Healthcare resource utilization following switch or discontinuation in multiple sclerosis patients on disease modifying drugs

Abstract

Objective: The objective of this study was to explore the cost and utilization in the period following discontinuations or switches of disease modifying drugs (DMDs) for patients with multiple sclerosis (MS). Secondary objectives included an assessment of the time to switch or discontinuation from index DMD treatment.

Methods: Cases were defined as a billed MS diagnosis in continuously enrolled patients initiated with interferon-β1a IM, interferon-β1b SC, glatiramer acetate, and interferon-β1a SC found in the PharMetrics Patient-Centric Database. Information on patient demographics, diagnoses, procedures, pharmacy-dispensed drugs, and costs was extracted; reasons for discontinuation and expenses outside of the healthcare system were not available. Treatment discontinuations and switches between study drugs were defined using pharmacy prescription patterns and analyzed by descriptive and regression methods. The non-pharmacy medical costs in the 18 months following switching or discontinuation were compared to the costs in a randomly selected similar period for those patients who did not switch or discontinue these agents.

Results: A total of 5,772 MS patients were continuously enrolled and were treated with one or more of the four drugs of interest, and about half of these patients switched drugs or discontinued treatment for at least 90 days. Patients initiated with interferon-β1b SC were more likely to discontinue treatment compared to interferon-β1a IM users. Non-pharmaceutical medical costs were highest for those switching treatments followed by those discontinuing DMDs in the 18 months following a switch or discontinuation, compared to persistent users of these drugs. Interferon β1b SC initiators had higher costs following changes or discontinuations, while glatiramer acetate and interferon-β1a SC users had lower subsequent costs compared to interferon-β1a IM users.

Limitations: Unfortunately, the reasons for stopping the initial treatment cannot be determined from analysis of an administrative claims database. Also, the MS cases followed in this analysis are billing diagnostic events unconfirmed through a review of medical records or other data sources. The results are unstratified in terms of severity and thus while treatment patterns may vary for patients with different types of MS (e.g., progressive vs. relapsing-remitting), this cannot be examined in this analysis.

Conclusion: Changing or discontinuing DMDs is common among MS patients and is associated with higher non-pharmaceutical medical costs that vary based on the initiating drug and other demographics characteristics.

Key words::

• costs
• costs analysis
• interferon-beta
• patient compliance

Introduction

Multiple sclerosis (MS) is an incurable, progressively disabling, neurodegenerative disease that affects the brain, spinal cord, and optic nerves. The estimated prevalence of MS in the US is between 350,000 and 450,000, with approximately 12,000 new cases diagnosed each year¹. Among privately insured populations it has been measured at rates of 24 per 10,000 enrollees, and there is a higher prevalence in the Medicare and Medicaid populations (36 and 71 per 10,000 enrollees, respectively)².

MS is an expensive disease because it requires continuous treatment and has an early onset. The lifetime costs per patient in the US have been estimated at $2.2 million (in 1994 US$), and the overall annual cost, between $6.8 and 11.9 billion³. The average annual cost in 2004 was $12,879, with almost two-thirds of this from prescriptions⁴. Costs increase with disease progression primarily due to increasing disability⁵.

The long-term goal of treatment is to slow disease progression with disease modifying drugs (DMDs). These agents are the co-polymer glatiramer acetate (Copaxone), and three forms of beta interferon (interferon-β1a IM [Avonex^†], interferon-β1b SC [Betaseron^‡], and interferon-β1a SC [Rebif^§]. Current treatment guidelines recommend initiating and maintaining one of the four approved first line DMDs as soon as a definite diagnosis of relapsing-remitting MS is confirmed, or if a single attack is highly suggestive of MS^6,7. There is a high inter-patient variation in treatment response and patient compliance to these drugs. The long-term efficacies of DMDs are not fully known. The guidelines offer latitude to neurologists in choosing between any one of these agents based on their relative efficacy, potential for generating self-neutralizing antibodies, and the patient's expected compliance with the treatment plan, because these drugs require one or more injections per week^8,9.

Adequate clinical benefit depends on disease severity and adhering to both the dose and the administration schedule of DMDs. Disease progression is evaluated in terms of the relapse rate and worsening disability. Disease progression or intolerance to the side-effects of DMDs are recognized reasons to switch medications, although there is no consensus on the specific criteria for evaluating a response to treatment^10–12. It is clinically plausible that a switch of therapy may affect disease control because DMDs take a period of time to become fully effective (3–6 months after initiation for interferons and up to 9 months for glatiramer acetate)⁹. In addition, patients treated with interferons may experience a transient period of increased disability before they respond to a new drug¹³.

A review by Clerico et al found that 17–41% of MS patients ultimately discontinued DMD therapy, with most of the discontinuations occurring within the first 2 years of treatment¹⁴. The review also found that 30–50% of discontinuations were reportedly due to a lack of efficacy, while 22–70% were due to side-effects. The main side-effects reported included injection site reactions, flu-like symptoms, and depression. A recent survey of MS patients found that 75% of patients switched their DMD drug at least once, 11% changed twice, and 14% switched three or more times¹⁵.

To the authors' knowledge, there are no studies that examined the costs and resource utilization following switches or discontinuations in DMDs. Because persistent treatment is highly associated with effective treatment of a disease, it is a clinical goal to choose the right treatments for patients and to maintain therapy long-term. Medication discontinuation and switching is hence indicative of ineffective or less effective treatment in real world patients and hence there will likely be associated medical costs. The primary objective of this study was to explore the cost and utilization in the period following discontinuations or switches of DMD therapy for patients with MS. Secondary objectives included an assessment of the time to switch or discontinuation from index DMD treatment.

Methods

Data source

This study utilized the PharMetrics Patient-Centric Database, which constitutes an integrated set of fully adjudicated medical and pharmaceutical claims for all covered services. It includes both inpatient and outpatient diagnoses and procedures, and both retail pharmacy and mail order prescription records. This dataset includes longitudinal, integrated, patient-level medical and pharmaceutical claims histories for more than 45 million commercially-insured patients over 85 US health plans. The national breadth of the PharMetrics Patient-Centric Database allows it to serve as a critical tool for national and regional benchmarking of resource utilization. The depth of the PharMetrics Patient-Centric Database allows for comparisons across a range of patient demographics, including age, gender, treatment patterns, and comorbidities.

Patient identification

Data were extracted for all patients in the PharMetrics database with a diagnosis of MS (ICD-9 code 340) from January 1, 1996 through June 30, 2005. The extracted data include patient demographic and enrollment characteristics; all diagnoses, procedures, and pharmacy-dispensed drugs; and all associated costs. Patients were required to have at least two years of continuous enrollment in the health plan, with gaps in enrollment of no more than 2 months allowed, to qualify for the study.

Exposure definitions

Using the full dataset of all MS patients, the authors determined all MS treatments given throughout each patient's duration in the health plan database. Patients were classified as initiators of one of the four study drugs– interferon-β1a IM, interferon-β1b SC, glatiramer acetate, or interferon-β1a SC – with initiation defined as a first pharmacy dispensing or injection procedure for that drug following at least 6 months of enrollment in the health plan database, with no record of the same drug at any prior time, and with at least 18 months of enrollment in the database after drug initiation. Enrollment gaps of up to 2 months during these windows were permitted. Patients who qualified as initiators of more than one study drug were classified as initiators of the drug started earliest. Individuals who did not qualify as initiators of any study drug were retained for analysis in the full MS patient population only.

Definition of baseline characteristics

Baseline characteristics of the full MS patient population were defined based on the 6 months immediately preceding first MS diagnosis. For those who received an MS diagnosis during the first 6 months enrolled, baseline characteristics were obtained based on the first 6 months of enrollment. Baseline characteristics for each drug initiator group were calculated based on the 6 months immediately preceding study drug initiation (the baseline period). The authors classified patients with respect to age, gender, region of the US, type of health plan, and prior treatment with drugs for MS during the baseline period. For the prior MS drug exposure, in addition to examining use of the specific study drugs, use of natalizumab was also included. Natalizumab is also indicated for treatment of MS but was infrequently used in this population and so was not included as one of the primary study drugs of interest.

Outcome ascertainment

Patients were followed until disenrollment from the health plan or database cutoff (December 31, 2006) for each outcome. Treatment discontinuation was defined as a gap in exposure to the initial MS drug of at least 90 days (e.g., a gap of at least 90 days following the date when the medication should next have been dispensed or administered). The date of discontinuation was calculated as the date when the medication should next have been dispensed or administered plus 90 days. A switch was defined as a discontinuation of the initial drug, with the start of a new MS medication occurring after the date of last dispensing or administration of the initial drug, and no later than the date of discontinuation of the initial drug. Time to switch/discontinuation was calculated as the date of switch or discontinuation minus the start date of the initial drug.

Medical costs associated with switching/discontinuing were calculated based on the 18 months following first switch or discontinuation, and on a randomly selected 18-month period after study drug initiation for those who did not switch or discontinue. For patients who had less than 18 total months of follow-up after switch or discontinuation, the costs were extrapolated to an 18-month total (e.g., costs for a patient with 9 months of follow-up post-switch or discontinuation would have been doubled). Medical costs associated with MS (i.e., with an MS diagnosis on the claim) and total medical costs regardless of diagnosis were summed for each patient and were broken out into outpatient, emergency room, inpatient, MRI, and pharmacy drug costs. All costs were adjusted to 2006 US dollars.

Analysis

The authors examined each baseline characteristic for the full MS patient population and separately for initiators of interferon-β1a IM, interferon-β1b SC, glatiramer acetate, and interferon-β1a SC. Time to first treatment switch and time to discontinuation of all MS treatment were compared among drug groups using Cox proportional hazards regression models, censoring patients without an event at disenrollment or study end. Costs associated with switching or discontinuing treatment, in categories of total, outpatient, MRI, emergency room, inpatient, and drug were tabulated for those who switched or discontinued and those who remained on the initial treatment, with predictors of total medical costs identified using linear regression analyses, adjusting for relevant baseline characteristics.

Results

The search for patients with a diagnosis of MS and at least 2 years of enrollment with no more than a 2-month gap during enrollment identified a total of 27,513 patients. Applying the criteria for identifying new drug initiators, the authors found 2,019 patients initiating interferon-β1a IM, 576 on interferon-β1b SC, 2,081 on glatiramer acetate, and 1,096 on interferon-β1a SC. The remaining 21,741 patients did not qualify as initiators of one of the study drugs but were followed as part of the full MS cohort.

Table 1 presents the baseline characteristics for the full MS cohort and each of the four drug initiator groups. Ages were fairly similar between groups, and the only significant gender difference was a lower proportion of females among the interferon-β1b SC initiators compared to the interferon-β1a IM group. However, considerable differences can be seen between the interferon-β1a IM initiators and initiators of each of the three comparator drugs in other characteristics. Regional differences were found across all drug pairs, with more interferon-β1b SC initiators in the Midwest, more glatiramer acetate users in the West, and more interferon-β1a SC users in the South and West compared to the distribution of interferon-β1a IM initiators. Interferon β1a SC initiators more often had a health maintenance organization providing their health insurance plan, glatiramer acetate initiators a point of service plan, and interferon-β1a SC initiators a preferred provider organization, compared to the interferon-β1a IM group. Prior treatment with any other MS drug was far more common in each of the comparator groups than among the interferon-β1a IM initiators, indicating that interferon-β1a IM was the most likely drug to be used as first-line treatment.

Table 1. Baseline characteristics of multiple sclerosis population.

Patient characteristic	All MS patients(n = 27,513)		Interferon-β1a IM(n = 2,019)		Interferon-β1b SC(n = 576)		Glatiramer acetate(n = 2,081)		Interferon-β1a SC(n = 1,096)
	Mean	SD	Mean	SD	Mean	SD	Mean	SD	Mean	SD
Age, years (mean, SD)	43.23	9.73	43.39	9.82	43.22	9.73	43.38	9.51	42.67	9.94
	n	%	n	%	n	%	n	%	n	%
Age, categorized (n, %)
0–17	28	0.49	13	0.23	3	0.05	8	0.14	4	0.07
18–24	172	2.98	62	1.07	18	0.31	45	0.78	47	0.81
25–34	940	16.29	306	5.30	93	1.61	347	6.01	194	3.36
35–44	1,869	32.38	671	11.63	172	2.98	696	12.06	330	5.72
45–54	2,038	35.31	707	12.25	222	3.85	718	12.44	391	6.77
55–64	725	12.56	260	4.50	68	1.18	267	4.63	130	2.25
65+	—	—	—	—	—	—	—	—	—	—
Gender (n, %)
Male	1,262	21.86	427	7.40	154	2.67	424	7.35	257	4.45
Female	4,510	78.14	1,592	27.58	422	7.31	1,657	28.71	839	14.54
Geographic region (n, %)
East	1,028	17.81	368	6.38	91	1.58	409	7.09	160	2.77
South	1,120	19.40	377	6.53	98	1.70	393	6.81	252	4.37
Midwest	2,671	46.28	967	16.75	321	5.56	916	15.87	467	8.09
West	953	16.51	307	5.32	66	1.14	363	6.29	217	3.76
Plan type (n, %)
Consumer directed health care	6	0.10	4	0.07	—	—	1	0.02	1	0.02
Health maintenance organization	2,512	43.52	874	15.14	322	5.58	948	16.42	368	6.38
Indemnity plan	200	3.47	76	1.32	27	0.47	45	0.78	52	0.90
Point of service plan	870	15.07	260	4.50	82	1.42	351	6.08	177	3.07
Preferred provider organization	2,143	37.13	789	13.67	140	2.43	723	12.53	491	8.51
Other or unknown	41	0.71	16	0.28	5	0.09	13	0.23	7	0.12
Index year (n, %)
1996	11	0.19	5	0.09	6	0.10	—	—	—	—
1997	15	0.26	5	0.09	5	0.09	5	0.09	—	—
1998	119	2.06	63	10.90	27	0.47	29	0.50	—	—
1999	222	3.85	132	2.29	28	0.49	62	10.70	—	—
2000	483	8.37	209	3.62	79	1.37	195	3.38	—	—
2001	502	8.70	225	3.90	87	1.51	190	3.29	—	—
2002	638	11.05	175	3.03	99	1.72	237	4.11	127	2.20
2003	1,656	28.69	553	9.58	182	3.15	529	9.16	392	6.79
2004	1,679	29.09	510	8.84	52	0.90	667	11.56	450	7.80
2005	447	7.74	142	2.46	11	0.19	167	2.89	127	2.20
Drug exposures during baseline (n, %)
Any MS treatment before index	910	15.77	98	1.70	115	1.99	322	5.58	375	6.50
Interferon-β1a IM	541	9.37	—	—	65	1.13	22	3.85	254	4.40
Interferon-β1b SC	127	2.20	25	0.43	—	—	74	1.28	28	0.49
Glatiramer acetate	187	3.24	52	0.90	42	0.73	—	—	93	1.61
Interferon-β1a SC	58	1.00	18	0.31	10	0.17	30	0.52	—	—
Natalizumab	4	0.07	3	0.05	—	—	—	—	1	0.02

n, number of patients with characteristic; SD, standard deviation; MS, multiple sclerosis; MRI, magnetic resonance imaging; NEC, necrotizing enterocolitis; SC, subcutaneous; IM, intramuscular.

Cox proportional hazards models (Table 2) showed that the difference in treatment changes between patients treated with glatiramer acetate and interferon-β1a IM as index drugs, remained significant after adjusting for important baseline factors. Prior use of MS drugs in the 6-month pre-index period was significantly associated with treatment changes, with patients who had prior MS therapy 2.5 times more likely to switch treatments than those with no prior MS therapies. Younger patients (aged 0–34 years) and female MS patients were more likely to change treatments. Treatment switching was also more common in the years 2001 and 2002 compared to other periods in the study.

Table 2. Cox proportional hazards model of time to treatment change or discontinuation.

Change treatment
Predictor	Hazard ratios	95% hazard ratio CI
Index drug
Glatiramer acetate	0.619	0.51	0.751
Interferon-β1b SC	1.249	0.969	1.612
Interferon-β1a SC	0.884	0.71	1.1
Interferon-β1a IM	ref
Prior use of MS drugs
Yes	2.46	2.06	2.9520
No	ref
Gender
Female	1.41	1.15	1.72
Male	ref
Age
0–34	1.63	1.34	1.9860
35–44	1.16	0.97	1.40
45–54	ref
55–64	0.71	0.53	0.95
Index year
1998–2000	0.75	0.57	0.97
2001–2002	ref
2003–2006	0.82	0.68	0.9890
Costs
Cost of emergency services	1.11	1.04	1.1980
Cost of MRI	1.05	1.01	1.1020
Cost of outpatient services	1.01	1.00	1.0180
Discontinuation
Variable	Hazard ratios	95% hazard ratio CI
Index drug
Glatiramer acetate	1.101	0.99	1.224
Interferon-β1b SC	2.458	2.141	2.822
Interferon-β1a SC	1.05	0.918	1.202
Interferon-β1a IM	ref
Health insurance provider
HMO	0.89	0.81	0.9810
Other	ref
Prior use of MS drugs
Yes	0.77	0.67	0.8750
No	ref
Gender
Female	0.99	0.89	1.10
Male	ref
Age
0–34	1.24	1.10	1.3950
35–44	1.02	0.92	1.14
45–54	ref
55–64	0.98	0.86	1.13
Region
South	1.18	1.06	1.31
Other	ref
Index year
1998–2002	ref
2003–2006	1.46	1.31	1.6250
Costs
Cost of MRI	0.87	0.84	0.8960

HMO, health maintenance organization; ref, reference level; SC, subcutaneous; IM, intramuscular.

The regression models also showed that patients treated with interferon-β1b SC as an index drug had a hazard ratio of 2.5 relative to interferon-β1a IM for discontinuing treatment. In comparison, the hazard for discontinuation for glatiramer acetate was only about 10% higher than interferon-β1a IM, and for interferon-β1a SC it was 5% higher than for interferon-β1a IM. Patients with prior use of a different MS drug before the start of the index drug were less likely to discontinue treatment than those who were naïve to MS therapy. Treatment discontinuations were more likely to occur during the later years of the study (2003–2006 as compared to 1998–2002).

In all, 5,722 patients (94%) had at least 6 more months of further enrollment at 18 months after the start of their index drug, which was the minimum follow-up period required to allow an assessment of costs following a treatment change or discontinuation. More than half (55%) of the patients remained on the index drug, 11% had switched to a different MS drug without a break in treatment, and 33% had discontinued MS treatment for at least 90 days. Table 3 presents costs over an 18-month period separately for patients who switched, discontinued, and persisted with the initial treatment. Patients who switched MS treatments had the highest mean medical costs over the 18 months following the switch ($15,532), followed by patients who discontinued MS treatment for at least 90 days ($12,551). Those who remained on the initial treatment had mean medical costs of $9,399 over a randomly selected 18-month period during follow-up. Linear regression models (Table 4) confirmed that the costs associated with switching and discontinuing treatment were significantly higher (p < 0.0001) than with persistence, even when adjusting for patient characteristics, MS drug use, region of the US, historic intervals, insurance type, and medical costs during the 6 months preceding treatment start. This pattern of increase in medical costs following a switch varied with the drug that the patient was initially started on. Patients who switched from interferon-β1b SC had significantly higher subsequent medical costs, compared to those who switched from interferon-β1a IM (p = 0.03), while the increase in the group that switched from interferon-β1a SC was significantly lower than interferon-β1a IM (p < 0.01). Compared to patients who were started on interferon-β1a IM, those who discontinued interferon-β1b SC had the highest medical costs and those starting on glatiramer acetate and interferon-β1a SC had lower costs following discontinuation, but these differences were not statistically significant.

Table 3. Cost ($) of medical services associated with treatment change or discontinuation.

Medical service	Continue initial treatment(n = 3,199)		Change treatment(n = 656)		Discontinue treatment(n = 1,917)
	Mean	SD	Mean	SD	Mean	SD
Total cost of services in 18-month period ($)
Outpatient – any diagnosis	6,723.68	12,348.14	10,267.00	18,702.78	8,360.01	17,656.50
Outpatient – MS diagnosis	3,480.07	9,419.74	6,473.84	15,718.04	4,329.56	12,123.96
Emergency room – any diagnosis	176.61	680.53	370.79	1,334.38	281.86	1,099.63
Emergency room – MS diagnosis	66.67	416.59	141.01	784.87	101.13	523.63
Inpatient – any diagnosis	1,613.42	8,524.09	3,752.17	17,487.28	3,135.12	12,510.82
Inpatient – MS diagnosis	1,087.79	6,041.03	2,505.61	9,582.92	1,824.05	8,083.88
MRI – any diagnosis	885.47	1,563.97	1,141.66	1,792.10	774.38	1,495.72
MRI – MS diagnosis	687.89	1,366.75	842.02	1,533.68	553.81	1,267.03
Pharmacy costs	19,919.84	12,557.02	17,730.40	11,352.34	7,823.07	11,140.99
Non-pharmacy medical costs	9,399.17	16,861.29	15,531.62	28,950.58	12,551.38	24,320.70
Total medical and pharmacy costs	29,319.01	20,748.54	33,262.02	31,043.89	20,374.45	27,756.15

n, number of patients; SD, standard deviation; MS, multiple sclerosis; MRI, magnetic resonance imaging.

Table 4. Linear regression model of total non-pharmacy medical costs associated with treatment change or discontinuation.

Change treatment
Predictor	Parameter estimate	Standard error	p-value
Intercept	13,862	1,408.09	< 0.0001
Health insurance provider
HMO	–1,882.31	678.68	0.0056
Other	ref
Gender
Female	–1,102.23	748.16	0.14
Male	ref
Age
0–34	–1,821.83	880.29	0.0380
35–44	–519.50	736.02	0.48
45–54	ref
55–64	1,664.05	992.95	0.09
Index year
1998–2000	–2,689.21	1,044.40	0.01
2001–2002	ref
2003–2006	1,555.33	837.99	0.06
Region
East	1,100.75	775.34	0.16
Other regions	ref
Index drug
Interferon-β1a IM	ref
Interferon-β1b SC	2,788.12	1,287.77	0.03
Glatiramer acetate	–870.33	720.44	0.23
Interferon-β1a SC	–2,564.46	906.18	0.0047
Baseline costs
Total medical costs	0.05	0.03	0.0417
Total pharmacy cost	0.24	0.12	0.0390
Cost of outpatient services	0.61	0.06	< 0.0001
Treatment persistence
Persist with initial drug	–5,497.99	817.20	< 0.0001
Changed treatment	ref
Discontinuation
Predictor	Parameter estimate	Standard error	p-value
Intercept	11,524	914.58	< 0.0001
Health insurance provider
HMO	–2,488.82	569.87	< 0.0001
Other	ref
Prior use of MS drugs
Yes	3,560.77	951.40	0.0002
No	ref
Gender
Female	161.47	650.80	0.80
Male	ref
Age
0–34	–3,598.27	774.23	< 0.0001
35–44	–1,378.20	654.22	0.04
45–54	ref
55–64	2,273.89	860.19	0.01
Index year
1998–2000	–3,002.90	816.79	0.00
2001–2006	ref
Region
East	1,305.00	692.71	0.06
Other Regions	ref
Index drug
Interferon-β1a IM	ref
Interferon-β1b SC	1,829.42	996.61	0.07
Glatiramer acetate	–749.11	643.75	0.24
Interferon-β1a SC	–657.59	826.10	0.43
Baseline costs
Total medical cost	0.06	0.02	0.0128
Total pharmacy cost	–0.24	0.10	0.0156
Cost of outpatient services	0.69	0.06	< 0.0001
Treatment persistence
Persist with initial drug	–2,882.94	562.74	< 0.0001
Discontinued treatment	ref

HMO, health maintenance organization; ref, reference level.

Discussion

The present study examined the frequency of DMD treatment switching and discontinuation in patients with MS. Results showed that only about half (55%) of patients remained on their index drug, while 11% had switched to a different MS drug without a break in treatment and 33% had discontinued MS treatment for at least 90 days. These high rates of switching or discontinuing therapy are similar to prior published studies of MS patients^14,15. In the context of other chronic, incurable diseases, such rates predict treatments that were abandoned due to ‘lack of efficacy’ or adverse events^16–19. Further, patients who switched or discontinued their initial MS treatment for at least 18 months had higher average healthcare costs than those who maintained their initial schedules. Both findings suggest that the first attempt at treatment with a DMD drug faltered either due to a lack of efficacy or tolerability. The rise in non-pharmacy medical costs (attributable to further investigations and hospital visits) after a discontinuation or change in the MS medication suggests that in at least some cases, disruption in care will lead to more costs to compensate for the lack of efficacy of the first drug and/or to manage its side-effects.

Unfortunately, the reasons for stopping the initial treatment cannot be determined from analysis of an administrative claims database. Also, the MS cases followed in this analysis are billing diagnostic events unconfirmed through a review of medical records or other data sources and unstratified in terms of severity. A patient with relapsing-remitting disease may be channeled to treatment options at different rates than those with progressive disease. Such patients would be expected to display unique profiles of safety, efficacy, and costs due to both the differences of such treatments and the innate natural course of disease.

The use of an automated healthcare claims database for the present study permitted efficient cost analysis of a large cohort of ‘real world’ MS patients on the drugs of interest, with follow-up for an 18-month period. Claims data are an excellent source of information on healthcare utilization and costs, regardless of site of service. This claims analysis, while confirming compliance rates of earlier observational studies, allowed costs to be assigned sequentially and per treatment, although out-of-pocket expenses, over-the-counter drugs, and drugs given outside of the healthcare system cannot be considered.

Conclusion

In a large cohort of MS patients followed in a healthcare claims analysis, 44% of patients switched from their index drug or else discontinued MS treatment for at least 90 days. These patients accumulated higher average healthcare costs over 18 months at minimum than those who remained on their initial MS treatment. Research to elucidate the reasons for high rates of DMD switching and discontinuation would likely promote better continued care for MS patients while maintaining associated costs.

Transparency

Declaration of funding: This study was funded by Biogen Idec.

Declaration of financial/other relationships: MWR has disclosed that he is an employee of United BioSource Corp, which received funding for this project, and that he is a consultant to Biogen Idec on other similar projects. RS has disclosed that he is an employee of United BioSource Corp. CS has disclosed that he is a consultant and advisor for United BioSource Corp. KR has disclosed that she is an employee of Biogen Idec.

The JME peer reviewers 1 and 2 have not received an honorarium for their review work on this manuscript. Both have disclosed that they have no relevant financial relationships.

Acknowledgments: The authors would like to acknowledge the scientific support and direction provided by Dr Beth Nordstrom of United BioSource Corporation.

Notes

* Copaxone is a registered trademark of Teva Pharmaceutical Industries Ltd, Petach Tikva, Israel.

† Avonex is a registered trademark of Biogen Idec, Cambridge, MA, USA.

‡ Betaseron is a registered trademark of Bayer HealthCare Pharmaceuticals, Leverkusen, Germany.

§ Rebif is a registered trademark of EMD Serono, Inc., Rockland, MA, USA.