Abstract
Introduction
Erenumab is indicated for migraine preventive treatment in adults. The objective of this study was to provide descriptive information on real-world use of erenumab including patient profile and treatment patterns.
Methods
We completed a retrospective review of US data (through May 2019) from the IBM MarketScan® Early View Databases, identifying adult patients newly treated with erenumab with a migraine claim in the year prior to first erenumab claim (index) and at least 1 year of continuous pre-index medical and pharmacy insurance coverage, to assess pre- and post-erenumab migraine characteristics, comorbidities, healthcare resource utilization, and associated costs. All data were summarized using descriptive statistics.
Results
A total of 9753 patients met inclusion criteria. The average (SD) age was 46 (12) years, 85% of patients were female, and 64% had at least one claim for chronic migraine; 70% of erenumab users had an initial dose of 70 mg; 77% of patients in the 6-month follow-up sample (n = 4437) remained on their initial erenumab dose. Persistence at 6-month follow-up was 47.3% with a mean (95% CI) proportion of days covered of 0.68 (0.67, 0.68). In the post-erenumab period, claims for comorbidities of non-migraine headaches and anxiety were reduced and there was a shift to decreased use of acute and preventive medications. Reductions in overall use and associated cost of healthcare resources such as inpatient hospitalization and outpatient office visits were minimal, with slightly more pronounced reductions in the subgroup of patients that were persistent to erenumab.
Conclusions
We observed reductions in claims for important migraine characteristics, comorbidities, and a shift to decreased use of acute and preventive migraine medications—observations indicative of the real-world effectiveness of erenumab. Further examination is required as persistence to erenumab, which may be influenced by dose titration, appears to be an important factor in changes to healthcare resource utilization and costs.
Supplementary Information
The online version contains supplementary material available at 10.1007/s40122-021-00319-z.
Keywords: Cost, Erenumab, Healthcare resource utilization, Migraine, Persistence
Key Summary Points
| Why carry out this study? |
| Erenumab-aooe was approved in the USA on 17 May 2018 for the preventive treatment of migraine in adults. |
| As a first-in-class option for migraine prevention, knowledge about the real-world experience of patients receiving erenumab is limited. |
| This study sought to examine the impact of erenumab treatment on migraine characteristics, comorbidities, healthcare resource utilization, and associated costs. |
| What was learned from the study? |
| Reductions in claims for important migraine characteristics, comorbidities, and a shift to decreased use of acute and preventive migraine medications in the post-erenumab follow-up period are indicative of the real-world effectiveness of erenumab. |
| Persistence to erenumab is an important factor requiring further examination. |
Introduction
Migraine is a neurological disorder characterized by recurrent headache attacks of moderate to severe pain. It affects up to 18% of the female population and 6% of the male population. Reduced quality of life, functional impairment, reduced productivity, and high economic burden, are all associated with migraine, highlighting the need for effective therapeutic strategies [1–4]. The clinical management of migraine relies on the use of preventive medications to preempt migraine attacks and the use of acute medications to abort migraine attacks. Many of the available preventive medications, including tricyclic antidepressants, anticonvulsants, beta blockers, and calcium channel blockers, were not developed specifically for migraine but have demonstrated efficacy in clinical trials [5, 6]. However, many patients have an inadequate response or do not continue treatment because of tolerability issues and thus fail to maintain control of their migraine [7, 8]. A recent retrospective analysis of claims for 8707 patients showed persistence of 25% after the first 6 months of treatment declining to 14% by 12 months, for the 14 most common medications used for migraine prevention [9]. These low rates of persistence may lead to increased burden of migraine as healthcare resource utilization has been shown to be greater among individuals who failed multiple migraine preventive therapies [10]. Importantly, there may also be substantial health risks when patients decide to alter their prescribed treatment plan without consulting their healthcare provider [11]. Thus, more effective and tolerable preventive treatments may promote wider use with the potential to reduce the burden of migraine to individuals and society.
Erenumab (erenumab-aooe in the USA) was approved in the USA by the Food and Drug Administration (FDA) on 17 May 2018 for the preventive treatment of migraine in adults. Information on the real-world experience of patients receiving erenumab is limited. We conducted a claims-based, retrospective observational study to examine the impact of erenumab treatment as a migraine preventive on migraine characteristics, comorbidities, healthcare resource utilization, and associated costs.
Methods
Study Design
We used administrative claims data from the 2017–2019 IBM MarketScan® Early View Commercial Claims and Encounters (CCAE) and Medicare Supplemental and Coordination of Benefits databases. The CCAE database contains employer-sponsored private health insurance records encompassing employees, their spouses, and dependents. With the MarketScan® Early View Data Set, data are available within 45 days of the end of the service month. The Medicare Supplemental and Coordination of Benefits database contains the data of retirees with employer-sponsored Medicare supplemental insurance. MarketScan® databases are in compliance with the Health Information Portability and Accountability Act of 1996 (HIPAA). Data from the MarketScan® database were used under a licensing agreement with IBM Watson Health. As this study did not involve the collection, use, or transmittal of individually identifiable data, it did not require institutional review board review or approval. All named authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship for this article, take responsibility for the integrity of the work as a whole, and have given their approval for this version to be published.
Patient Identification
Adult patients (at least 18 years of age) with at least one National Drug Code (NDC) claim for erenumab from May 2018 through May 2019 and a migraine diagnosis in the year prior to the first erenumab claim were identified. The date of the first claim of erenumab was set as the index date. Migraine diagnosis was based on the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes G43.xx and prescription claims using NDC or Healthcare Common Procedure Coding System (HCPCS) codes for acute migraine medications (e.g., triptans or ergotamines). Patients were required to have at least 1 year of continuous enrollment prior to the index date with both medical and pharmacy insurance coverage.
Study Variables
Patient demographic characteristics assessed at the index date included age, sex, geographic region, and provider specialty. Provider specialties were identified on the basis of inpatient/outpatient migraine diagnosis (ICD-10-CM) and/or procedure (HCPCS) claims within 90 days prior and most proximal to or on the erenumab index date. Patients may have had multiple provider specialties per claim.
Patient clinical characteristics for the full study population were characterized using ICD-10-CM codes for the 12-month pre-index period (baseline). Erenumab dispensing patterns included initial dose prescribed (70 mg vs 140 mg), switches in dose and time to switch (in days), and persistence. Persistence was evaluated using 30-day and 45-day allowable gaps in therapy as well as the proportion of days covered (PDC) method using stockpiling algorithm [12]. PDC was calculated as the ratio of number of days the patient was covered by erenumab in a period to the total number of days in the period. PDC was described at 3- and 6-months post-index.
In a sample of patients with 6-month follow-up data, pre- and post-erenumab migraine characteristics, comorbidities, treatment characteristics, healthcare resource utilization, and associated costs in US dollars ($) were described 6 months pre- and post-index. Pre- and post-erenumab treatment characteristics were based on prescription claims using NDC or HCPCS codes. Acute treatments were identified as triptans, nonsteroidal anti-inflammatory drugs (NSAIDs), opioids, analgesics (non-NSAID, non-opioid analgesics), and ergotamines. Preventive treatments were identified as anticonvulsants, antidepressants, antihypertensives, botulinum toxins, and others. Analysis included measures of the number and type of acute and preventive treatments. Results include stratification by persistence and non-persistence to erenumab with persistence defined as having continuously filled erenumab prescriptions through day 180 of the follow-up period with a 30-day allowable gap between fills. Healthcare utilization was measured as prescription of medications used for the acute and preventive treatment of migraine, emergency room visits (service place code 23), outpatients care (service place code 22 [outpatient hospital] or 11 [office]), in-patient hospitalizations (service code 21) and inpatient length of hospital stay, brain imaging studies (Current Procedural Terminology [CPT] code or HCPCS code for an imaging procedure [e.g., CT Head 70,450–70,492, MRI Head 70,336, 76,390, 70,540–70,559] containing HCPCS modifier code 26 to avoid double counting) and their associated costs for 6 months prior to and 6 months after the first claim of erenumab. For both commercial and Medicare patients in the study, healthcare costs were total reimbursed amount, including health plan payment and patients' out-of-pocket payment (copay, deductible, and coinsurance).
Statistical Analysis
All data were summarized using descriptive statistics. Categorical variables were presented as frequencies and percentages. Continuous variables were presented as mean/medians with corresponding standard deviations or standard errors. Subgroup analyses based on provider specialty were conducted for select baseline comorbidities and migraine characteristics. For these select subgroup analyses, patients were grouped on the basis of specialist visit during the pre-index baseline period as (1) patients who saw a neurologist and (2) patients who saw a family practice/internal medicine physician but no neurologist and no pain specialist/anesthesiologist.
Results
Sample Size
After inclusion and exclusion criteria were met, 9753 patients with an erenumab claim comprised the full study population. Of these, 7280 patients had 3-month follow-up data and 4437 had 6-month follow-up data.
Demographics and Baseline Characteristics
The average (SD) age at first erenumab claim for the full study population was 46.5 (12.1) years and 85.4% of users were women. Demographic characteristics for patients with 6-month follow-up data were similar to that of the full study population (Table 1). The 45–54 years age group made up the largest proportion of patients in the full study population (32.1%). Overall, 69.4% of patients had at least one comorbidity (Table 1). The most common comorbidities observed in the baseline period were other (non-migraine) headaches (29.0%), anxiety (21.8%), hypertension (18.4%), depression (16.5%), low back pain (15.5%), and osteoarthritis/spondylosis (14.5%).
Table 1.
Patient demographics and baseline clinical characteristics in the full study population and 6-month follow-up sample
| All patients | 6-month follow-up patients* | |
|---|---|---|
| N = 9753 | N = 4437 | |
| Age, mean (SD) | 46.5 (12.1) | 46.7 (11.8) |
| Age category, n (%) | ||
| 18–34 | 1550 (15.9) | 655 (14.8) |
| 35–44 | 2458 (25.2) | 1104 (24.9) |
| 45–54 | 3134 (32.1) | 1497 (33.7) |
| 55–64 | 2180 (22.4) | 1000 (22.5) |
| 65–74 | 353 (3.6) | 148 (3.3) |
| 75+ | 78 (0.8) | 33 (0.7) |
| Sex (female), n (%) | 8329 (85.4) | 3808 (85.8) |
| Geographic region, n (%) | ||
| Northeast | 1047 (10.7) | 507 (11.4) |
| Midwest | 2134 (21.9) | 934 (21.1) |
| South | 2998 (30.7) | 1287 (29.0) |
| West | 1373 (14.1) | 633 (14.3) |
| Unknown | 2201 (22.6) | 1076 (24.3) |
| Year of index date | ||
| 2018 | 6185 (63.4) | 4437 (100.0) |
| 2019 | 3568 (36.6) | 0 (0.0) |
| Comorbidities | ||
| Asthma | 727 (7.5) | 174 (3.9) |
| Hypertension | 1790 (18.4) | 457 (10.3) |
| Acute myocardial infarction | 37 (0.4) | 8 (0.2) |
| Ischemic stroke | 77 (0.8) | 23 (0.5) |
| Constipation | 306 (3.1) | 76 (1.7) |
| Irritable bowel syndrome | 232 (2.4) | 48 (1.1) |
| Complications of constipation | 187 (1.9) | 44 (1.0) |
| Anxiety | 2125 (21.8) | 646 (14.6) |
| Depression | 1606 (16.5) | 500 (11.3) |
| Other (non-migraine) headaches | 2829 (29.0) | 810 (18.3) |
| Low back pain | 1511 (15.5) | 463 (10.4) |
| Osteoarthritis/spondylosis | 1411 (14.5) | 393 (8.9) |
| Fibromyalgia | 549 (5.6) | 171 (3.9) |
| No comorbidities | 2981 (30.6) | 2056 (46.3) |
| Migraine characteristics | ||
| Migraine with aura, including persistent aura | 2611 (26.8) | 915 (20.6) |
| Migraine with aura, excluding persistent aura | 2530 (25.9) | 890 (20.1) |
| Chronic migraine | 6274 (64.3) | 2904 (65.4) |
| Menstrual migraine | 259 (2.7) | 103 (2.3) |
| Status migrainosus | 2607 (26.7) | 1103 (24.9) |
SD standard deviation
*Comorbidities and migraine characteristics for the 6-month follow-up period were based on a 6-month baseline period whereas the same elements for all patients were based on a 12-month baseline period
Neurologists were the most common (47.3%) provider specialty seen by patients with migraine initiating erenumab, followed by other specialties including family practice (9.6%), nurse practitioner/physician’s assistant (6.7%), and internal medicine (5.0%) (data not shown). As expected, additional analysis by provider type indicates that patients who saw a neurologist during the baseline period had a higher frequency of comorbidities in the year prior to erenumab use compared to patients who saw a family practice/internal medicine physician but no neurologist and no pain specialist/anesthesiologist during the 12-month baseline period (Supplemental Table 1).
Erenumab Dispensing Patterns and Persistence
A majority (70.2%) of erenumab users were administered an initial dose of 70 mg; 76.9% of patients in the 6-month follow-up sample remained on their initial erenumab dose; 833 patients (26.6%) starting on 70 mg switched to 140 mg and 194 patients (14.9%) starting on 140 mg switched to 70 mg (Table 2). Persistence at 3-month follow-up was 67.1% using a 30-day gap and 72.6% using a 45-day gap; persistence at 6-month follow-up was 41.8% and 47.3%, respectively. Mean (95% CI) PDC for the 3-month follow-up after erenumab initiation was 0.80 (0.79, 0.80) and for the 6-month follow-up was 0.68 (0.67, 0.68) (Table 2). There was no difference in persistence based on erenumab dose or provider specialty (data not shown). There were, however, notable differences in erenumab dose switching for persistent and non-persistent patients. Thirty two percent of persistent patients, versus 17% of non-persistent patients, switched from their initial dose. Seventy five percent of persistent patients and 67% of non-persistent patients started at 70 mg; 36% of the persistent patients starting at 70 mg switched to 140 mg at a mean of 135 days and 19% of non-persistent starting at 70 mg switched to 140 mg at a mean of 126 days. Of the 25% of persistent patients starting at 140 mg, 21% switched to 70 mg at a mean of 130 days. Of the 33% of non-persistent patients starting at 140 mg, 12% switched to 70 mg at a mean of 121 days.
Table 2.
Erenumab dispensing patterns and persistence in the full study population and 6-month follow-up sample stratified by persistence
| All patients (N = 9753) | 6-month follow-up patients | |||
|---|---|---|---|---|
| All (N = 4437) | Erenumab persistent (N = 1853) | Erenumab non-persistent (N = 2584) | ||
| Remained on initial dose, n (%) | 8236 (84.4) | 3410 (76.9) | 1261 (68.1) | 2149 (83.2) |
| Switched from initial dose, n (%) | 1517 (15.6) | 1027 (23.1) | 592 (31.9) | 435 (16.8) |
| 70 mg initial dose | n = 6845 (70.2) | n = 3136 (70.7) | n = 1394 (75.2) | n = 1742 (67.4) |
| Switched to 140 mg, n (%) | 1202 (17.6) | 833 (26.6) | 497 (35.7) | 336 (19.3) |
| Time to switch (days) | ||||
| Mean (SD) | 114.2 (65.0) | 131.3 (67.8) | 135.0 (64.9) | 125.8 (71.6) |
| 95% CI | (110.5, 117.9) | (126.7, 135.9) | (129.3, 140.7) | (118.2, 133.5) |
| 140 mg initial dose | n = 2908 (29.8) | n = 1301 (29.3) | n = 459 (24.8) | n = 842 (32.6) |
| Switched to 70 mg, n (%) | 315 (10.8) | 194 (14.9) | 95 (20.7) | 99 (11.8) |
| Time to switch (days) | ||||
| Mean (SD) | 104.3 (71.2) | 125.6 (78.3) | 130.1 (79.7) | 121.3 (77.7) |
| 95% CI | (96.5, 112.2) | (114.5, 136.7) | (114.0, 146.2) | (105.8, 136.8) |
| Erenumab proportion of days covered | 3-month follow-up | |||
| n | 7280 | 4437 | 1853 | 2584 |
| Mean (SD) | 0.80 (0.24) | 0.68 (0.28) | 0.94 (0.07) | 0.48 (0.21) |
| 95% CI | (0.79, 0.80) | (0.67, 0.68) | (0.94, 0.95) | (0.47, 0.49) |
CI confidence interval, SD standard deviation
Pre- and Post-erenumab Comorbidities and Migraine Characteristics
In the sample of patients with 6-month follow-up data, claims submitted for migraine with aura including and excluding persistent aura, chronic migraine, menstrual migraine, and status migrainosus were reduced in the 6-month post-index period as compared with the 6-month pre-index period with reductions in claims for chronic migraine and status migrainosus being slightly more pronounced for persistent patients (Table 3). Medical claims for comorbidities submitted during the 6-month pre-index period and the 6-month post-index period remained relatively consistent except for non-migraine headaches (18.3% vs 13.2%) and anxiety (14.6% vs 13.6%), which decreased in the 6-month post-compared with the 6-month pre-index period; these decreases were consistent for persistent and non-persistent patients (Table 3).
Table 3.
Patient migraine characteristics and comorbidities in the 6 months pre- and post-erenumab stratified by persistence
| Pre-erenumab | Post-erenumab | |||||
|---|---|---|---|---|---|---|
| All (N = 4437) | Erenumab persistent (N = 1853) | Erenumab non-persistent (N = 2584) | All (N = 4437) | Erenumab persistent (N = 1853) | Erenumab non-persistent (N = 2584) | |
| Comorbidities | ||||||
| Asthma | 174 (3.9) | 68 (3.7) | 106 (4.1) | 176 (4.0) | 63 (3.4) | 113 (4.4) |
| Hypertension | 457 (10.3) | 186 (10.0) | 271 (10.5) | 425 (9.6) | 168 (9.1) | 257 (9.9) |
| Acute myocardial infarction | 8 (0.2) | 1 (0.1) | 7 (0.3) | 9 (0.2) | 3 (0.2) | 6 (0.2) |
| Ischemic stroke | 23 (0.5) | 7 (0.4) | 16 (0.6) | 20 (0.5) | 8 (0.4) | 12 (0.5) |
| Constipation | 76 (1.7) | 28 (1.5) | 48 (1.9) | 85 (1.9) | 29 (1.6) | 56 (2.2) |
| Irritable bowel syndrome | 48 (1.1) | 20 (1.1) | 28 (1.1) | 52 (1.2) | 26 (1.4) | 26 (1.0) |
| Complications of constipation | 44 (1.0) | 18 (1.0) | 26 (1.0) | 52 (1.2) | 19 (1.0) | 33 (1.3) |
| Anxiety | 646 (14.6) | 238 (12.8) | 408 (15.8) | 604 (13.6) | 224 (12.1) | 380 (14.7) |
| Depression | 500 (11.3) | 192 (10.4) | 308 (11.9) | 477 (10.8) | 173 (9.3) | 304 (11.8) |
| Other (non-migraine) headaches | 810 (18.3) | 299 (16.1) | 511 (19.8) | 584 (13.2) | 201 (10.8) | 383 (14.8) |
| Low back pain | 463 (10.4) | 192 (10.4) | 271 (10.5) | 440 (9.9) | 158 (8.5) | 282 (10.9) |
| Osteoarthritis/spondylosis | 393 (8.9) | 168 (9.1) | 225 (8.7) | 431 (9.7) | 166 (9.0) | 265 (10.3) |
| Fibromyalgia | 171 (3.9) | 62 (3.3) | 109 (4.2) | 146 (3.3) | 53 (2.9) | 93 (3.6) |
| No comorbidities | 2056 (46.3) | 893 (48.2) | 1163 (45.0) | 2207 (49.7) | 972 (52.5) | 1235 (47.8) |
| Migraine characteristics, n (%) | ||||||
| Migraine with aura, including persistent aura | 915 (20.6) | 372 (20.1) | 543 (21.0) | 700 (15.8) | 287 (15.5) | 413 (16.0) |
| Migraine with aura, excluding persistent aura | 890 (20.1) | 364 (19.6) | 526 (20.4) | 682 (15.4) | 280 (15.1) | 402 (15.6) |
| Chronic migraine | 2904 (65.4) | 1216 (65.6) | 1688 (65.3) | 2531 (57.0) | 1041 (56.2) | 1490 (57.7) |
| Menstrual migraine | 103 (2.3) | 37 (2.0) | 66 (2.6) | 77 (1.7) | 33 (1.8) | 44 (1.7) |
| Status migrainosus | 1103 (24.9) | 442 (23.9) | 661 (25.6) | 806 (18.2) | 296 (16.0) | 510 (19.7) |
Pre- and Post-erenumab Medications, Healthcare Resource Utilization, and Costs
In the follow-up sample at baseline, almost all patients were using at least one acute and one preventive migraine medication prior to the erenumab index date with most patients using two or more (Table 4). The majority of patients were using triptans and a substantial proportion were using NSAID, opioids, or other analgesics, with few using ergotamines. During baseline, the most commonly used preventive migraine medications were anticonvulsants and antidepressants, followed by antihypertensives and botulinum toxin. Use of acute and preventive medications in the baseline period was similar for persistent and non-persistent patients with the exception of triptans where 72.4% of persistent patients were using triptans in the pre-index period versus 65.6% of non-persistent patients.
Table 4.
Patient use of acute and preventive migraine treatment in the 6 months pre- and post-erenumab stratified by persistence
| Pre-erenumab | Post-erenumab | |||||
|---|---|---|---|---|---|---|
| All (N = 4437) | Erenumab persistent (N = 1853) | Erenumab non-persistent (N = 2584) | All (N = 4437) | Erenumab persistent (N = 1853) | Erenumab non-persistent (N = 2584) | |
| Number of acute medications used, n (%) | ||||||
| 0 | 402 (9.1) | 146 (7.9) | 256 (9.9) | 610 (13.7) | 242 (13.1) | 368 (14.2) |
| 1 | 1294 (29.2) | 554 (29.9) | 740 (28.6) | 1373 (30.9) | 593 (32.0) | 780 (30.2) |
| 2 | 114 (25.8) | 488 (26.3) | 656 (25.4) | 1077 (24.3) | 446 (24.1) | 631 (24.4) |
| 3 | 777 (17.5) | 332 (17.9) | 445 (17.2) | 674 (15.2) | 284 (15.3) | 390 (15.1) |
| 4 | 404 (9.1) | 153 (8.3) | 251 (9.7) | 355 (8.0) | 162 (8.7) | 193 (7.5) |
| 5+ | 416 (9.4) | 180 (9.7) | 236 (9.1) | 348 (7.8) | 126 (6.8) | 222 (8.6) |
| Type of acute medication used, n (%) | ||||||
| Triptans | 3037 (68.4) | 1342 (72.4) | 1695 (65.6) | 2712 (61.1) | 1196 (64.5) | 1516 (58.7) |
| Ergotamines | 214 (4.8) | 90 (4.9) | 124 (4.8) | 168 (3.8) | 65 (3.5) | 103 (4.0) |
| NSAIDs | 1646 (37.1) | 685 (37.0) | 961 (37.2) | 1517 (34.2) | 627 (33.8) | 890 (34.4) |
| Opioids | 1682 (37.9) | 680 (36.7) | 1002 (38.8) | 1699 (38.3) | 688 (37.1) | 1011 (39.1) |
| 2+ opioid claims | 1150 (25.9) | 457 (24.7) | 693 (26.8) | 1125 (25.4) | 456 (24.6) | 669 (25.9) |
| Analgesicsa | 825 (18.6) | 329 (17.8) | 496 (19.2) | 741 (16.7) | 280 (15.1) | 461 (17.8) |
| Number of preventive medications used, n (%) | ||||||
| 0 | 564 (12.7) | 235 (12.7) | 329 (12.7) | 878 (19.8) | 402 (21.7) | 476 (18.4) |
| 1 | 1131 (25.5) | 515 (27.8) | 616 (23.8) | 1235 (27.8) | 528 (28.5) | 707 (27.4) |
| 2 | 1195 (26.9) | 491 (26.5) | 704 (27.2) | 1110 (25.0) | 461 (24.9) | 649 (25.1) |
| 3 | 804 (18.1) | 312 (16.8) | 492 (19.0) | 673 (15.2) | 260 (14.0) | 413 (16.0) |
| 4 | 451 (10.2) | 186 (10.0) | 265 (10.3) | 334 (7.5) | 138 (7.4) | 196 (7.6) |
| 5+ | 292 (6.6) | 114 (6.2) | 178 (6.9) | 207 (4.7) | 64 (3.5) | 143 (5.5) |
| Type of preventive medication used, n (%) | ||||||
| Anticonvulsant | 2249 (50.7) | 908 (49.0) | 1341 (51.9) | 1863 (42.0) | 735 (39.7) | 1128 (43.7) |
| Antidepressant | 2206 (49.7) | 879 (47.4) | 1327 (51.4) | 2001 (45.1) | 779 (42.0) | 1222 (47.3) |
| Antihypertensive | 1540 (34.7) | 642 (34.6) | 898 (34.8) | 1335 (30.1) | 542 (29.2) | 793 (30.7) |
| Botulinum toxin | 1302 (29.3) | 575 (31.0) | 727 (28.1) | 1048 (23.6) | 452 (24.4) | 596 (23.1) |
| Anti-CGRP (excluding erenumab) | 0 (0.0) | 0 (0) | 0 (0) | 552 (12.4) | 43 (2.3) | 509 (19.7) |
| Otherb | 744 (16.8) | 277 (14.9) | 467 (18.1) | 690 (15.6) | 249 (13.4) | 441 (17.1) |
| Associated costs, mean (SD) | ||||||
| Acute migraine treatment costs | $1206 ($8164) | $1463 ($12,298) | $1021 ($2437) | $1054 ($7214) | $1245 ($10,845) | $916 ($2235) |
| Preventive migraine treatment costs (excluding anti-CGRPs) | $1321 ($2028) | $1346 ($2034) | $1303 ($2024) | $1124 ($1959) | $1120 ($1908) | $1127 ($1995) |
| Total acute and preventive migraine treatment costs (excluding anti-CGRPs) | $2526 ($8453) | $2809 ($12,470) | $2324 ($3334) | $2178 ($7540) | $2365 ($11,028) | $2043 ($3225) |
| Anti-CGRP costs (excluding erenumab) | $0 ($0) | $0 ($0) | $0 ($0) | $171 ($6) | $24 ($2) | $276 ($7) |
aNon-NSAID/non-opioid analgesics
bOther agents prescribed for migraine prevention such as memantine, milnaciprin, etc.
The use of these acute medications declined from pre-index to post-index period (Table 4). There was a greater percentage of patients using zero or one acute medication in the post-index period compared to the pre-index period (44.6% vs 38.3%) with an accompanying reduction in the percentage of patients using each class of medication except for opioids which increased slightly from 37.9% pre-index to 38.3% post-index. Closer inspection of these results using subgroup analyses for patients who were persistent and non-persistent at 6 months revealed a small difference in the post-index proportion of patients on zero or one acute medication, increasing by 7.3% (from 37.8% to 45.1%) for persistent patients and 5.9% (from 38.5% to 44.4%) for non-persistent patients. Similar small differences were observed between persistent and non-persistent patients and the use of triptans, decreasing by 7.9% (from 72.4% to 64.5%) for persistent patients and 6.9% (from 65.6% to 58.7%) for non-persistence patients. The slight increase in opioid use was consistent for persistent, increasing by 0.4% (from 36.7% to 37.1%), and non-persistent patients, increasing by 0.3% (from 38.8% to 39.1%). For preventive medications, the proportion of patients using anticonvulsants in the post-erenumab period decreased by 9.3% (from 49.0% to 39.7%) for persistent patients and 8.2% (from 51.9% to 43.7%) for non-persistent patients.
Differences in the costs associated with acute medications aligned with the overall reduction in use between pre- and post-erenumab index periods. Overall, a minor reduction in costs was observed from an average of $1206 in the 6-month period pre-index versus $1054 in the 6-month post-erenumab period. Again, closer inspection of cost results by persistent and non-persistent patients identified important differences. Reflective of higher use of triptans in the pre-index period, costs for acute medications in the pre-index period were $1463 for persistent patients and $1021 for non-persistent patients. In the post-index period, reductions in costs were slightly higher for persistent patients (− $218, $1245 post versus $1463 pre) than non-persistent patients (− $105, $916 post versus $1021 pre).
As was observed for acute medications, a similar picture of overall reduction from pre- to post-erenumab periods was seen with the preventive medications used prior to erenumab (Table 4). There was a greater percentage of patients using zero or one preventive medication in the post-index period (47.6% post versus 38.2% pre) with an accompanying reduction in the percentage of patients using each class of medication. These reductions were slightly higher for persistent versus non-persistent patients.
With these changes, the average costs for preventive medication decreased slightly between pre- and post-erenumab periods with slightly higher reductions for persistent than non-persistent patients; − $226 ($1120 post versus $1346 pre) for persistent patients and − $176 ($1127 post versus $1303 pre) for non-persistent patients. Importantly, we noted some concomitant use of non-erenumab CGRP pathway antagonists in the persistent group (n = 43 or 2.3%), but also switching to a non-erenumab CGRP pathway antagonist in non-persistent patients (n = 509 or 19.7%). We completed a sensitivity analysis excluding patients prescribed non-erenumab CGRP pathway antagonists from the analysis and noted similar patterns of change as the primary analysis for both acute and preventive medications. As expected, removing a larger number of non-erenumab CGRP pathway antagonists from the non-persistent subgroup resulted in a reduction in associated costs for preventive medications in the subgroup (data not shown).
Aligned to overall reductions in the use of acute and preventive medications in the pre-index period, total costs for acute and preventive medications were reduced with slightly higher reductions in costs for persistent (− $444, $2365 post versus $2809 pre) versus non-persistent (− $281, $2043 post versus $2324 pre) patients.
The use of inpatient, ER, and imaging resource was low in the 6-month baseline period, for example 0.1 hospitalizations and 0.5 ER visits on average (Table 5). The most used resource was outpatient office visits, with a mean 8.5 visits in the 6-month pre-erenumab period. Associated costs ranged from a mean (SD) of $170 ($606) for imaging to $1680 ($9995) for inpatient hospital admissions contributing to a total cost of $3651 ($11,736) across non-medication healthcare resources.
Table 5.
Patient use of healthcare utilization and associated costs in the 6 months pre- and post-erenumab stratified by persistence
| Pre-erenumab | Post-erenumab | |||||
|---|---|---|---|---|---|---|
| All (N = 4437) | Erenumab persistent (N = 1853) | Erenumab non-persistent (N = 2584) | All (N = 4437) | Erenumab persistent (N = 1853) | Erenumab non-persistent (N = 2584) | |
| Inpatient hospitalizations | ||||||
| Mean (SD) | 0.1 (0.3) | 0.1 (0.3) | 0.1 (0.3) | 0.1 (0.3) | 0.0 (0.2) | 0.1 (0.3) |
| 95% CI | (0.1, 0.1) | (0.0, 0.1) | (0.1, 0.1) | (0.0, 0.1) | (0.0, 0.1) | (0.1, 0.1) |
| Length of stay, days | ||||||
| Mean (SD) | 5.2 (3.9) | 4.6 (2.5) | 5.6 (4.5) | 5.2 (4.3) | 4.4 (3.2) | 5.5 (4.7) |
| 95% CI | (4.7, 5.7) | (4.0, 5.1) | (4.9, 6.3) | (4.6, 5.8) | (3.7, 5.2) | (4.7, 6.3) |
| Mean (SD) costs | $1680 ($9995) | $1403 ($8316) | $1879 ($11,040) | $1696 ($19,939) | $957 ($6515) | $2227 ($25,527) |
| ER visits | ||||||
| Mean (SD) | 0.5 (2.2) | 0.5 (2.9) | 0.4 (1.4) | 0.4 (1.4) | 0.4 (1.4) | 0.4 (1.4) |
| 95% CI | (0.4, 0.5) | (0.4, 0.6) | (0.4, 0.5) | (0.3, 0.4) | (0.3, 0.4) | (0.4, 0.5) |
| Mean (SD) costs | $657 ($3581) | $636 ($4227) | $673 ($3035) | $533 ($2401) | $500 ($2738) | $557 ($2127) |
| Outpatient office visits | ||||||
| Mean (SD) | 8.5 (7.9) | 8.2 (7.5) | 8.7 (8.2) | 8.0 (7.8) | 7.7 (7.2) | 8.3 (8.2) |
| 95% CI | (8.2, 8.7) | (7.9, 8.5) | (8.4, 9) | (7.8, 8.2) | (7.3, 8.0) | (8.0, 8.6) |
| Mean (SD) costs | $1143 ($1951) | $1128 ($2381) | $1154 ($1572) | $1049 ($1473) | $1024 ($1676) | $1067 ($1308) |
| Brain imaginga studies | ||||||
| Mean (SD) | 0.2 (0.5) | 0.2 (0.5) | 0.2 (0.5) | 0.1 (0.4) | 0.1 (0.4) | 0.1 (0.4) |
| 95% CI | (0.2, 0.2) | (0.1, 0.2) | (0.2, 0.2) | (0.1, 0.1) | (0.1, 0.1) | (0.1, 0.2) |
| Mean (SD) costs | $170 ($606) | $166 ($643) | $173 ($577) | $110 ($503) | $84 ($459) | $129 ($531) |
| Total associated costsb mean (SD) | $3651 ($11,736) | $3334 ($10,921) | $3879 ($12,285) | $3389 ($20,457) | $2564 ($7786) | $3980 ($25,969) |
a(CT + MRI + Nuclear Med + Angiography)
bTotal associated costs include inpatient, ER, office visits, and brain imaging studies
For the overall population, only minor differences in the use and associated costs for non-medication healthcare resources were observed in the 6 months post-erenumab compared to the 6 months pre-erenumab (Table 5). Closer inspection of differences in the use and associated costs for these resources pre-index by persistent and non-persistent patients revealed a slightly longer mean length of stay during inpatient hospitalizations and higher use of outpatient visits for non-persistent versus persistent patients: 5.6 days versus 4.6 days and 8.7 visits versus 8.2 visits, respectively. Aligned to this, the associated pre-index costs for these services were higher in non-persistent versus persistent patients. Post-index reductions in the use of resources were slightly more pronounced for persistent patients with the exception of brain imaging studies which were almost identical for persistent and non-persistent patients. For costs estimates, a notable finding was the substantial uncertainty (large standard deviation relative to the mean) in post-index inpatient hospitalization cost estimates for non-persistent patients, suggesting that a few outlier patients with higher costs likely contributed to higher mean post-index costs. Acknowledging this, and that the uncertainty in cost estimates overall was substantial, there remained a pattern of greater reductions in costs for persistent patients compared to non-persistent patients. This contributed to an overall reduction in total costs of − 23% for persistent patients ($3334 pre-index to $2564 post-index) and a slight increase in total costs of 3.0% for non-persistent patients ($3879 pre-index to $3980 post-index) with the caveat that this increase will be influenced by the skewed cost estimate for inpatient hospitalization. Of note, this pattern was still observed in our non-erenumab CGRP pathway antagonist censored sensitivity analysis (data not shown).
Discussion
We completed a retrospective, claims-based analysis designed to understand the real-world impact of erenumab as a migraine preventive including changes in migraine characteristics, comorbidities, healthcare resource utilization, and associated costs. In the overall population of patients, the average age of the predominately female population was 46 years. Comorbid non-migraine headaches, hypertension, anxiety, and depression were common. Almost two-thirds of patients had chronic migraine. Demographic and migraine characteristics were similar in the 6-month follow-up sample where we observed reductions in claims for migraine with aura, chronic migraine, menstrual migraine, and status migrainosus. Similarly claims for comorbidities of hypertension, anxiety, depression, and non-migraine headaches were reduced in the post-erenumab period. There was a shift to decreased use of acute and preventive medications. Reductions in the overall use and associated cost of healthcare resources such as inpatient hospitalization and outpatient office visits were minimal. Reductions that were observed in the 6-month follow-up period were generally more pronounced for the subgroup of patients that were persistent to erenumab. A notable exception was observed in the use of opioids which increased slightly overall but consistently for persistent and non-persistent patients. We also note some potentially important differences in erenumab prescribing patterns between persistent and non-persistent patients that suggest dose titration may have a role to play in persistence.
The main strengths of our study include the relatively large sample size of patients initiating preventive treatment for migraine with erenumab, the comprehensive capture of healthcare resource utilization in the MarketScan database, and the pre–post design. There are also inherent weaknesses with claims-based analyses. Use of ICD-10-CM codes and NDC/HCPCS codes to identify comorbid conditions and concomitant medications, respectively, may misclassify comorbid conditions because of sensitivity and specificity of ICD-10-CM code algorithms. Although we can assess use of prescription medications, we cannot assess use of over-the-counter prescription medications used for migraine treatments including non-prescription NSAIDs and other supplements. No data on migraine frequency or severity were available nor were reasons for non-persistence. Our results also represent patients predominantly younger than 65 years of age with commercial health insurance. Results may not be representative of the full US population. A high level of variability was observed in all cost estimates, limiting comparisons between time periods and subgroups.
Our results are generally consistent with other early real-world use reports. As expected, our erenumab patient profile was almost identical to that observed in a preliminary analysis of 1181 patient also based on MarketScan® Early View data [13] and similar to the profile observed in early analysis of other databases [14, 15]. A preliminary review of changes in acute migraine-specific medications pre- and post-erenumab initiation using pharmacy data also observed reductions in acute medication use post-erenumab [16]. Using a combined electronic health record and claims database, Tepper and colleagues [17] observed significant reductions in migraine-specific acute medication use and healthcare resource utilization. Reporting interim results from a retrospective chart-review study with select US headache centers and focusing on clinical versus medication use outcomes, Faust and colleagues [18] observed reductions in the mean number of migraine/headache days per month and the average duration of migraine/headache attacks, but also noted the continued use of a polypharmacy approach to management in their chronic migraine population. Similarly, on the basis of data from headache centers in Italy, Ornello and colleagues [19] observed reductions in monthly migraine days, median monthly days of analgesic use, and median monthly days of triptan use.
We note that 6-month persistence was less than 50%. Although this is higher than claims-based analysis for classes of non-migraine specific medications traditionally used for migraine prevention (around 34% at 6 months for beta-blockers, tricyclic antidepressants, and topiramate with a 90-day gap to define persistence) [20], it is still lower than ideal; however, the context of our analysis is important. The time frame for our analysis covers a period of initial approval and use of a totally new class of medications and the associated evolution of understanding of this new class and the place of the class in migraine prevention. Although there was an overall reduction in the use of acute medications, preventive medications, and healthcare resource utilization for migraine in the 6-month post-erenumab period, it is important to consider the differences observed between persistent and non-persistent patients. For acute medications, the higher use of more expensive triptans in the post-erenumab period by erenumab-persistent patients was aligned with a small increase in associated costs. At the same time, increased use of less expensive opioids in non-persistent patients was aligned with a decrease in costs. As has been shown in more recent research, patients using opioids have greater overall healthcare costs compared to patients not using opioids—up to double for patients using the highest doses of opioids [21]. Thus, the trade-off between increased use of a more expensive migraine-specific acute medication and the reduced use of a less expensive but more costly overall non-migraine-specific and addictive pain medication may be more than justified.
Limitations
This was a study of medical claims from US-centric databases. The findings from this study are specific to the US healthcare system and may not be generalizable to other countries. Claims data lack information on frequency or severity of migraine attacks and do not distinguish among migraine subtypes. Although there is a specific ICD-10-CM code for chronic migraine (G43.7), there is no similar code for episodic migraine. This limits the ability to stratify patients by migraine category, thus potential imbalances in healthcare resource utilization between patients with varying degrees of migraine severity may make the estimates calculated less precise. However, focusing the analysis on all migraine patients is consistent with previous studies and the US prescribing information for erenumab. Additional research with adjudicated claims, longer follow-up time, and larger sample size will likely support a better characterization of the potential economic benefits of erenumab.
Conclusion
In this claims-based, retrospective observational study we observed reductions in claims for important migraine characteristics, comorbidities, and a shift to decreased use of acute and preventive migraine medications in the post-erenumab follow-up period—observations indicative of the real-world effectiveness of erenumab. Further examination is required as persistence to erenumab, which may be influenced by dose change, appears to be an important factor in changes to healthcare resource utilization and costs.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
Funding
The study and the journal’s Rapid Service Fee were funded by Amgen Inc.
Medical Writing and Editorial Assistance
Medical writing and editorial assistance for this article was provided by Jon Nilsen, PhD (Amgen Inc.) funded by Amgen Inc. The authors would like to thank Robert Urman and Soeren Rasmussen (both of Amgen Inc.) for their insightful comments in reviewing the manuscript.
Authorship
All named authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship for this article, take responsibility for the integrity of the work as a whole, and have given their approval for this version to be published.
Author Contributions
David Chandler: Concept and design, interpretation of data, drafting of manuscript. Christine Szekely: Concept and design, interpretation of data, drafting of manuscript. Shivani Aggarwal: Statistical analysis, drafting of the manuscript. Lori Cyprien: Statistical analysis, drafting of the manuscript. Mark Bensink: Concept and design, interpretation of data, drafting of manuscript.
Disclosures
David Chandler, Christine Szekely, Shivani Aggarwal, Lori Cyprien, and Mark Bensink are/were employees and stockholders of Amgen Inc. The journal’s Rapid Service Fee was funded by Amgen Inc. Christine Szekely current affiliation Brigham and Women's Hospital, Boston, MA, USA. Mark Bensink current affiliation is Benofit Consulting PTY LTD, Brisbane, Australia.
Compliance with Ethics Guidelines
MarketScan® databases are in compliance with the Health Information Portability and Accountability Act of 1996 (HIPAA). Data from the MarketScan® database were used under a licensing agreement with IBM Watson Health. As this study did not involve the collection, use, or transmittal of individually identifiable data, it did not require institutional review board review or approval.
Data Availability
In alignment with the IBM Watson Health licensing agreement, the dataset generated and analyzed during the current study is not publicly available.
References
- 1.Lipton RB, Bigal ME, Diamond M, et al. Migraine prevalence, disease burden, and the need for preventive therapy. Neurology. 2007;68:343–349. doi: 10.1212/01.wnl.0000252808.97649.21. [DOI] [PubMed] [Google Scholar]
- 2.Blumenfeld AM, Varon SF, Wilcox TK, et al. Disability, HRQoL and resource use among chronic and episodic migraineurs: results from the International Burden of Migraine Study (IBMS) Cephalalgia. 2011;31:301–315. doi: 10.1177/0333102410381145. [DOI] [PubMed] [Google Scholar]
- 3.Hazard E, Munakata J, Bigal ME, et al. The burden of migraine in the United States: current and emerging perspectives on disease management and economic analysis. Value Health. 2009;12:55–64. doi: 10.1111/j.1524-4733.2008.00404.x. [DOI] [PubMed] [Google Scholar]
- 4.Lanteri-Minet M, Duru G, Mudge M, et al. Quality of life impairment, disability and economic burden associated with chronic daily headache, focusing on chronic migraine with or without medication overuse: a systematic review. Cephalalgia. 2011;31:837–850. doi: 10.1177/0333102411398400. [DOI] [PubMed] [Google Scholar]
- 5.Chessman AW. ACP journal club. Review: approved and some off-label preventive drugs reduce migraine frequency in episodic migraine. Ann Intern Med. 2013;159:JC11. doi: 10.7326/0003-4819-159-8-201310150-02011. [DOI] [PubMed] [Google Scholar]
- 6.Starling AJ, Vargas BB. A narrative review of evidence-based preventive options for chronic migraine. Curr Pain Headache Rep. 2015;19:49. doi: 10.1007/s11916-015-0521-0. [DOI] [PubMed] [Google Scholar]
- 7.Hepp Z, Bloudek LM, Varon SF. Systematic review of migraine prophylaxis adherence and persistence. J Manag Care Pharm. 2014;20:22–33. doi: 10.18553/jmcp.2014.20.1.22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hepp Z, Dodick DW, Varon SF, et al. Adherence to oral migraine-preventive medications among patients with chronic migraine. Cephalalgia. 2015;35:478–488. doi: 10.1177/0333102414547138. [DOI] [PubMed] [Google Scholar]
- 9.Hepp Z, Dodick DW, Varon SF, et al. Persistence and switching patterns of oral migraine prophylactic medications among patients with chronic migraine: a retrospective claims analysis. Cephalalgia. 2017;37:470–485. doi: 10.1177/0333102416678382. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Buse DC, Yugrakh MS, Lee LK, et al. Burden of illness among people with migraine and >/= 4 monthly headache days while using acute and/or preventive prescription medications for migraine. J Manag Care Spec Pharm. 2020;26:1334–1343. doi: 10.18553/jmcp.2020.20100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kawata AK, Shah N, Poon J-L, et al. Understanding the migraine treatment landscape prior to the introduction of calcitonin gene-related peptide inhibitors: results from the assessment of tolerability and effectiveness in MigrAINe patients using preventive treatment (ATTAIN) study. Headache J Head Face Pain. 2021 doi: 10.1111/head.14053. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Leslie SR, Gwadry-Sridhar F, Thiebaud P, et al. Calculating medication compliance, adherence and persistence in administrative pharmacy claims databases. Pharm Program. 2008;1:13–19. doi: 10.1179/175709208X334614. [DOI] [Google Scholar]
- 13.Chia V, Szekely C, Bensink M, et al. Characteristics of early users of erenumab in a large US administrative claims database P268LB. Headache. 2019;59:125–126. [Google Scholar]
- 14.Hines DM, Shah S, Multani JK, et al. Erenumab prescription early view: patient characteristics, treatment patterns and medication adherence in the United States P270LB hines. Headache. 2019;59:199–200. doi: 10.1111/head.14068. [DOI] [PubMed] [Google Scholar]
- 15.Bogdanov A, Chia V, Szekely C, et al. Early use of erenumab in US real world practice P268LB. Headache. 2019;59:25158163211020419. [Google Scholar]
- 16.Hines D, Shah S, Multani J, et al. Changes in acute migraine-specific medications after initiating erenumab: results from a real-world retrospective cohort study in the United States (1685). Neurology. 2020;94:1685.
- 17.Tepper S, Fang J, Vo P, et al. Impact of erenumab on acute medication usage and healthcare resource utilization among migraine patients: a US claims database study. Cephalalgia. 2020;40:97–99. doi: 10.1186/s10194-021-01238-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Faust E, Pivneva I, Yang K, et al. Reductions in migraine frequency and duration in patients with chronic migraine treated with erenumab: interim results from a real-world multicenter chart-review study of US headache centers MTV20-DP-001. Cephalalgia. 2020;40:18–110. [Google Scholar]
- 19.Ornello R, Casalena A, Frattale I, et al. Real-life data on the efficacy and safety of erenumab in the Abruzzo region, central Italy. J Headache Pain. 2020;21:32. doi: 10.1186/s10194-020-01102-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Woolley JM, Bonafede MM, Maiese BA, et al. Migraine prophylaxis and acute treatment patterns among commercially insured patients in the United States. Headache. 2017;57:1399–1408. doi: 10.1111/head.13157. [DOI] [PubMed] [Google Scholar]
- 21.Kangethe A, Polson M, Evangelatos TM, et al. Real-world assessment of concomitant opioid utilization and associated trends in patients with migraine. Am J Manag Care. 2020;26:S8–S14. doi: 10.37765/ajmc.2020.42544. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
In alignment with the IBM Watson Health licensing agreement, the dataset generated and analyzed during the current study is not publicly available.