dosing regimen is essential to reduce the risk of adverse thromboembolic and haemorrhagic events. Indeed, a recent randomized controlled trial of a one-off behavioural-educational programme, the TREAT study, demonstrated a significant improvement in adherence to warfarin, evidenced by time in therapeutic range among warfarin-naive AF patients, by increasing patients understanding of the necessity of warfarin for stroke prevention and reducing their perception of medication harm[34]. Lastly, the finding that women were significantly more adherent than men is contrary to the finding of no gender differences in other studies of anticoagulant therapy [35, 36]. However, this apparent gender

This article is protected by copyright. All rights reserved.

Accepted Article

difference could be related to the improved adherence among patients with a higher morbidity, since females with AF are known to be both older and with more comorbidities than males [37]. Indeed, a slightly higher morbidity among females was also found in the present study.

Limitations While adherence measures based on prescription refills in a closed pharmacy system can be accurate at a population level[10], they lack distinction between purchase and consumption, making them less reliable for use at an individual level and in smaller subgroups. Additionally, they cannot be used to address the issue of overuse, which in the case of dabigatran is an equally serious concern because of the increased bleeding risk. Our indirectly defined population of AF patients initiating dabigatran therapy relied on registry information and may, despite a high validity[38], be subject to misclassification errors. We took into account bridging by excluding hospitalized days when calculating number of days covered, and by excluding patients who purchased other anticoagulants during the first year; however, we cannot rule out minor inaccuracies in adherence estimates due to bridging. Lastly, a deliberate design choice rather than a limitation per se, resulted in a selected, slightly ‘healthier’ study population compared with the source population, since patients who died or experienced bleeding during the first year after initiating dabigatran therapy were excluded. Future research is needed to investigate how adherence affects outcomes; owing to the longitudinal nature of the adherence process, time-dependent confounding is a serious concern and appropriate analysis is both complex and prone to bias[18]. Future studies are also needed to assess the impact of socioeconomic variables on adherence which, considering the high cost of dabigatran, may be substantial in countries where patients have to bear medication costs[39]. In conclusion, in this nationwide prescription refill study of dabigatran adherence patterns, more than 75% of patients were adherent with dabigatran for >80 % of the time during the first year of usage. Patients at high stroke risk (CHA2DS2-VASc score ≥2) exhibited better medication adherence as their intermediate risk

(CHA2DS2-VASc score=1) and patients with prior bleeding were not less adherent. Regular contact with the

This article is protected by copyright. All rights reserved.

Accepted Article

health care system as well as patient understanding the necessity of therapy may improve adherence with non-vitamin K antagonist oral anticoagulants. Identification of the extent and form of non-adherence in clinical subgroups is essential for targeting patient education and improving overall medication adherence.

Addendum A. Gorst-Rasmussen, F. Skjøth, T. B. Larsen and D. A. Lane provided the idea for the article. A. GorstRasmussen and F. Skjøth conducted the analyses. A. Gorst-Rasmussen wrote the first version of the manuscript and D. A. Lane contributed to writing and subsequent revisions. F. Skjøth, T. B. Larsen, L. H. Rasmussen and G. Y. H. Lip contributed to manuscript revisions. A. Gorst-Rasmussen and T. B. Larsen are the guarantors.

Competing interests statement This study was supported by The Obel Family Foundation. The sponsor had no role in the study design; in the collection, analysis, and interpretation of the data; in the writing of this report; or in the decision to submit the paper for publication. A. Gorst-Rasmussen and F. Skjøth declare no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; or any other relationships or activities that could appear to have influenced the submitted work. T. B. Larsen has served as an investigator for Janssen Scientific Affairs, LLC and Boehringer Ingelheim and has been on the speaker bureaus for Bayer Healthcare, Bristol Myers Squibb/Pfizer, Roche Diagnostics, Boehringer Ingelheim and Takeda Pharma. L. H. Rasmussen has been on the speaker bureaus for Bayer Healthcare, Bristol Myers Squibb/Pfizer, Roche Diagnostics, Boehringer Ingelheim and Takeda Pharma. G. Y. H. Lip has served as a consultant for Bayer Healthcare, Astellas, Merck, AstraZeneca, Sanofi Aventis, Bristol Myers Squibb/Pfizer and Boehringer Ingelheim, and has been on the speaker bureaus for Bayer, Bristol Myers Squibb/Pfizer, Boehringer

This article is protected by copyright. All rights reserved.

Accepted Article

Ingelheim and Sanofi Aventis. D. A. Lane has received an investigator initiated educational grants from Bayer Healthcare and Boehringer Ingelheim and has been on the speaker bureau for Boehringer Ingelheim, Bayer, and Bristol Myers Squibb/ Pfizer. D. A. Lane has also received royalties from BMJ during the conduct of the study.

REFERENCES

1

Kneeland PP, Fang MC. Current issues in patient adherence and persistence: focus on anticoagulants for the treatment and prevention of thromboembolism. Patient Prefer Adherence 2010; 4: 51–60.

2

Camm AJ, Kirchhof P, Lip GYH, Schotten U, Savelieva I, Ernst S, Van Gelder IC, Al-Attar N, Hindricks G, Prendergast B, Heidbuchel H, Alfieri O, Angelini A, Atar D, Colonna P, De Caterina R, De Sutter J, Goette A, Gorenek B, Heldal M, et al. Guidelines for the management of atrial fibrillation: the Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Eur Heart J 2010; 31: 2369–429.

3

Camm AJ, Lip GYH, De Caterina R, Savelieva I, Atar D, Hohnloser SH, Hindricks G, Kirchhof P, Bax JJ, Baumgartner H, Ceconi C, Dean V, Deaton C, Fagard R, Funck-Brentano C, Hasdai D, Hoes A, Knuuti J, Kolh P, McDonagh T, et al. 2012 focused update of the ESC Guidelines for the management of atrial fibrillation: An update of the 2010 ESC Guidelines for the management of atrial fibrillation - Developed with the special contribution of the European Heart Rhythm Association. Europace 2012; 14: 1385–413.

4

De Caterina R, Husted S, Wallentin L, Andreotti F, Arnesen H, Bachmann F, Baigent C, Huber K, Jespersen J, Kristensen SD, Lip GYH, Morais J, Rasmussen LH, Siegbahn A, Verheugt FWA, Weitz JI. Vitamin K antagonists in heart disease: current status and perspectives (Section III). Position paper of the ESC Working Group on Thrombosis--Task Force on Anticoagulants in Heart Disease. Thromb Haemost 2013; 110: 1087–107.

5

Rodriguez R a, Carrier M, Wells PS. Non-adherence to new oral anticoagulants: a reason for concern during long-term anticoagulation? J Thromb Haemost 2013; 11: 390–4.

6

Husted S, De Caterina R, Andreotti F, Arnesen H, Bachmann F, Huber K, Jespersen J, Kristensen SD, Lip GYH, Morais J, Rasmussen LH, Siegbahn A, Storey RF, Weitz JI, Esc Working Group On Thrombosis Task Force On Anticoagulants In Heart Disease OBOT. Non-vitamin K antagonist oral anticoagulants (NOACs): No longer new or novel. Thromb Haemost 2014; 111.

7

Schulman S. New oral anticoagulant agents - general features and outcomes in subsets of patients. Thromb Haemost; 2014; 111: 575–82.

This article is protected by copyright. All rights reserved.

Accepted Article

8

Ten Cate H. Monitoring new oral anticoagulants, managing thrombosis, or both? Thromb Haemost 2012; 107: 803–5.

9

Alberts MJ, Eikelboom JW, Hankey GJ. Antithrombotic therapy for stroke prevention in non-valvular atrial fibrillation. Lancet Neurol Elsevier Ltd; 2012; 11: 1066–81.

10

Osterberg L, Blaschke T. Adherence to medication. N Engl J Med 2005; 353: 487–97.

11

Sabaté E (editor). Adherence to long-term therapies: evidence for action. WHO. Geneva: World Health Organisation; 2003. p. 1–194.

12

Bangalore S, Kamalakkannan G, Parkar S, Messerli FH. Fixed-dose combinations improve medication compliance: a meta-analysis. Am J Med 2007; 120: 713–9.

13

Srivastava K, Arora A, Kataria A, Cappelleri JC, Sadosky A, Peterson AM. Impact of reducing dosing frequency on adherence to oral therapies: a literature review and meta-analysis. Patient Prefer Adherence 2013; 7: 419–34.

14

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: 567–74.

15

Schulman S, Shortt B, Robinson M, Eikelboom JW. Adherence to anticoagulant treatment with dabigatran in a real-world setting. J Thromb Haemost 2013; 11: 1295–9.

16

Sanz G, Fuster V. Fixed-dose combination therapy and secondary cardiovascular prevention: rationale, selection of drugs and target population. Nat Clin Pract Cardiovasc Med 2009; 6: 101–10.

17

Cramer JA, Roy A, Burrell A, Fairchild CJ, Fuldeore MJ, Ollendorf DA, Wong PK. Medication compliance and persistence: terminology and definitions. Value Health 2008; 11: 44–7.

18

Yu AP, Yu YF, Nichol MB. Estimating the effect of medication adherence on health outcomes among patients with type 2 diabetes--an application of marginal structural models. Value Health 2010; 13: 1038–45.

19

Kildemoes HW, Sørensen HT, Hallas J. The Danish National Prescription Registry. Scand J Public Health 2011; 39: 38–41.

20

Lynge E, Sandegaard JL, Rebolj M. The Danish National Patient Register. Scand J Public Health 2011; 39: 30–3.

21

Pedersen CB. The Danish Civil Registration System. Scand J Public Health 2011; 39: 22–5.

22

Larsen TB, Gorst-Rasmussen A, Rasmussen LH, Skjøth F, Rosenzweig M, Lip GYH. Bleeding events among new starters and switchers to dabigatran compared with warfarin in atrial fibrillation. Am J Med 2014; 127: 650–656.e5.

This article is protected by copyright. All rights reserved.

Accepted Article

23

Andrade SE, Kahler KH, Frech F, Chan KA. Methods for evaluation of medication adherence and persistence using automated databases. Pharmacoepidemiol Drug Saf 2006; 15: 565–74; discussion 575–7.

24

Ho PM, Bryson CL, Rumsfeld JS. Medication adherence: its importance in cardiovascular outcomes. Circulation 2009; 119: 3028–35.

25

Roughead EE, Ramsay E, Priess K, Barratt J, Ryan P, Gilbert AL. Medication adherence, first episode duration, overall duration and time without therapy: the example of bisphosphonates. Pharmacoepidemiol Drug Saf 2009; 18: 69–75.

26

Tsai K, Erickson SC, Yang J, Harada AS, Solow BK, Lew HC. Adherence, persistence, and switching patterns of dabigatran etexilate. Am J Manag Care 2013; 19: e325–32.

27

Schulman S, Kearon C, Kakkar AK, Mismetti P, Schellong S, Eriksson H, Baanstra D, Schnee J, Goldhaber SZ. Dabigatran versus warfarin in the treatment of acute venous thromboembolism. N Engl J Med 2009; 361: 2342–52.

28

Schulman S. Advantages and limitations of the new anticoagulants. J Intern Med 2014; 275: 1–11.

29

January CT, Wann LS, Alpert JS, Calkins H, Cleveland JC, Cigarroa JE, Conti JB, Ellinor PT, Ezekowitz MD, Field ME, Murray KT, Sacco RL, Stevenson WG, Tchou PJ, Tracy CM, Yancy CW. 2014 AHA/ACC/HRS Guideline for the Management of Patients With Atrial Fibrillation: Executive Summary: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol 2014; .

30

Deedwania PC. New oral anticoagulants in elderly patients with atrial fibrillation. Am J Med 2013; 126: 289–96.

31

Vaughan Sarrazin MS, Cram P, Mazur A, Ward M, Reisinger HS. Patient perspectives of dabigatran: analysis of online discussion forums. Patient 2014; 7: 47–54.

32

Miller NH. Compliance with treatment regimens in chronic asymptomatic diseases. Am J Med 1997; 102: 43–9.

33

Horne R, Weinman J, Barber N. Concordance, adherence and compliance in medicine taking. Report for the National Coordinating Centre for NHS Service Delivery and Organisation (NCCSDO); 2005.

34

Clarkesmith DE, Pattison HM, Lip GYH, Lane DA. Educational intervention improves anticoagulation control in atrial fibrillation patients: the TREAT randomised trial. PLoS One 2013; 8: e74037.

35

Pamboukian SV, Nisar I, Patel S, Gu L, McLeod M, Costanzo MR, Heroux A. Factors associated with non-adherence to therapy with warfarin in a population of chronic heart failure patients. Clin Cardiol 2008; 31:30.

This article is protected by copyright. All rights reserved.

Accepted Article

36

Avgil Tsadok M, Jackevicius CA, Rahme E, Humphries KH, Behlouli H, Pilote L. Sex differences in stroke risk among older patients with recently diagnosed atrial fibrillation. J Am Med Assoc. 2012; 307: 1952-1958.

37

Dagres N, Nieuwlaat R, Vardas PE, Andresen D, Lévy S, Cobbe S, Kremastinos DT, Breithardt G, Cokkinos DV, Crijns HJ. Gender-related differences in presentation, treatment, and outcome of patients with atrial fibrillation in Europe: a report from the Euro Heart Survey on Atrial Fibrillation. J Am Coll Cardiol. 2007; 49: 572-577.

38

Rix TA, Riahi S, Overvad K, Lundbye-Christiensen S, Schmidt EB, Joensen AM. Validity of the diagnoses atrial fibrillation and atrial flutter in a Dansih patient registry. Scand Cardiovasc J 2012; 46: 149–53.

39

Avorn J. The relative cost-effectiveness of anticoagulants: obvious, except for the cost and the effectiveness. Circulation 2011; 123: 2519–21.

Figure legends Figure 1. Dabigatran therapy histories for five example patients with one year of follow-up. Solid lines indicate periods during which dabigatran would be available under a twice-daily one-pill regimen. Gap counts refer to the number of periods where dabigatran would be unavailable; PDC refers to the relative period during which dabigatran would be available i.e., proportion of days covered. Figure 2. Flowchart of study population selection process. Figure 3. Proportion of patients above a given level of proportion of days covered (PDC). Solid line represents result for main analysis, dotted line for the analysis incorporating a 3-day grace period. The dashed line marks the proportion of patients with a PDC above 80%. Figure 4. Proportion of days covered (PDC) and PDC ratios (calculated using robust Poisson regression) according to clinical subgroups during the first year of dabigatran therapy. Overall PDC was 83.9%.

This article is protected by copyright. All rights reserved.

Accepted Article

Abbreviations: CI: confidence interval; PDC: proportion of days covered; CHA2DS2-VASc: Cardiac failure or dysfunction, Hypertension, Age≥75 [Doubled], Diabetes, Stroke [Doubled]-Vascular disease, Age 65-74, and Sex category [female] stroke risk score, previously described[22]; SE: systemic embolic event; TIA: transient ischemic attack. * PDC incorporating a 3-day grace period. † Prescription purchases in the code C section of the Anatomical Therapeutic Chemical Classification System. Figure 5. Gap rates and gap rate ratios (calculated using robust Poisson regression) according to clinical subgroups during the first year of dabigatran therapy (assuming a one pill twice daily regimen) Adjusted for PDC (categorical: 0-50%, 50%-60%, 60%-70%, 70%-80%, 80%-90%, 90%-100%). Overall gap rate was 1.4. Abbreviations: CI: confidence interval;CHA2DS2-VASc: Cardiac failure or dysfunction, Hypertension,

Age≥75 [Doubled], Diabetes, Stroke [Doubled]-Vascular disease, Age 65-74, and Sex category [female] stroke risk score, previously described[22]; SE: systemic embolic event; TIA: transient ischemic attack. * Gap rate incorporating a 3-day grace period. † Prescription purchases in the code C section of the Anatomical Therapeutic Chemical Classification System. Figure 6. Dabigatran therapy restart rates (smooth line) and Kaplan-Meier restart probability (dashed line) during gap periods occurring in the first year of dabigatran therapy.

This article is protected by copyright. All rights reserved.

Accepted Article

Table 1. Baseline demographic and clinical characteristics. Number of patients, N

2,960

Female sex, % (N)

45.8 (1356)

Age, years, mean (SD)

72.1 (10.8)

Age, years, % (N) ≤54

6.4 (189)

55-64

17.2 (510)

65-74

34.9 (1034)

≥75

41.5 (1227)

Initial dose 150 mg, % (N)

61.1 (1812)

CHA2DS2-VASc score, mean (SD)

2.8 (1.6)

CHA2DS2-VASc score, % (N) 0

7.0 (206)

1

14.2 (419)

≥2

78.9 (2335)

History of bleeding, % (N)

11.7 (345)

History of stroke/SE/TIA, % (N)

18.6 (550)

Cardiovascular prescriptions, past year*, % (N)

68.5 (2028)

Hospitalisation, past year, % (N)

80.8 (2392)

This article is protected by copyright. All rights reserved.

Accepted Article

Abbreviations: SD: standard deviation; CHA2DS2-VASc: Cardiac failure or dysfunction, Hypertension, Age≥75 [Doubled], Diabetes, Stroke [Doubled]-Vascular disease, Age 65-74, and Sex category [female] stroke risk score, previously described[22]; SE: systemic embolic event; TIA: transient ischemic attack. *Prescription purchases in the code C section of the Anatomical Therapeutic Chemical Classification System.

Table 2. Cox model restart rate ratios, stratified by time since gap period start. Restart rate ratio (95% confidence interval)

Female sex

Rapid restart

Intermediate restart

Late restart

75 (vs. 55-64)

1.24 (1.06-1.45)

1.45 (1.22-1.72)

1.40 (0.88-2.22)

0.90 (0.81-1.01)

0.81 (0.72-0.91)

0.83 (0.59-1.17)

0 (vs. 1)

0.62 (0.47-0.81)

0.65 (0.49-0.88)

0.49 (0.25-0.94)

≥2 (vs. 1)

1.12 (0.95-1.31)

1.39 (1.16-1.68)

1.46 (0.95-2.23)

Age, years

Original dose 150mg CHA2DS2-VASc score

This article is protected by copyright. All rights reserved.

0.98 (0.83-1.17)

0.98 (0.81-1.19)

1.49 (0.96-2.33)

1.14 (0.99-1.32)

1.34 (1.15-1.57)

0.97 (0.59-1.59)

year*

1.14 (1.01-1.28)

1.29 (1.14-1.47)

1.32 (0.94-1.84)

Hospitalization, past year

0.91 (0.80-1.04)

1.00 (0.86-1.17)

1.10 (0.74-1.64)

2nd (vs. 1st)

1.05 (0.95-1.17)

1.88 (1.63-2.16)

2.23 (1.51-3.30)

3rd or later (vs. 1st)

0.86 (0.76-0.96)

1.97 (1.73-2.23)

4.12 (2.71-6.27)

Accepted Article

History of bleeding History of stroke/SE/TIA Cardiovascular prescriptions, past

Gap period index

Abbreviations: CHA2DS2-VASc: Cardiac failure or dysfunction, Hypertension, Age≥75 [Doubled], Diabetes,

Stroke [Doubled]-Vascular disease, Age 65-74, and Sex category [female] stroke risk score, previously described[22]; SE: systemic embolic event; TIA: transient ischemic attack.* Prescription purchases in the code C section of the Anatomical Therapeutic Chemical Classification System

This article is protected by copyright. All rights reserved.

Accepted Article This article is protected by copyright. All rights reserved.

Accepted Article This article is protected by copyright. All rights reserved.

Accepted Article This article is protected by copyright. All rights reserved.

Accepted Article This article is protected by copyright. All rights reserved.

Accepted Article This article is protected by copyright. All rights reserved.

Dabigatran adherence in atrial fibrillation patients during the first year after diagnosis: a nationwide cohort study.

There is a perception among physicians that lack of routine monitoring with non-vitamin K antagonist oral anticoagulants (NOACs) may lead to poor adhe...
511KB Sizes 4 Downloads 8 Views