• Center on Health Equity & Access
  • Clinical
  • Health Care Cost
  • Health Care Delivery
  • Insurance
  • Policy
  • Technology
  • Value-Based Care

Adherence, Persistence, and Switching Patterns of Dabigatran Etexilate

Publication
Article
The American Journal of Managed CareSeptember 2013
Volume 19
Issue 9

We present a descriptive analysis utilizing pharmacy claims from a managed care population to quantify adherence, persistence, and switching patterns for patients initiating dabigatran.

Objective:

To investigate adherence, persistence, and switching in patients initiating dabigatran.

Study Design:

Descriptive analysis using pharmacy claims databases.

Methods:

Patients with a claim for dabigatran and who were continuously enrolled in pharmacy benefits for 180 days prior to and 180 days following the initiation of dabigatran were identified and stratified by whether there was a history of warfarin treatment prior to dabigatran initiation. Medication adherence was calculated as the proportion of days covered (PDC). Persistence to treatment at 180 days was measured. Among patients who discontinued dabigatran, time to initiating warfarin was determined.

Results:

In the overall population, 39.9% of 17,691 patients were nonpersistent to dabigatran. The PDC for the warfarin-naïve cohort was 0.674 (standard deviation [SD] 0.364), and 0.712 (SD 0.354) for the warfarin-experienced cohort. For patients persistent to dabigatran, the PDCs for warfarin-naïve and warfarin-experienced cohorts were 0.935 (SD 0.075) and 0.937 (SD 0.074), respectively. In patients discontinuing dabigatran, 16.1% of warfarin-naïve and 41.1% of warfarinexperienced patients initiated warfarin. Among patients discontinuing dabigatran, the mean time to discontinuation in warfarin-naïve and warfarinexperienced cohorts, respectively, was 59.8 (SD 36.2) and 59.6 (SD 36.2) days. The mean time to initiating warfarin in warfarin-naïve and warfarin experienced cohorts, respectively, was 62.5 (SD 47.2) and 60.5 (SD 43.0) days.

Conclusions: Two in 5 patients discontinued dabigatran therapy within 6 months, and the majority of these patients were not anticoagulated with warfarin upon discontinuation. These findings highlight potential gaps in the care of patients treated with dabigatran in routine practice.

Am J Manag Care. 2013;19(9):e325--e332Dabigatran has been shown in trials to reduce the risk of stroke and systemic embolism in patients with non-valvular atrial fibrillation. However, patterns of dabigatran use inroutine practice remain unknown. The objective of the study is to investigate adherence, persistence, and switching to warfarin upon dabigatran discontinuation, in patients initiating dabigatran therapy.

  • Findings may highlight gaps in the care of patients treated with dabigatran.

  • Due to the dependence on adherence and persistence to dabigatran for prevention of stroke and thromboembolic events, there may be a need for healthcare professionals to provide additional patient support with this medication.

Dabigatran, an oral direct thrombin inhibitor, was approved by the US Food and Drug Administration (FDA) in October 2010 to reduce the risk of stroke and systemic embolism in patients with non-valvular atrial fibrillation. It was the first oral anticoagulant to be introduced in the over 50 years since warfarin was approved. Dabigatran does not share warfarin’s potential limitations of having a variable response, extensive food and drug interactions, lab monitoring requirements, and narrow therapeutic index concerns. However, dabigatran is dosed twice daily rather than once daily for warfarin, and is likely to have an increased prescription cost burden to patients and payers.

Adherence is defined as the extent to which a patient follows directions for a prescribed medication, whereas persistence is defined as the act of continuing the treatment for the prescribed duration.1 In the setting of prescription claims analyses, drug adherence and persistence could be markers of drug efficacy and tolerability, and therapy switches may indicate unsatisfactory treatment response and/or unacceptable adverse effects.2 Understanding patient adherence and persistence to anticoagulation therapy in routine practice is important, since nonadherence and nonpersistence to anticoagulation therapy could potentially result in subtherapeutic responses and increased riskof thromboembolic events. Although warfarin has a long istory as a standard of care in anticoagulation, warfarin nonadherence continues to be a substantial problem that has been linked to risk of ischemic stroke. In a prospective cohort study of adults initiating warfarin in anticoagulation clinics, patients exhibited nonadherence on 21% of the measured patient-days.3 Another study demonstrated a correlation between warfarin nonadherence and the risk for ischemic stroke.4 In a recent retrospective medical claims analysis of 14,149 Medicare patients with a diagnosis of non-valvular atrial fibrillation, adherence to warfarinwas measured using proportion of days covered (PDC). the risk for ischemic stroke was significantly lower in those with better adherence to warfarin, with a PDC >80%, compared with those with a lower adherence to warfarin, with a PDC <0% (adjusted odds ratio, 0.59; 95% confidence interval, 0.48-0.72; P <.001).

There have been few findings regarding the use of dabigatran outside clinical trials. One study, which used medical and pharmacy claims, assessed demographic and clinical characteristics, prescriber specialty, and persistence to dabigatran therapy over a 4-month period.5 Results showed that 17% of all patients (n = 1143) were nonpersistent over a 4-month follow-up period, and 71% of these patients had a history of warfarin use in the 6 months prior. However, these results may be limited because of the short identification period of 2 months immediately following FDA approval. In a New Zealand study of inpatients starting dabigatran, the authors found that over a follow-up period of 1 to 7 months, 24% of all patients discontinued, with a median treatment duration of 140.5 days.6 Approximately 10% of the overall population discontinued due to the side effects of dabigatran (eg, gastrointestinal side effects, bleeding, major adverse cardiovascular events). An additional 8% of patients stopped dabigatran due to planned discontinuation of therapy after electrical cardioversion, adverse media coverage, or medication adherence issues. Results of this study may be limited by a smaller study size of 70 patients. Therefore, there is a need to study dabigatran in a larger population over a longer time period to further explore the real-world use of dabigatran. The objective of our study is to investigate adherence and persistence to dabigatran therapy, and switching to warfarin upon dabigatran discontinuation, in patients initiating dabigatran therapy.

METHODSStudy Design and Identification Periods

This was a descriptive, retrospective, observational analysis using administrative pharmacy claims from a large pharmacy benefits manager (PBM). The study was conducted using prescription claims from Medicare Part D Prescription Drug Plan (PDP), Medicare Advantage Prescription Drug (MAPD), commercial, and Medicaid plan patients. Patients were identified by a first fill (index fill) of dabigatran during the identification period of October 29, 2010, through June 30, 2011. The pre-index period was defined as 180 days prior to the index date, and the post-index period was defined as 180 days after the index date. The inclusion criteria were that patients had at least 1 fill of dabigatran during the identification period, and that they were continuously eligible for pharmacy benefits during the pre-index and postindex periods.

Study Cohorts

Patients were stratified by presence of a warfarin claim (“warfarin-experienced”) or absence of a warfarin claim (“warfarin-naïve”) in the pre-index period to analyze the cohorts separately. It was hypothesized that there could potentially be differences in behavior between the 2 cohorts, based on previous use of anticoagulants.

Study Measures

Appendix

The study measures included a description of various demographic characteristics (ie, age, gender, plan type, and geographical region) and clinical characteristics (ie, frequency of chronic heart failure [CHF], diabetes mellitus [DM], hypertension [HTN], coronary heart disease [CHD], and dyslipidemia, Chronic Disease Score [CDS],7 and number of concomitant medications, measured by the number of prescriptions with days supply on the index date). The presence of concomitant diseases was detected by proxy of specific medication classes in the pre-index period (details can be found in the ; www.ajmc.com).

The primary outcomes measures included persistence, as defined by continuous use of dabigatran with no days supply gap of 30 days or more during the post-index period, and adherence, as described by the PDC calculated as the number of calendar days that patients had a supply of dabigatran during the post-index period divided by 180 days. In patients nonpersistent to dabigatran therapy, time to discontinuation (number of days from the index date to the last day of supply prior to a gap in therapy of 30 days or greater), incidence of switching to warfarin (number of patients who had a claim for warfarin in the post-index period), and time to switch to warfarin (number of days from the index date to the first fill of warfarin in the post-index period) were measured.

Statistical Methods

All statistical tests were performed using SAS version 9.2. Means were tested for variance and compared by independent t tests, and proportions were compared using the χ2 test. All comparisons were 2-sided and performed at a 0.05 level of significance.

RESULTS

Table 1

There were 7322 patients identified for the warfarin-naïve cohort, and 10,369 patients identified for the warfarin-experienced cohort (). The average age of the study population was 76 years, and the majority of patients were enrolled in a PDP plan (77.2% for the warfarin-naïve cohort and 80.6% for the warfarin-experienced cohort, P <.001). Chronic concomitant conditions included hypertension (88.9% and 92.6% of the warfarin-naïve and warfarin-experienced cohorts, respectively, P <.001), CHD (69.8% and 71.9% of the warfarin-naïve and warfarin-experienced cohorts, respectively, P = .002), and dyslipidemia (58.2% and 63.0% of the warfarin-naïve and warfarin-experienced cohorts, respectively, P <.001). Warfarin-naïve patients had a mean of 6.4 total prescriptions (including dabigatran), and warfarin-experienced patients had a mean of 7.5 total prescriptions (including dabigatran) with days supply on the index date (P <.001). The mean Chronic Disease Scores were 4.2 and 6.4 (P <.001) for tively. While other statistically significant differences were observed for characteristics between the cohorts, a number of the statistically significant differences may be driven by the large population size and may not be of clinical significance.

Table 2

The number of patients persistent to dabigatran therapy was 4137 (56.5%) in the warfarin-naïve cohort and 6492 (62.6%) in the warfarin-experienced cohort (P <.001) (). Of all 17,691 patients, 7062 (39.9%) were nonpersistent to dabigatran therapy over the 6-month period. Adherence was determined using PDC for the overall population and also for patients persistent to dabigatran therapy. The mean PDC for patients persistent to dabigatran therapy was 0.935 (standard deviation [SD] 0.075) for the warfarin-naïve cohort and 0.937 (SD 0.074) for the warfarin-experienced cohort (P = .28). In the overall population, the mean PDC was 0.674 (SD 0.364) for the warfarin-naïve cohort and 0.712 (SD 0.354) for the warfarin-experienced cohort (P <.001).

Table 3

Figure 1

Figure 2

Among patients discontinuing dabigatran therapy, the mean time to discontinuation in warfarin-naïve and warfarinexperienced cohorts, respectively, was 59.8 (SD 36.2) days the warfarin-naïve and warfarin-experienced cohorts, respectively, was 59.8 (SD 36.2) days and 59.6 (SD 36.2) days (P = .83), with approximately 50% of patients discontinuing after the first 30-day fill of dabigatran. The mean time to initiating warfarin in warfarin-naïve and warfarin-experienced cohorts, respectively, was 62.5 (SD 47.2) and 60.5 (SD 43.0) days (P = .35) (, , ). The number of patients nonpersistent to dabigatran therapy who switched to warfarin was 513 (16.1%) in the warfarin-naïve cohort and 1595 (41.1%) in the warfarinexperienced cohort (P <.001). The Kaplan-Meier curves for the time to discontinuation and time to switch to warfarin are shown in Figures 1 and 2.

DISCUSSION

Patients with a history of warfarin use prior to initiating dabigatran therapy had a greater burden of comorbidity than those without prior warfarin use, as demonstrated by a greater prevalence of CHF, DM, HTN, CHD, dyslipidemia, a higher CDS score, and a greater total number of medications compared with warfarin-naïve patients. Other statistically significant differences between the cohorts were likely driven by the large size of the study population. Examples of variables with small, but statistically significant, differences include mean age (76.0 years in warfarin-naïve versus 76.7 years in warfarin-experienced patients, P <.001), patients with concomitant diabetes mellitus (21.2% in warfarin-naïve vs 22.5% in warfarin-experienced patients, P = .03), and the percentage of patients with PDP coverage (77.2% in warfarin-naïve vs 80.6% in warfarin-experienced patients, P <.001). Among patients persistent to dabigatran therapy, both cohorts demonstrated good adherence with mean PDC values of 0.935 (warfarin-naïve) and 0.937 (warfarin-experienced). The suboptimal adherence of the overall cohort (including both patients persistent and nonpersistent to dabigatran therapy) appears to be attributed to nonpersistence. In the overall study cohort, 40% of patients were nonpersistent to dabigatran therapy (ie, discontinued dabigatran therapy). In patients nonpersistent to dabigatran therapy, the mean trial of dabigatran (ie, time to discontinuation) was a length of 60 days.

Among patients who were nonpersistent to dabigatran therapy, approximately 50% of patients discontinued by 30 days and approximately 80% had discontinued by 90 days. The observed decline in persistence at these points likely reflects discontinuation after only the first fill (either a 30- or 90-day supply). Of patients switching to warfarin, the mean time to switch was also approximately 60 days. In patients nonpersistent to dabigatran therapy, the majority did not switch to warfarin in either cohort. This suggests that most patients may discontinue dabigatran without being switched to warfarin, despite the need for patients with atrial fibrillation to be persistent to chronic anticoagulation therapy. It is possible that patients may have discontinued dabigatran therapy at their own discretion without informing their healthcare providers. If this is the case, then patients taking dabigatran still may require closer monitoring, in spite of the lack of need for laboratory monitoring with dabigatran therapy. It may be necessary for healthcare professionals and managed care organizations to provide additional support and counseling for patients on dabigatran therapy.

The results of this study should be interpreted with caution, as it was not possible to verify reasons for dabigatran discontinuation. Potential reasons may include adverse event occurrence (eg, a bleeding episode or dyspepsia as was observed in the RE-LY trial8), a switch to aspirin or rivaroxaban, or use of dabigatran for an off-label indication with a shorter duration of treatment. In terms of adverse events, in the RE-LY trial,8 reasons for discontinuation that were reported included gastrointestinal bleeding, gastrointestinal symptoms, a serious adverse event, an outcome event, and patient’s decision. (For reference, in the RE-LY trial,8 gastrointestinal bleeding was cited as the reason for discontinuation in 1.3% of patients taking dabigatran 150 mg vs 0.9% of patients who were taking warfarin [P value not reported]. Gastrointestinal side effects led to discontinuation in 2.1% of patients taking 150 mg dabigatran vs 0.6% of patients who were taking warfarin [P value not reported].) Rivaroxaban was initially approved in the United States on July 1, 2011, for the prevention of deep vein thrombosis (DVT) in patients undergoing knee or hip replacement surgery. According to the time frame of this study (with the latest possible day of the post-index period as December 31, 2011), some patients could have switched from dabigatran to rivaroxaban, accounting for some of the nonpersistence. However, this is less likely, as a very low incidence of switching from dabigatran to rivaroxaban was observed in a prescription claims study examining switch rates among oral anticoagulants (warfarin, dabigatran, and rivaroxaban).9 The study found that only 0.02% of switchers changed from dabigatran to rivaroxaban, while 87.3% of switchers changed from warfarin to dabigatran. Lastly, it is possible that some discontinuation occurred appropriately and may be accounted for by patients who were treated with dabigatran, off label, for a fi nite duration (eg, DVT prophylaxis in patient undergoing knee or hip replacement surgery). However, it is less likely that this accounts for the majority of discontinuations observed in this study, for several reasons. First, dabigatran required Prior Authorization (PA) approval for benefit coverage in this population. A review of PA cases among Part D patients during a 4-month period found that less than 1% had off-label indications for finite duration of dabigatran therapy, and the majority of indications were for atrial fi brillation,requiring chronic dabigatran therapy. Secondly, the findings of an observational study reported that during the identification period of our study (October 29, 2010, through June 30, 2011), the majority of patients in the United States were using dabigatran for the atrial fi brillation indication.10 The research utilized the IMS Health National Disease and Therapeutic Index, a nationally representative audit of office-based providers, to quantify patterns of oral anticoagulant use for the prevention of thromboembolism. The authors found that dabigatran has been rapidly adopted into ambulatory practice in the United States, primarily for the treatment of atrial fi brillation, but increasingly is being used for off-label indications. They found that the proportion of US patients who used dabigatran for the indication of atrial fi brillation was 92% in the last quarter of 2010 and 63% in the last quarter of 2011. Thus, we believe that the majority of discontinuations observed in our study are due to true nonpersistence, rather than appropriate discontinuations for off-label indications. A follow-up study, with pharmacy and medical claims, could offer explanations for nonpersistence (eg, capturing a medical claim with the ICD-9-CM diagnosis code for hemorrhage of gastrointestinal tract may indicate that a patient discontinued because of a gastrointestinal bleed).

Limitations of this prescription claims database study include lack of information regarding diagnoses and medical claims. Comorbidities were inferred from pharmacy claims, by assuming that taking certain medications indicates the presence of a particular disease state. (Validation of the proxy measures was done by reviewing published studies and referencing an internal process performing similar identifications of disease states through pharmacy claims.) Secondly, measures of adherence and persistence were based on claims data, which assumes that the patient is taking the medication rather than stockpiling it. It is not possible in a claims analysis to determine whether patients discontinued at their own discretion or upon the direction of their healthcare provider, nor is it possible to determine reasons behind discontinuations of therapy. Having this information could aid in the design of potential specific interventions to increase adherence and persistence to therapy. Lastly, claims data do not capture plan exclusion OTC utilization or other patient out-of-pocket purchases, which could contribute to a more comprehensive understanding of patient practices.

CONCLUSIONS

Two in 5 patients discontinued dabigatran therapy within 6 months in this observational, retrospective analysis using administrative pharmacy claims from a large PBM. The majority of patients discontinuing dabigatran therapy were not anticoagulated with warfarin during the post-index period. These findings highlight potential gaps in the care of patients treated with dabigatran therapy in routine practice. Patients should be educated to report intolerance, adverse events, or cost burden to their healthcare provider, instead of discontinuing therapy on their own. Due to the dependence on good adherence and persistence to dabigatran therapy for prevention of stroke and thromboembolic events, there may be a need for healthcare professionals and managed care organizations to provide additional patient support and counseling on this medication. Further research is needed to explain the reasons for nonpersistence to dabigatran therapy.Author Affiliations: From OptumRx (KT, SCE, JY, ASMH, BKS, HCL), Irvine, CA.

Funding Source: None.

Author Disclosures: The authors (KT, SCE, JY, ASMH, BKS, HCL) report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (KT, SCE, BKS, HCL); acquisition of data (KT, JY); analysis and interpretation of data (KT, SCE, JY, ASMH, BKS); drafting of the manuscript (KT, SCE, ASMH); critical revision of the manuscript for important intellectual content (KT, SCE, ASMH); statistical analysis (KT, JY); provision of study materials or patients (KT); administrative, technical, or logistic support (ASMH, BKS, HCL); and supervision (SCE, BKS, HCL).

Address correspondence to: Kimberly Tsai, PharmD, 2300 Main St, CA 134-0404, Irvine, CA 92614. E-mail: kimberly.tsai@optum.com.

1. Cramer J, Roy A, Burrell A, et al. Medication compliance and persistence: terminology and definitions. Value Health. 2008;11(1):44-47.

2. Vegter S, Nguyen NH, Visser ST, et al. Compliance, persistence, and switching patterns for ACE inhibitors and ARBs. Am J Manag Care. 2011;17(9):609-616.

3. Platt AB, Localio AR, Brensinger CM, et al. Risk factors for nonadherence to warfarin: results from the IN-RANGE study. Pharmacoepidemiol Drug Saf. 2008;17(9):853-860.

4. Fitch K, Broulette J, Pyenson BS, Iwasaki K, Kwong WJ. Utilization of anticoagulation therapy in Medicare patients with nonvalvular atrial fibrillation. Am Health Drug Benefits. 2012;5(3):157-168.

5. Stanek E, Agatep BC, Herrera V, et al. Assessment of initial U.S. dabigatran utilization and short-term persistence [abstract]. Circulation.2011;124:A16675.

6. Michel J, Mundell D, Boga T, Sasse A. Dabigatran for anticoagulation in atrial fibrillation — early clinical experience in a hospital population and comparison to trial data. Heart Lung Circ. 2013;22(1):50-55.

7. Von Korff M, Wagner E, Saunders K. A chronic disease score from automated pharmacy data. J Clin Epidemiol. 1992;45(2):197-203.

8. Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med. 2009;361(12): 1139-1151.

9. Theodorou AA, Palmieri A, Szychowski JA, Sehman ML, Swarna V. Prescription utilization of the oral anticoagulants. Am J Pharm Benefits. 2012;4(3):120-123.

10. Kirley K, Qato DM, Kornfield R, Stafford SS, Alexander GC. National trends in oral anticoagulant use in the United States, 2007 to 2011. Circ Cardiovasc Qual Outcomes. 2012;5(5)615-621.

Related Videos
Wayne Jonas, MD
Wayne Jonas, MD
Wayne Jonas, MD
Wayne Jonas, MD
Wayne Jonas, MD
Dr Wayne Jonas
Related Content
© 2024 MJH Life Sciences
AJMC®
All rights reserved.