Unauthorized Use Prohibited Unauthorized Use

Adv Ther (2009) 26(8):. DOI 10.1007/s12325-009-0054-1

ORIGINAL RESEARCH

National Assessment of Medicare Prescription Plan Coverage Gaps Among Patients with Atrial Fibrillation in the US Michael H. Kim · Jay Lin · Charles Kreilick

Unauthorized Use Prohibited

Received: June 22, 2009 / Published online: / Printed: © Springer Healthcare Communications 2009

ABSTRACT Introduction: The Medicare Part D prescription plan has a coverage gap from $2250 to $5100 per year (2006 thresholds) in which patients pay full drug costs (the “donut hole”). We examined prescription costs in patients aged ≥65 years with atrial fibrillation (AF), and the proportions entering/ exiting the donut hole in 2006. Methods: This retrospective cohort study used claims data from the Integrated Healthcare Information Systems National Managed Care Benchmark Database from January 2005 to December 2006. Medicare Advantage patients aged ≥65 years who were hospitalized with a primary/secondary discharge diagnosis of AF, or had ≥2 outpatient AF claims without hospitalization in 2005, were identified. Total 2006 prescription costs were calculated

by summing health plan payments and patient copayments. Results: Of 16,655 patients included in the analysis, 1527 were hospitalized with a primary AF diagnosis, 6067 with a secondary AF diagnosis, and 9061 had outpatient-managed AF in 2005. Mean 2006 per-patient prescription costs were $3457.16. In total, 58.8% of patients reached the $2250 threshold in a mean of 199 days; including 59.9% of primary AF hospitalization, 63.5% of secondary AF hospitalization, and 55.5% of patients with outpatient-managed AF. Of the overall population, 21.2% reached the $5100 threshold in a mean of 257 days; including 21.3% of primary AF hospitalization, 27.3% of secondary AF hospitalization, and 17.2% of patients with outpatient-managed AF. AF-related drugs accounted for only 15% of the total prescription costs, with the majority of costs relating to the treatment of non-AF comorbidities. Conclusions: In 2006, total annual prescription costs in this population of Medicare Advantage AF patients were considerable. In this sample population, over half of patients would have had prescription costs in excess of the 2006 Part D donut hole threshold. The costs of AF treatment itself were relatively low, with the majority of prescription costs relating to the treatment of non-AF comorbidities.


Michael H. Kim () Associate Professor, Division of Cardiology, Northwestern University, Feinberg School of Medicine, 251 E Huron St, Feinberg Pavilion, Suite 8-503, Chicago, IL 60611, USA. Email: [email protected] Jay Lin sanofi-aventis, Bridgewater, New Jersey, USA Charles Kreilick ProUnlimited, Boca Raton, Florida, USA

2

Keywords: atrial fibrillation; comorbidities; coverage gap; donut hole; Medicare Part D; prescription costs

INTRODUCTION Atrial fibrillation (AF) is the most common cardiac dysrhythmia, affecting approximately 2.4 million individuals in the US.1 In addition to a reduced quality of life,2 patients with AF frequently have cardiovascular comorbidities and have an elevated risk of cardiovascular events and mortality compared with individuals without the condition.3-5 As a result, hospitalization rates among patients with AF are high, with approximately 460,000 hospital discharges per year in the US with a diagnosis of AF.6 In order to reduce hospitalizations and mortality and improve the quality of life in patients with AF, effective treatment strategies need to include therapies for AF, common comorbidities (eg, heart failure, hypertension, and diabetes), and the prevention of stroke or other cardiovascular events.4,5 Thus, the prescription burden among patients with AF is likely to be considerable and to result in substantial costs to both the patient and health plans. More than three-quarters of patients with AF are over the age of 65 years1,7 and are thus eligible for healthcare and prescription coverage through Medicare, a US governmental insurance program. Overall, approximately 6% of all Medicare enrollees have AF.8 In 2006, the Medicare Part D drug benefit plan was introduced, arising from the Medicare Prescription Drug Improvement and Modernization Act of 2003. The introduction of the Part D plan was intended to improve the quality of healthcare for older individuals, offering Medicare enrollees the opportunity to purchase subsidized prescription coverage, including additional subsidization for those patients with the highest prescription costs.9 The standard Part D

Adv Ther (2009) 26(8):.

plan is structured such that, following an initial deductible of $250, the plan covers 75% of annual prescription costs up to a defined threshold ($2250 in 2006). Patients are then required to pay full drug costs up to a second threshold ($5100 in 2006), after which they qualify for catastrophic coverage and pay only 5% of subsequent prescription costs (Figure 1).9-11 Based on these criteria, a patient reaching but not exceeding the threshold of $2250 would pay only $750 in prescription costs ($250 deductible, $500 coinsurance), while a patient reaching the catastrophic coverage threshold of $5100 would pay at least $3600 ($250 deductible, $500 coinsurance, plus $2850 uncovered costs during the coverage gap). The coverage gap, which has been termed the “donut hole”, is thus likely to have a significant impact on patients’ overall cost burden, and potentially influence their medication utilization.11‑13 A previous substudy in 472 patients with AF aged ≥65 years estimated that 27%-46% of patients (depending on the cost assumptions used) reached the donut hole threshold in 2006, and would thus have incurred full prescription costs for the remainder of the year (or until qualifying for catastrophic coverage), if enrolled in the Medicare Part D prescription plan. 14 However, to the best of our knowledge, no large



Figure 1. Structure of the Medicare Part D Standard Drug Benefit Plan in 2006. $250

$2250

$5100

Donut hole

Patient pays 100% (deductible)

Catastrophic coverage

Patient pays Patient pays Patient pays Cumulative prescription 5% 25% 100% (coinsurance/ costs (coinsurance/ copayments) copayments)

Adv Ther (2009) 26(8):.

study has provided a comprehensive evaluation of Medicare-eligible patients with AF across the US who would potentially have increased outof-pocket costs as a result of the donut hole. We therefore examined cumulative prescription costs due to AF treatments and all other treatments in a sample population of Medicare Advantage patients with AF in US-managed care organizations, and the proportions reaching the donut hole and catastrophic coverage thresholds, in 2006. To give an indication of how prescription costs may differ in hospitalized patients compared with those treated as outpatients, costs were evaluated separately for patients hospitalized with a primary discharge diagnosis of AF, those hospitalized with a secondary diagnosis of AF, and patients with outpatient-managed AF.

3

METHODS

Ninth Revision [ICD-9]-CM diagnosis code 427.31). Patients were required to have both medical and pharmacy benefits data available and to be continuously enrolled in the database from January 1, 2005 to December 31, 2006. Patients meeting these criteria, who would be enrolled in the Medicare Advantage plan, were used as a surrogate for the general Medicare Part D population. Three patient populations were defined: (1) primary AF diagnosis hospitalization patients, defined as those hospitalized in 2005 with a primary discharge diagnosis of AF; (2) secondary AF diagnosis hospitalization patients, defined as those hospitalized in 2005 with a secondary diagnosis of AF; (3) patients with outpatientmanaged AF, defined as those with ≥2 outpatient medical claims with an AF diagnosis in 2005, without a hospitalization during this period.

Study Design

Data Collection and Analyses

This retrospective cohort study used data from the Integrated Healthcare Information Systems National Managed Care Benchmark Database drawn from between January 1, 2005 and December 31, 2006. The database is derived from over 30 health plans, covering seven census regions, and includes approximately 25 million individuals. Data from this database have been used previously for other health outcomes and health economics studies.15-18 In accordance with the Health Information Portability and Accountability Act (HIPAA) of 1996 all patient data were de-identified.

Cumulative prescription costs in 2006 US$ were calculated by summing the recorded health plan payment and the patient copayment. Outcome measures included the proportion of patients reaching the donut hole ($2250) and catastrophic coverage ($5100) thresholds in 2006, and the time to reach the donut hole and catastrophic coverage thresholds. All statistical analyses were conducted using SAS Release 9.1 (SAS Institute Inc., Cary, NC, USA).


Unauthorized Use Prohibited RESULTS

Study Population

Study Population and Baseline Clinical Characteristics

Health insurance claims between January 1, 2005 and December 31, 2005 were screened to identify patients aged ≥65 years with an AF diagnosis (International Classification of Diseases,

In total, 16,655 patients with a diagnosis of AF were included in the analysis; 1527 were hospitalized with a primary discharge diagnosis of AF, 6067 with a secondary AF diagnosis,

4

and 9061 had outpatient-managed AF (Table 1). In the overall population, the mean age was 74 years and 57% of patients were men. Over half of patients (60.2%) were from the northeast of the US (Table 2). A high proportion of patients across the three groups had comorbidities: 31.6% had heart failure, 25.7% had diabetes, 19.4% had stroke/transient ischemic attack/cerebrovascular disease, and 11.9% had coronary artery disease. In general, a lower proportion of patients with outpatient-managed AF had comorbidities than patients with primary or secondary AF hospitalizations (Table 2). Across the three patient groups, 74% of patients with AF received rate-control agents, 69% received anticoagulants, and 21% received antiarrhythmics. The proportion of patients prescribed antiarrhythmics was highest in patients hospitalized with a primary diagnosis of AF (39% vs. 19% across the other two groups).

Adv Ther (2009) 26(8):.

(eg, carvedilol); angiotensin-converting enzyme inhibitors/angiotensin receptor blockers (eg, lisinopril, ramipril, valsartan); other antihypertensives (eg, amlodipine); antiplatelet drugs (eg, clopidogrel); antidiabetic drugs (eg, rosiglitazone, pioglitazone); chronic obstructive pulmonary disease treatments (eg, fluticasonesalmeterol, albuterol-ipratropium); and Alzheimer’s treatments (eg, donepezil). Overall, the highest mean per-patient costs for individual drugs (averaged across the whole population) were for atorvastatin ($185-$221), simvastatin ($88-$99), and clopidogrel ($48-$126).


Cumulative Prescription Costs In 2006, each patient filled an average of 46 prescriptions (764,817 in 16,655 patients). In total, 14.4% of all prescriptions were for ratecontrol drugs, 10.0% were for anticoagulants, and 2.6% were for rhythm-control drugs (Table 3). Mean total prescription costs were $3457.16 per patient. Rhythm-control drugs accounted for 4% of mean total per-patient prescription costs ($146.09 of $3457.16), rate-control drugs for 6% ($217.97 of $3457.16), and anticoagulants for 5% ($177.41 of $3457.16) (Table 3); thus, collectively, AF-related drugs accounted for only 15% of total prescription costs. Consistent with the comorbidity profile and age of the study population, among non-AFrelated drugs, the treatments accounting for the greatest proportions of total prescription costs included: lipid-lowering drugs (eg, atorvastatin, simvastatin, ezetimibe); beta-blockers

Proportion of Patients Reaching the Donut Hole and Catastrophic Coverage Thresholds In 2006, 58.8% of patients with AF reached the $2250 threshold, including 59.9% of patients with primary AF hospitalizations, 63.5% of patients with secondary AF hospitalizations, and 55.5% of patients with outpatient-managed AF (Figure 2). Of patients entering the donut hole, a high proportion had prescription costs close to the donut hole threshold, with declining numbers of patients with costs approaching the catastrophic coverage threshold (Figure 3). In total, 21.2% of all patients (3537 of 16,655 patients) reached the $5100 threshold in 2006, including 21.3% of patients with primary AF hospitalizations, 27.3% of patients with secondary AF hospitalizations, and 17.2% of patients with outpatient-managed AF (Figure 2).

Unauthorized Use Prohibited Time to Reach the Donut Hole and Catastrophic Coverage Thresholds

Among patients reaching the $2250 threshold in 2006, the mean±SD time to reach the threshold was 199±83 (median 193, range 2-364) days for the overall population; 197±80 (median 194, range 12-363) days for patients with primary

Adv Ther (2009) 26(8):.

5

Table 1. Patient identification. All patients with AF

Primary AF Secondary AF Outpatienthospitalization hospitalization managed AF

Total number of patients aged ≥65 years Exclusion criteria

112,842

7008

34,197

71,637

Not continuously enrolled with medical benefit

54,441

3543

18,079

32,819

Not continuously enrolled with pharmacy benefit

34,206

1938

9681

22,587


Outpatients hospitalized for AF in 2005

Secondary AF patients hospitalized with a primary diagnosis of AF in 2005 Patients included in the analysis

7170

NA

NA

7170

370

NA

370

NA

16,655

1527

6067

9061

AF=atrial fibrillation; NA=not applicable.

Table 2. Patient characteristics. All patients with Primary AF AF hospitalization (n=16,655) (n=1527)

Secondary AF hospitalization (n=6067)

Outpatientmanaged AF (n=9061)

Age, years mean (SD) Age group, n (%) 65-74 years ≥75 years Male, n (%) Region, n (%) South Midwest Northeast West Other Comorbidities,* n (%)

74.0 (4.5)

73.3 (4.6)

74.5 (4.3)

73.7 (4.5)

7396 (44.4) 9259 (55.6) 9427 (56.6)

789 (51.7) 738 (48.3) 740 (48.5)

2356 (38.8) 3711 (61.2) 3371 (55.6)

4251 (46.9) 4810 (53.1) 5316 (58.7)

3983 (23.9) 950 (5.7) 10,025 (60.2) 1690 (10.1) 7 (0.04)

414 (27.1) 98 (6.4) 908 (59.5) 106 (6.9) 1 (0.07)

1458 (24.0) 382 (6.3) 3623 (59.7) 601 (9.9) 3 (0.05)

2111 (23.3) 470 (5.2) 5494 (60.6) 983 (10.8) 3 (0.03)

Unauthorized Use Prohibited Coronary artery disease Stroke/TIA/cerebrovascular disease Hypertension with heart failure Hypertension without heart failure Heart failure Cardiac dysrhythmias/conduction disorders Diabetes mellitus Hypothyroidism Structural heart disease Emphysema/COPD

1979 (11.9) 3234 (19.4) 5334 (32.0) 3051 (18.3) 5257 (31.6) 4781 (28.7) 4282 (25.7) 1810 (10.9) 3353 (20.1) 2712 (16.3)

270 (17.7) 297 (19.4) 598 (39.2) 746 (48.9) 590 (38.6) 681 (44.6) 349 (22.9) 226 (14.8) 393 (25.7) 276 (18.1)

1174 (19.4) 1690 (27.9) 2971 (49.0) 2057 (33.9) 2942 (48.5) 2222 (36.6) 1981 (32.7) 759 (12.5) 1601 (26.4) 1600 (26.4)

535 (5.9) 1247 (13.8) 1765 (19.5) 248 (2.7) 1725 (19.0) 1878 (20.7) 1952 (21.5) 825 (9.1) 1359 (15.0) 836 (9.2)

*Defined according to relevant ICD-9-CM codes. AF=atrial fibrillation; COPD=chronic obstructive pulmonary disease; SD=standard deviation; TIA=transient ischemic attack.

6

Adv Ther (2009) 26(8):.

Table 3. Cumulative prescription costs in 2006. All patients with Primary AF AF hospitalization (n=16,655) (n=1527)


Outpatientmanaged AF (n=9061)

310,259 (100) 3948.68

383,015 (100) 3126.64

Total Number of prescriptions, n (%) Total per-patient prescription cost, mean US$ Rhythm-control agents Number of prescriptions, n (%) Total per-patient prescription cost, mean US$ Rate-control agents Number of prescriptions, n (%) Total per-patient prescription cost, mean US$ Anticoagulants Number of prescriptions, n (%) Total per-patient prescription cost, mean US$ Other Number of prescriptions, n (%) Total per-patient prescription cost, mean US$

764,817 (100) 3457.16

71,543 (100) 3465.52

Unauthorized Use Prohibited 19,866 (2.6) 146.09

3282 (4.6) 259.69

6423 (2.1) 117.12

10,161 (2.7) 146.34

110,150 (14.4) 217.97

10,643 (14.9) 246.54

38,935 (12.5) 200.08

60,572 (15.8) 225.14

76,860 (10.0) 177.41

7133 (10.0) 160.00

23,688 (7.6) 140.52

46,039 (12.0) 205.04

557,941 (73.0) 2915.69

50,485 (70.6) 2799.31

241,213 (77.7) 266,243 (69.5) 3490.97 2550.12

AF=atrial fibrillation. Figure 2. Proportion of patients reaching the donut hole and catastrophic coverage thresholds in 2006. AF=atrial fibrillation.

50

Proportion of patients (%)

40

Below donut hole threshold ($5100) 58.8 41.2

59.9 40.1

37.6

38.6

63.5

36.5

44.5

55.5 38.3

36.2

Unauthorized Use Prohibited 30

20

27.3

21.2

21.3

17.2

10

0

All patients (n=16,655)

Primary AF hospitalization (n=1527)


Outpatient-managed AF (n=9061)

Adv Ther (2009) 26(8):.

7

Figure 3. Distribution of patients stratified by prescription costs at $50 intervals. 7.2

58.8% reached donut-hole threshold ($2250)

Proportion of patients (%)

7.0 1.2

21.2% reached catastrophic coverage threshold ($5100)

1.0 0.8

Unauthorized Use Prohibited 0.6

0.4

0.2

0 $8000

Total prescription costs in 2006 ($50 intervals)

AF hospitalizations; 186±84 (median 177, range 2-364) days for patients with secondary AF hospitalizations; and 209±82 (median 206, range 2-364) days for patients with outpatientmanaged AF. Among patients reaching the $5100 threshold in 2006, the mean±SD time to reach the threshold was 257±71 (median 267, range 9-364) days for the overall population; 260±73 (median 274, range 51-364) days for patients with primary AF hospitalizations; 250±74 (median 260, range 12-364) days for patients with secondary AF hospitalizations; and 264±67 (median 272, range 9-364) days for patients with outpatient-managed AF.

prescription expenditure in excess of the $2250 donut hole threshold, and 21.2% reached the $5100 catastrophic coverage threshold. The mean time to reach the donut hole threshold was 6 months and patients reaching the catastrophic coverage threshold did so in approximately 8 months. Total prescription costs among patients with AF in this study were high, with mean costs of $3457.16 per patient in 2006. Using the data in Figure 3 for the distribution of patients by total prescription costs, and assuming a similar distribution in the general population of Medicare Part D enrollees receiving standard benefit (in which the patient covers 25% of costs up to the donut hole [plus $250 deductible], 100% of costs during the donut hole, and 5% of costs during catastrophic coverage), we estimated the mean prescription costs that would be covered by the patient at approximately $1709. Depending on whether patients reached the donut hole and catastrophic coverage thresholds, we estimated mean costs to the patient of $424 for the 41.2% remaining below the donut hole threshold, $2000 for the 38% entering and remaining in the


DISCUSSION

The results of this study suggest that a high proportion of elderly patients with AF have prescription costs in excess of the Medicare Part D donut hole threshold, and would thus be required to pay the full cost of their drugs for at least a portion of the year under the Part D plan. In 2006, 58.8% of the Medicare Advantage patients with AF included in this analysis had

8

Adv Ther (2009) 26(8):.

donut hole, and $3689 for the 21.2% of patients reaching the catastrophic coverage threshold. Both total prescription costs and the proportion of patients entering the donut hole were higher in our study than reported in a previous study in patients with AF by Evans-Molina et al.14 Evans-Molina et al. estimated annual prescription costs of between $1723 and $2622 for patients with AF aged ≥65 years (depending on cost assumptions relating to the dose of medication used), with 27%-45% of patients reaching the 2006 donut hole threshold ($2250 in total drug costs) and 3%-11% reaching a catastrophic coverage threshold of $3600 in true out-of-pocket costs to the patient (approximately equivalent to the $5100 Part D threshold depending on copayment costs prior to the donut hole), based on an AARP-endorsed plan.14 These differences may

was more than twice that estimated for patients in the overall Medicare population who filled at least one prescription in 2007 (26%, based on prescription costs only, without accounting for whether patients were on low income subsidy [LIS] plans); the proportion reaching the catastrophic coverage threshold (21.2%) was approximately five times that in the overall population (4%).11 Furthermore, although the percentage of patients with AF entering the donut hole was slightly lower than among patients receiving medications for the treatment of Alzheimer’s disease (64% in 2007), it was greater than among patients grouped according to the use of common cardiovascular treatments, including oral antidiabetics (51%), antidepressants (45%), angiotensin receptor blockers (42%), or statins (39%).11

be due to variations in the methodology used to evaluate prescription costs or differences in the patient populations between these two studies. In particular, the study by Evans-Molina et al. was based on a small subset of participants in a trial to assess stroke prevention in patients with AF, which may be less representative of all patients with AF than the broader managed care population in our study.14 However, it should also be noted that all patients in our study were enrollees in commercial Medicare Advantage plans and could potentially have higher annual prescription costs than would be observed in the broader Part D population. Although some caution should be taken in generalizing the observations in the Medicare Advantage patients in our study to the overall population of Part D enrollees, we believe that it is important to highlight the high prescription burden in the patients with AF in our analysis in relation to the overall Medicare population and patients with other conditions. For example, the proportion of patients with AF reaching the donut hole threshold in our study (58.8%)

Prescriptions relating to AF management accounted for only 15% of total prescription costs (rhythm-control drugs, 4%; rate-control drugs, 6%; and anticoagulants, 5%), indicating that the treatment of comorbidities, such as heart failure, diabetes, hypertension, and stroke, among patients with AF is the primary driver of the high prescription cost burden in this population. Consistent with this, prescription costs were highest in patients hospitalized with a secondary diagnosis of AF, for whom comorbid conditions were the primary reason for hospitalization, and these patients were more likely to reach the donut hole and catastrophic coverage thresholds than those hospitalized with a primary AF diagnosis. Patients with outpatient-managed AF had the lowest prescription costs and were the least likely to reach the payment thresholds, suggesting that, as might be expected, hospitalized patients were treated more intensively than those treated as outpatients. The low proportion of costs associated with AF treatments in relation to other drugs suggests that AF treatment itself may not be



Adv Ther (2009) 26(8):.

receiving sufficient attention, and that physicians are instead concentrating on treating or preventing the clinical sequelae of AF, such as stroke or heart failure. Further investigation is required to determine whether novel treatment strategies to improve the management of AF can reduce the number of patients developing these conditions, and thus potentially lower patients’ prescription cost burden in the long term. By reducing hospitalizations and mortality due to AF, improved treatment strategies for AF management would also have the potential to reduce the national cost burden of AF, which is currently in the region of $6.65 billion annually in the US.19

9

to AF and comorbid cardiovascular conditions due to medication under-use. This is of particular concern in this patient group, as those most at risk of entering the donut hole due to a high prescription burden are also likely to be those with the greatest number of cardiovascular comorbidities, and thus at the greatest risk of adverse clinical outcomes. It is currently unclear which medications are the most likely to be stopped/underused in patients with AF entering the donut hole and which conditions would be most affected. Further investigation is critical to evaluate the effect of the donut hole on pat-


The impact of the increased cost burden for patients with AF entering the donut hole on the potential discontinuation of AF-related drugs and other treatments is not currently known. However, it is likely that these patients would be more reluctant to take medications in the knowledge that they need to pay full prescription costs on reaching the donut hole. Recent studies in Medicare Part D plan enrollees have demonstrated that on entering the donut hole, medication utilization is reduced on average by 187 days of treatment,12 or by approximately 14%,13 compared with drug use prior to the coverage gap. An analysis by the Kaiser Family Foundation among Medicare Part D enrollees taking one of eight classes of medication for common chronic conditions, indicated that of those who reached the donut hole in 2007, 15% stopped taking their medication, 5% switched to an alternative drug in the same class, and 1% reduced their medication use.11 Other studies have also demonstrated an association between increased copayment costs and reduced adherence with medication.20‑22 An increase in drug discontinuation in patients with AF as a result of the donut hole could potentially increase hospitalizations due

terns of drug utilization in patients with AF and fully understand the impact on quality of care and clinical outcomes. Limitations As mentioned above, all patients in this study were identified from the Integrated Healthcare Information Systems National Managed Care Benchmark Database and were thus enrollees in commercial Medicare Advantage plans. These patients typically account for only one-third of all Part D enrollees23 and may not be fully representative of the general Medicare Part D population. Depending on their specific plan, some Medicare Advantage patients will have greater prescription coverage than under the standard Part D benefit plan, which may encourage higher levels of prescription use in these patients. As a result, prescription cost estimates in this analysis may be slightly higher than would be observed in the broader Part D population. In addition, the donut hole thresholds were based on the 2006 Medicare Part D standard benefit plan, while many available plans have an alternative structure, which may impact on the costs to the patient and the time before entering the donut hole. For example, some plans with an increased monthly premium offer coverage


10

for some or all prescription charges during the donut hole. Furthermore, approximately 36% of all Part D enrollees receive LIS benefit and are not subject to donut hole payments.23 If we exclude this proportion of patients from those who had prescription costs in excess of the donut hole and catastrophic coverage thresholds in our analysis, the estimated proportion of patients with AF subject to donut hole payments is 38%, with 14% reaching catastrophic coverage. Moreover, given the fact that individuals with the lowest income are often those with the poorest overall health,24 and who are thus likely to have the highest prescription expenditure, the proportion of Part D enrollees subject to donut hole payments in the real world may be even lower. Other limitations of the study include the fact that it was not possible to capture patients’ activity or medication use outside of that recorded in the Integrated Healthcare Information Systems National Managed Care Benchmark Database, and that only limited data were available in the database on patients’ clinical profile. Additionally, the database may not be fully representative of the patients and practice patterns in hospitals across the US, and health plans in the Midwest and West are underrepresented.

Adv Ther (2009) 26(8):.

population, with the majority of prescription costs being related to the treatment of non-AF comorbid conditions.

ACKNOWLEDGMENTS Financial and editorial support for this manuscript has been provided by sanofi-aventis US, Inc. Dr. Kim is a research consultant to sanofi-aventis. Dr. Lin is an employee of sanofiaventis. Mr. Kreilick is an employee of ProUnlimited, which has a research consulting agreement with sanofi-aventis. We would like to acknowledge Mehul Dalal for his thoughtful review of the draft manuscript. Elizabeth Harvey, PhD, of UBC Scientific Solutions, provided editorial support. Parts of these data were presented as a poster at the 9th Scientific Forum on Quality of Care and Outcomes Research in Cardiovascular Disease and Stroke, Washington DC, USA, April 23-25, 2009.


CONCLUSIONS

REFERENCES 1.

Go AS, Hylek EM, Phillips KA, et al. Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and stroke prevention: the AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study. JAMA. 2001;285:2370-2375.

2.

Thrall G, Lane D, Carroll D, Lip GY. Quality of life in patients with atrial fibrillation: a systematic review. Am J Med. 2006;119:448 e441-419.

3.

Benjamin EJ, Wolf PA, D’Agostino RB, Silbershatz H, Kannel WB, Levy D. Impact of atrial fibrillation on the risk of death: the Framingham Heart Study. Circulation. 1998;98:946-952.

4.

Krahn AD, Manfreda J, Tate RB, Mathewson FA, Cuddy TE. The natural history of atrial fibrillation: incidence, risk factors, and prognosis in the Manitoba Follow-Up Study. Am J Med. 1995;98:476-484.

5.

Wolf PA, Mitchell JB, Baker CS, Kannel WB, D’Agostino RB. Impact of atrial fibrillation on mortality, stroke, and medical costs. Arch Intern Med. 1998;158:229-234.

Unauthorized Use Prohibited Patients with AF aged ≥65 years have a high burden of comorbidities and, as such, annual prescription costs in this population are considerable. Although some care should be taken in generalizing our results to the broader Medicare population, among the Medicare Advantage patients with AF included in this study, more than half would have had prescription costs in excess of the Part D donut hole threshold in 2006. The costs of AF treatment itself are relatively low in this

Adv Ther (2009) 26(8):.

6.

7.

Lloyd-Jones D, Adams R, Carnethon M, et al. Heart disease and stroke statistics–2009 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation. 2009;119:e21-181. Miyasaka Y, Barnes ME, Gersh BJ, et al. Secular trends in incidence of atrial fibrillation in Olmsted County, Minnesota, 1980 to 2000, and implications on the projections for future prevalence. Circulation. 2006;114:119-125.

11

Results from a retrospective cohort study. Int J Impot Res. 2004;16:350-353. 16. Kessler RC, Cantrell CR, Berglund P, Sokol MC. The effects of copayments on medication adherence during the first two years of prescription drug treatment. J Occup Environ Med. 2007;49:597-609. 17. Brixner DI, Jackson KC, 2nd, Sheng X, Nelson RE, Keskinaslan A. Assessment of adherence, persistence, and costs among valsartan and hydrochlorothiazide retrospective cohorts in free- and fixed-dose combinations. Curr Med Res Opin. 2008;24:2597-2607.


8.

9.

Lakshminarayan K, Solid CA, Collins AJ, Anderson DC, Herzog CA. Atrial fibrillation and stroke in the general Medicare population: a 10-year perspective (1992 to 2002). Stroke. 2006;37:1969-1974.

Bach PB, McClellan MB. A prescription for a modern Medicare program. N Engl J Med. 2005;353:2733-2735.

18. Spyropoulos AC, Lin J. Direct medical costs of venous thromboembolism and subsequent hospital readmission rates: an administrative claims analysis from 30 managed care organizations. J Manag Care Pharm. 2007;13:475-486.

10. Medicare. The Official U.S. Government Site for People with Medicare http://www.medicare.gov/. Accessed February 24, 2009.

19. Coyne KS, Paramore C, Grandy S, Mercader M, Reynolds M, Zimetbaum P. Assessing the direct costs of treating nonvalvular atrial fibrillation in the United States. Value Health. 2006;9:348-356.

11. Hoadley J, Hargrave E, Cubanski J, Neuman T. The Medicare Drug Benefit: the Medicare Part D coverage gap: costs and consequences in 2007. The Henry J. Kaiser Family Foundation; August 2008.

20. Taira DA, Wong KS, Frech-Tamas F, Chung RS. Copayment level and compliance with antihypertensive medication: analysis and policy implications for managed care. Am J Manag Care. 2006;12:678-683.

12. Sun SX, Lee KY. The Medicare Part D doughnut hole: effect on pharmacy utilization. Manag Care Interface. 2007;20:51-55, 59.

21. Pedan A, Varasteh L, Schneeweiss S. Analysis of factors associated with statin adherence in a hierarchical model considering physician, pharmacy, patient, and prescription characteristics. J Manag Care Pharm. 2007;13:487-496.

13. Zhang Y, Donohue JM, Newhouse JP, Lave JR. The effects of the coverage gap on drug spending: a closer look at Medicare Part D. Health Aff (Millwood). 2009;28:w317-w325. 14. Evans-Molina C, Regan S, Henault LE, Hylek EM, Schwartz GR. The new Medicare Part D prescription drug benefit: an estimation of its effect on prescription drug costs in a Medicare population with atrial fibrillation. J Am Geriatr Soc. 2007;55:1038-1043.

22. Chernew ME, Shah MR, Wegh A, et al. Impact of decreasing copayments on medication adherence within a disease management environment. Health Aff (Millwood). 2008;27:103-112.


15. Blumentals WA, Gomez-Caminero A, Joo S, Vannappagari V. Should erectile dysfunction be considered as a marker for acute myocardial infarction?

23. The Henry J. Kaiser Family Foundation. Medicare: a primer. January, 2009; Available at http:// www.kff.org/medicare/upload/7615-02.pdf. Accessed June 2, 2009. 24. Pamuk E, Makuc D, Heck K, Reuben C, Lochner K. NCHS: Socioeconomic status and health chartbook. Health, United States, 1998. Hyattsville, MD: National Center for Health Statistics; 1998.