Journal of Medical Economics

ISSN: 1369-6998 (Print) 1941-837X (Online) Journal homepage: http://www.tandfonline.com/loi/ijme20

Treatment patterns in multiple sclerosis:

administrative claims analysis over 10 years

MerriKay Oleen-Burkey, Anissa Cyhaniuk & Eric Swallow

To cite this article: MerriKay Oleen-Burkey, Anissa Cyhaniuk & Eric Swallow (2013) Treatment
patterns in multiple sclerosis: administrative claims analysis over 10 years, Journal of Medical
Economics, 16:3, 397-406, DOI: 10.3111/13696998.2013.764309

To link to this article: http://dx.doi.org/10.3111/13696998.2013.764309

Accepted author version posted online: 09

Jan 2013.
Published online: 22 Jan 2013.

Submit your article to this journal

Article views: 232

View related articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=ijme20

Download by: [TEVA Pharmaceutical Ltd] Date: 16 February 2016, At: 08:17
 Journal of Medical Economics Vol. 16, No. 3, 2013, 397–406

1369-6998 Article 0136.R1/764309

doi:10.3111/13696998.2013.764309 All rights reserved: reproduction in whole or part not permitted

Original article
Treatment patterns in multiple sclerosis:
administrative claims analysis over 10 years

d
ite
l u nl n
MerriKay Oleen-Burkey Abstract
Downloaded by [TEVA Pharmaceutical Ltd] at 08:17 16 February 2016

na ow io

,
py us is Lim

d
Outcomes Scribe, LLC, Leawood, KS, USA, and Teva

se oa
so d ut
Pharmaceuticals, Kansas City, MO, USA Objective:
Treatment patterns for the MS disease-modifying therapies (DMT) have changed over time. The objective of

b
Anissa Cyhaniuk

K
this study was to examine and describe treatment patterns in MS over a 10-year period.

rp c i
Eric Swallow

fo ers tr
OptumInsight Life Sciences, Eden Prairie, MN, USA
aU

er n
Methods:

a
rm

D
MS patients who filled a DMT prescription between January 1, 2001 and December 31, 2010 were
Address for correspondence: identified from Clinformatics for DataMart affiliated with OptumInsight. Two cohorts were identified:
le is al
rin ted m fo

MerriKay Oleen-Burkey, PhD, Principal, Outcomes those with a DMT prescription in 2003 and those with a DMT prescription in 2008. Treatment patterns
ng or i
Scribe, LLC, 13701 Belinder Road, Leawood, KS were examined 2 years before and after the anchor prescriptions for each cohort.
si th rc
d p ibi om In

co ed
66224, USA.
Tel.: 913-302-1211; Fax: 816-508-5013;
t a . Au e
13

moburkey@kc.rr.com Results:
Comparing treatment patterns prior to the two anchor prescriptions, interferon-beta (IFN )-1a IM (Avonex)
20

and IFN -1b (Betaseron) gained the most users in 2001–2003, while IFN -1a IM and IFN -1a SC (Rebif)
Key words: gained the most users from 2006–2008. In the 2 years following the two anchor prescriptions, treatment
Multiple sclerosis – Disease-modifying therapy –
la d le ©

patterns changed. From 2003–2005, 21.2% of IFN -1a SC users and more than 15.0% of IFN -1a IM and
C

Immunomodulatory therapy – Treatment patterns

IFN -1b users changed to another interferon or glatiramer acetate (GA; Copaxone), while 12.5% of GA users
ht

ew p r

changed to an interferon, most often IFN -1a SC. From 2008–2010 the largest proportion of changes from
vi e o

Accepted: 4 January 2013; published online: 21 January 2013

an h

Citation: J Med Econ 2013; 16:397–406

ig

each of the interferons and natalizumab (NZ; Tysabri) were to GA, while those switching from GA were most
o

often changed to IFN -1a SC. Those with a 2008 anchor prescription for NZ were most often changed (57%)
r
yr
sp ize a

to GA.
di or S
y, us
No p
Co

Limitations:
au fo

In retrospective database analyses the presence of a claim for a filled prescription does not indicate that
the drug was consumed, and reasons for changes in therapy are not available. The study design looking
Un t
th

forward and backward from the anchor prescriptions may have contributed to differences in the proportion of
patients seen with no observable change in DMT. Claims-based data are also constrained by coverage
limitations that determine the data available and limit the generalizability of results to managed care patients.

Conclusions:
Changes in treatment patterns in the first half of the observation period were reflective of the addition of
IFN -1a SC to the market in 2002. Following the 2003 and 2008 anchor prescriptions there were
differences in treatment patterns, with more IFN -1a IM users being changed to IFN -1a SC after the
2003 anchor DMT, and more of each of the interferons and NZ being changed to GA following the 2008
anchor DMT. With the introduction of oral therapies for MS, treatment patterns will again be impacted.

Introduction
Multiple sclerosis (MS) is a central nervous system disorder with acute focal
inflammatory demyelination and axonal loss1. Relapsing remitting MS (RRMS)
is the most common form of the disease and is characterized by disease

! 2013 Informa UK Ltd www.informahealthcare.com/jme Multiple sclerosis treatment patterns Oleen-Burkey et al. 397
 Journal of Medical Economics Volume 16, Number 3 March 2013
exacerbations (new or worsening symptoms) and periods of
remission. Approximately 400,000 individuals in the US
have MS, with diagnosis typically occurring between 20–
50 years of age2. MS occurs most frequently in females, and
in Caucasians2.
Currently there is no cure for MS and treatment consists
primarily of the use of disease-modifying therapies (DMTs)
for reduction of relapses and slowing of disease progression
in conjunction with symptomatic treatment and support-
ive care3. The first DMT approved by the FDA for RRMS
was interferon beta-1b for subcutaneous injection (IFN -
1b; Betaseron, Bayer Healthcare Pharmaceuticals, Inc.,
Montville, NJ) in 1993 followed by intramuscular inter-
feron beta-1a (IFN -1a IM; Avonex, Biogen Idec, Inc.,
Cambridge, MA) and glatiramer acetate injection (GA;
Downloaded by [TEVA Pharmaceutical Ltd] at 08:17 16 February 2016
Copaxone, Teva Pharmaceuticals USA, Inc., North
Wales, PA) in 1996 and subcutaneous interferon-beta 1a
(IFN -1a SC; Rebif, EMD Serono, Inc., Rockland, MA) in
2002. Natalizumab injection (NZ; Tysabri, Elan
Pharmaceuticals, Inc., South San Francisco, CA) was
first approved for treating relapsing forms of MS in 2004,
removed from the market in 2005 due to a safety issue, and
returned to the market in 2006. In 2009, a second IFN -1b
(Extavia, Novartis AG, Novartis Pharmaceutical
Corporation, East Hanover, NJ) was approved as a new
brand. Fingolimod (FG; Gilenya, Novartis
Pharmaceutical Corporation, East Hanover, NJ), the first
oral therapy for MS, was approved for use in September
2010, and teriflunomide (TF; Aubagio, Genzyme
Corporation, Cambridge, MA), a second oral therapy,
was approved for use in September 2012. Neither oral
agent was present in the data used for this treatment pat-
terns analysis.
The National MS Society Consensus Statement recom-
mends that treatment with an IFN or GA should be initi-
ated as soon as possible following a diagnosis of RRMS4.
NZ is generally recommended for patients with an inade-
quate response or who are unable to tolerate other MS
therapies3. There are currently no US clinical guidelines
for the use of DMTs in MS. How patients are moved from
one DMT to another and the changes in treatment pat-
terns over time have not been well described since 2005.
Reasons for changing or stopping DMTs most commonly
include inadequate response to treatment, side-effects, and
lack of tolerability5–8.
Earlier studies that assessed DMT treatment patterns in
MS include Reynolds et al.9, who examined discontinua-
tion or switches of DMT and healthcare resource utiliza-
tion and related costs in an analysis of administrative
claims data for new drug initiators from 1996–2005. The
authors reported that patients who had used MS drugs pre-
viously were 2.5-times more likely to change treatments
than patients with no prior use of MS drugs. In a second
report from the same data designed to look at DMT adher-
ence, Reynolds et al.10 included 6134 MS patients
398 Multiple sclerosis treatment patterns Oleen-Burkey et al.
initiating one of the IFNs or GA. Over time the proportion
of patients who changed therapy increased and there were
differences in the proportion of changes by DMT.
Using a more contemporary commercial managed care
population from 2001–2007, Margolis et al.11 reported on a
retrospective cohort study of newly diagnosed MS patients.
The most common index prescriptions in their analysis
were for GA, IFN -1a IM, and IFN -1a SC, and of
those who initiated an index prescription, 78.7% remained
on their index therapy for the remainder of follow-up,
while 21.3% switched to another DMT. No information
was provided on therapy changes.
In an analysis of data from the National Ambulatory
Medical Care Survey (NAMCS) examining outpatient
prescribing patterns, Avasarala et al.12 reported on MS
treatment prescriptions from 1998–2003. This study did
not examine changes in treatment, but rather looked at
trends in MS treatment prescribing. Over the observation
period there did not appear to be a significant change in
the proportion of prescriptions for the commercially avail-
able MS drugs; however, the authors reported that the use
of IFN -1a IM declined from 27% in 1998 to 11% in 2000,
while use of GA and IFN -1b increased.
Two studies13,14 examined treatment patterns in the
North American Research Committee on Multiple
Sclerosis (NARCOMS) registry database. Vollmer
et al.13 focused on differences between matched cohorts
of veterans receiving care in the Veterans Health
Administration (VHA) and non-veterans with MS in
2001. Lo et al.14 conducted an analysis of NARCOMS
data from November 2003 to February 2004 examining
MS treatment patterns in veterans and non-veterans.
Although neither study examined therapy changes or rea-
sons for stopping therapy, they did examine prevalent
treatment patterns for the various cohorts at a given
point in time.
While the published literature on DMT treatment pat-
terns includes several studies with large sample sizes prior
to 2007, it does not reflect the changes in treatment pat-
terns that have occurred in recent years. Specifically, NZ
was reintroduced to the US market in 2006, and the
impact of this new therapy on treatment patterns has not
been described in the literature. The objective of this study
is to describe and compare treatment patterns in MS over
10 years looking at 2 years either side of anchor prescrip-
tions for DMT in 2003 and 2008.
Methods
Data source
The retrospective administrative patient claims data used
in this study included medical claims, pharmacy claims,
and patient eligibility information from United Health
www.informahealthcare.com/jme ! 2013 Informa UK Ltd
 Journal of Medical Economics Volume 16, Number 3 March 2013

Group (UHG) as well as data from non-UHG plans. The treatment identified during the study period. Eligibility
individuals covered by these health plans, 32 million did not require patients to be newly diagnosed.
annual lives in 2010, are geographically diverse across
the US, with greatest representation in the South and
Midwest US census regions. The plans provide fully
insured coverage for professional (e.g., physician), facility
Patient demographic characteristics
(e.g., hospital), and outpatient prescription medication Age and gender for the health plan enrollees with MS were
services. Outpatient pharmacy claims provide National captured from enrollment data. Table 1 describes patient
Drug Codes (NDC) for dispensed medications, quantity demographic characteristics for the overall study
dispensed, drug strength, days supply, provider specialty population.
code, and health plan and patient costs.
No identifiable protected health information was
extracted or accessed during the course of the study. Treatment patterns cohorts
Pursuant to the Health Insurance Portability and
Two cohorts were identified to contrast treatment patterns
Downloaded by [TEVA Pharmaceutical Ltd] at 08:17 16 February 2016

Accountability Act, the use of de-identified data does

for DMTs. The first cohort included patients who filled at
not require Institutional Review Board approval or
least one DMT prescription in 2003 and their first DMT
waiver of authorization15.
prescription was defined as the 2003 anchor prescription.
The second cohort included patients who filled at least
one DMT prescription in 2008; the first DMT prescription
Patient identification
in 2008 was defined as the 2008 anchor prescription. For
The study population was selected from commercial and each cohort, the treatment patterns were examined in
Medicare health plan members with evidence of MS (a two time periods (see Figure 1):
diagnostic code ICD-9 340. and a treatment with a mar-
2003 Cohort:
keted MS therapy) during the identification period of
 Period 1: From 1/1/2001 to the patient’s 2003 anchor
January 1, 2001 to December 31, 2010. Eligible patients
prescription.
had at least one filled prescription for a DMT (IFN -1a
 Period 2: From the 2003 anchor prescription to 12/31/
IM, IFN -1a SC, IFN -1b, GA, or NZ) and at least 24
2005.
months of insurance eligibility following the first MS
2008 Cohort:
 Period 1: From 1/1/2006 to the patient’s 2008 anchor
prescription.
Table 1. Patient demographic characteristics.  Period 2: From the 2008 anchor prescription to 12/31/
2010.
All MS patients
n % Analysis
Gender DMT treatment changes before and after the 2003 and
Female 16,818 76.7%
Age
2008 anchor prescriptions were analyzed descriptively.
18 and below 151 0.69% Combination therapy with DMTs in MS was observed
19–35 4626 21.1% less than 1% of the time. Therefore, when patients filled
36–50 11,292 51.5%
51–62 5372 24.5% a new DMT prescription prior to completion of the anchor
63 and above 504 2.3% prescription it was treated as a therapy change and not as
combination therapy.

Figure 1. Study period.

! 2013 Informa UK Ltd www.informahealthcare.com/jme Multiple sclerosis treatment patterns Oleen-Burkey et al. 399
 Journal of Medical Economics Volume 16, Number 3 March 2013
Downloaded by [TEVA Pharmaceutical Ltd] at 08:17 16 February 2016
Figure 2. DMT use prior to initiation of 2003 anchor prescription.
Results
Sample characteristics
There were 8323 patients with an anchor DMT
prescription in 2003; 18,910 patients had an anchor
DMT prescription in 2008.
2003 Cohort
Of the 8323 patients with an anchor DMT prescription in
2003, 62.9% had no observable change in therapy between
January 1, 2001, and the anchor DMT prescription in
2003. Of the 3088 patients who did experience a change
in therapy in Period 1, Figure 2 shows the proportions of
patients using another DMT in the 2001–2003 time frame
before initiating therapy with a specific 2003 anchor DMT
prescription.
In Period 2, looking forward from the 2003 anchor
prescription, 34.2% of the patients did not experience an
observable change in DMT in the 2003–2005 time frame.
Figure 3 presents the proportions of patients changing to
each DMT in the 2003–2005 time frame from each 2003
anchor DMT prescription.
IFN -1a IM
In Period 1, looking back from 2003 to January
2001, 12.1% of the 8323 patients changed to IFN -1a
400 Multiple sclerosis treatment patterns Oleen-Burkey et al.
IM. Of these patients, 43.5% were previously on
IFN -1a SC, 38.4% on GA, and 18.1% on IFN -1b.
From 2003–2005 (Period 2), 16.9% with a 2003
anchor prescription for IFN -1a IM changed to a
different DMT: 51.6% of patients changed to IFN -1a
SC, 35.6% to GA, and 12.5% to IFN -1b.
IFN -1a SC
From 2001–2003 (Period 1), 7.8% of the 8323 patients
changed to IFN -1a SC as their 2003 anchor prescription.
Of these patients, 54.6% were previously on IFN -1a IM,
34.6% on GA, and 10.9% from IFN -1b. In Period 2, from
2003–2005, 21.2% of patients with a 2003 anchor prescrip-
tion for IFN -1a SC changed to a different DMT. Of these
patients, 51.4% changed to GA, 34.3% to IFN -1a IM,
and 13.7% to IFN -1b.
IFN -1b
Of the 8323 patients, 9.4% changed to IFN -1b as their
2003 anchor DMT prescription during Period 1 (2001–
2003). Of these patients, 50.8% were previously on GA,
30.3% on IFN -1a IM, and 18.9% on IFN -1a SC.
Looking forward from 2003 (Period 2), 15.2% of patients
with a 2003 anchor prescription for IFN -1b changed to a
different DMT. Of these patients, 57.5% changed to GA,
22.4% changed to IFN -1a SC, and 20.1% changed to
IFN -1a IM.
www.informahealthcare.com/jme ! 2013 Informa UK Ltd

Downloaded by [TEVA Pharmaceutical Ltd] at 08:17 16 February 2016
Figure 3. Changes in DMT following the 2003 anchor prescription.
GA
During Period 1, from 2001–2003, 7.8% of the 8323
patients changed to GA as their 2003 anchor prescription:
37.1% were previously on IFN -1a IM, 34.9% on IFN -1a
SC, and 28.0% on IFN -1b. In Period 2, 2003–2005,
12.5% of patients with a 2003 anchor prescription for
GA changed to a different DMT. Of these patients,
52.2% changed to IFN -1a SC, 29.9% to IFN -1a IM,
and 17.5% to IFN -1b.
2008 Cohort
Of the 18,910 patients with an anchor DMT prescrip-
tion in 2008, 46.7% had no observable change in therapy
between January 1, 2006, and the anchor DMT prescrip-
tion in 2008. Of the 10,079 patients who did experience a
change in therapy in Period 1, Figure 4 shows the propor-
tions of patients using another DMT in the 2006–2008
time frame before initiating therapy with a specific 2008
anchor DMT prescription.
In Period 2, looking forward from the 2008 anchor pre-
scription, 35.0% of the patients did not experience an
observable change in DMT in the 2008–2010 time
frame. Figure 5 presents the proportions of patients chang-
ing to each DMT in the 2008–2010 time frame from each
2008 anchor DMT prescription.
IFN -1a IM
In Period 1, looking back from 2008 to January 1, 2006,
14.0% of the 10,079 patients changed to IFN -1a IM. Of
! 2013 Informa UK Ltd www.informahealthcare.com/jme
Journal of Medical Economics Volume 16, Number 3 March 2013
these patients, 34.7% were previously on IFN -1a SC,
32.6% on GA, 26.5% on NZ, and 6.3% on IFN -1b.
From 2008–2010 (Period 2), 14.0% with a 2008 anchor
prescription for IFN -1a IM changed to a different DMT:
38.0% of patients changed to GA, 31.5% to IFN -1a SC,
21.1% to NZ, and 9.4% to IFN -1b.
IFN -1a SC
From 2006–2008 (Period 1), 13.6% of the 10,079
patients changed to IFN -1a SC as their 2008 anchor
prescription. Of these patients, 50.0% were previously on
GA, 32.9% on NZ, 11.3% on IFN -1a IM, and 5.8% on
IFN -1b. In Period 2 from 2008–2010, 15.7% of patients
with a 2008 anchor prescription for IFN -1a SC changed
to a different DMT. Of these patients, 54.9% changed to
GA, 25.0% to NZ, 11.6% to IFN -1a IM, and 8.5% to
IFN -1b.
IFN -1b
Of the 10,079 patients, 11.9% changed to IFN -1b
as their 2008 anchor prescription during Period 1
(2006–2008). Of these patients, 40.5% were previously
on GA, 33.1% on NZ, 14.8% on IFN -1a SC, and
11.7% on IFN -1a IM. Looking forward from 2008–2010
(Period 2), 13.8% of patients with a 2008 anchor
prescription for IFN -1b changed to a different DMT.
Of these patients, 48.1% changed to GA, 24.8% to NZ,
Multiple sclerosis treatment patterns Oleen-Burkey et al. 401
 Journal of Medical Economics Volume 16, Number 3 March 2013
Downloaded by [TEVA Pharmaceutical Ltd] at 08:17 16 February 2016

Figure 4. DMT use prior to initiation of 2008 anchor prescription.

Figure 5. Changes in DMT following the 2008 anchor prescription.

402 Multiple sclerosis treatment patterns Oleen-Burkey et al. www.informahealthcare.com/jme ! 2013 Informa UK Ltd

Downloaded by [TEVA Pharmaceutical Ltd] at 08:17 16 February 2016
Figure 6. DMT utilization by product from 2001–2010.
16.2% changed to IFN -1a SC, and 10.8% changed to
IFN -1a IM.
GA
During Period 1, from 2006–2008, 9.1% of the 10,079
patients changed to GA as their 2008 anchor prescription:
33.7% were previously on IFN -1a SC, 30.7% on NZ,
20.6% on IFN -1a IM, and 14.9% on IFN -1b. In
Period 2, 2008–2010, 10.1% of patients with a 2008
anchor prescription for GA changed to a different DMT.
Of these patients, 41.3% changed to IFN -1a SC, 24.8%
to NZ, 17.9% to IFN -1b, and 16.1% to IFN -1a IM.
NZ
In Period 1, from 2006–2008, 4.7% of the 10,079 patients
changed to NZ as their 2008 anchor prescription. Of these
patients, 42.9% were previously on GA, 22.4% on IFN -
1a IM, 20.4% on IFN -1a SC, and 14.3% on IFN -1b.
During Period 2 from 2008–2010, 11.4% of patients with
a 2008 anchor prescription for NZ changed to a different
DMT. Of these patients, 57.0% changed to GA, 19.2% to
IFN -1a IM, 15.4% to IFN -1a SC, and 8.5% to IFN -1b.
Treatment comparison between time periods
In the 2001–2003 time period, 62.9% of the patients had
no observable change in DMT prior to their anchor DMT
! 2013 Informa UK Ltd www.informahealthcare.com/jme
Journal of Medical Economics Volume 16, Number 3 March 2013
prescription, compared to 46.7% in the 2006–2008 time
period. About a third of each DMT anchor cohort in 2003
and 2008 had no changes in DMT in the 2 years following
their anchor prescriptions.
Regarding the changes in DMT observed over the
various periods of the study, IFN -1a IM and IFN -1b
gained the most users in the 2001–2003 time frame,
while IFN -1a IM and IFN -1a SC gained the most
users in the 2006–2008 time frame. The most changes in
DMT in each of the follow-up periods from the anchor
DMT prescriptions occurred with interferon users.
Between 2003–2005, 21.2% of IFN -1a SC users
and more than 15.0% of IFN -1a IM and IFN -1b
users changed to another interferon or GA, while
12.5% of GA users changed to an interferon, most often
IFN -1a SC. Those using IFN -1a IM also changed most
often to IFN -1a SC, while those using IFN -1a SC or
IFN -1b changed to GA more than 50% of the time.
Between 2008–2010 the largest proportion of changes
from each of the interferons and NZ were to GA. Of
those with a 2008 anchor prescription for NZ, 11.4%
had a change in therapy, with more than half of them
receiving GA.
Figure 6 shows the shift in treatment patterns for the
treated MS patients over the 2001–2010 time frame.
The proportion of patients utilizing GA increased, while
the proportion utilizing IFN -1a IM decreased. Use
of IFN -1a SC increased after its introduction to the
US market in 2002, and has remained fairly stable
Multiple sclerosis treatment patterns Oleen-Burkey et al. 403
 Journal of Medical Economics Volume 16, Number 3 March 2013
since 2003. Use of NZ has been increasing since its
reintroduction in 2006. IFN -1b use declined after the
introduction of IFN -1a SC, and remained fairly stable
since 2003.
Discussion
This study assessed treatment patterns in four time periods,
from 2001–2003, 2003–2005, 2006–2008, and 2008–2010.
From 2001–2003, the largest proportions of MS patients
were started on IFN -1a IM and IFN -1b, with GA and
IFN -1a SC having the lowest percentages of therapy
gains. From 2003–2005, the largest proportions of patients
Downloaded by [TEVA Pharmaceutical Ltd] at 08:17 16 February 2016
were changed from IFN -1a SC to a different DMT. This is
probably due to the introduction of IFN -1a SC to the US
market in 2002 and the fact that patients who are begin-
ning a new therapy are most vulnerable to discontinuation
during the first 3–6 months16. Early discontinuation is
often the result of unrealistic expectations about the effec-
tiveness of the therapy in the short-term or a lack of tol-
erability. The lowest percentage of patients were being
changed from GA to a different DMT during the 2003–
2005 time period. Between 2006–2008, the largest propor-
tion of patients were being changed to IFN -1a IM and
IFN -1a SC. NZ, the newest entry to the MS market in
2006, was beginning to see increased use. From 2008–
2010, there was increased use of GA as therapy changes
from IFN -1a SC, IFN -1a IM, IFN -1b and NZ were
most often to GA. In part this may be attributed to the
publications of head-to-head trials which showed GA to
be as effective in terms of relapse reduction and slowing of
disability progression as two of the interferons, IFN -1a
SC and IFN -1b17,18.
Consistent with the results from Reynolds et al.9, who
reported that of patients on IFN -1a IM during the base-
line period (1996–2005), the most common change was to
IFN -1a SC, the analysis presented here shows most
patients with a IFN -1a IM anchor prescription in 2003
changed to IFN -1a SC in 2003–2005.
There was also consistency in these findings relative to
other studies that have reported lower rates of therapy
change from GA than from other DMTs. Patients with
GA as their anchor prescription in either 2003 or 2008
were less likely to have changed therapies than patients on
other treatments. Reynolds et al.9 reported that, compared
to patients with IFN -1a IM as their index drug, patients
with GA as their index drug were 40% less likely to change
therapies (HR: 0.619; 95% CI: 0.51, 0.751). A separate
analysis by Reynolds et al.10 reported that patients initiat-
ing GA had the lowest rates of therapy change within the
first 6 months after initiation (3.9%) and within 12
months after initiation (6.4%) compared to the beta-
interferons.
404 Multiple sclerosis treatment patterns Oleen-Burkey et al.
The published literature on MS therapy changes does
not include treatment pattern information on recently
available MS therapies such as NZ, FG, and TF. While
the use of FG and TF were not evaluated in this study
due to a lack of data in the study population, this study
does provide a picture of treatment patterns prior to and
after the use of NZ. From 2006–2008, 5% of patients
were changed from their anchor prescription to NZ,
most often changing from GA, IFN -1a IM, and
IFN -1a SC. From 2008–2010, more than 11% of
patients stopped NZ and moved to another DMT.
Over 50% of patients changed from NZ to GA, while
19% began IFN -1a IM and 15% started IFN -1a SC.
The discontinuation of NZ was related to concerns
about the increasing number of cases of progressive mul-
tifocal leukoencephalopathy (PML) associated with the
drug. Unlike in the early years of this study interval
when there were only three MS therapies available
and fewer alternatives when one therapy appeared inef-
fective or intolerable, the additional array of therapies
gave physicians and patients more treatment options in
these later years that led to more therapy changes.
Over the entire span of the study from 2001–2010, the
use of IFN -1a IM declined, while the use of GA
increased. These findings are consistent with a trend that
was first reported in an earlier analysis of the National
Ambulatory Medical Care Survey (NAMCS)12. In this
study, Avasarala et al.12 reported that, from 1998–2000,
the use of IFN -1a IM declined, while use of IFN -1b
and GA showed an upward trend. IFN -1a SC did not
become commercially available until 2002, and was insig-
nificantly represented in their study.
Reynolds et al.9 examined healthcare resource utiliza-
tion in MS patients after a treatment change or treat-
ment discontinuation and compared it to those
continuing initial treatment. Approximately half
(55%) of the patients remained on their index prescrip-
tion, while 11% changed MS therapy and 33% discon-
tinued MS treatment for at least 90 days. The 18-month
medical costs associated with changing therapies or stop-
ping them were significantly higher (p50.0001) than
those for patients who continued their initial treatment.
Changing DMTs may result in increased healthcare
costs, since the more common reasons for therapy
changes are lack of effectiveness or troublesome side-
effects that may require additional outpatient services.
In addition, a new therapy could require medical man-
agement of its side-effects9. However, as Gajofatto
et al.19 have shown, changing a first-line DMT in
patients who are not responding may be effective in
many cases. Future research examining the specific rea-
sons why patients change from one DMT to another
may assist in determining the optimal sequence of MS
treatments and contribute to the development of clini-
cal guidelines.
www.informahealthcare.com/jme ! 2013 Informa UK Ltd

Limitations
Claims database analysis allows for estimation of real-
world treatment patterns, and the strength of our analysis
derives from the large, geographically diverse population
studied. However, all retrospective database analyses are
subject to certain limitations, and the results of this study
must be interpreted with appropriate consideration of
these limitations. Claims data are collected primarily for
payment purposes, not research, and are subject to coding
errors. The presence of a claim for a filled prescription does
not necessarily indicate that the medication was consumed
or that it was taken as prescribed. The changing patterns
described here are based on filled prescriptions. Claims-
based data are constrained by coverage limitations which
Downloaded by [TEVA Pharmaceutical Ltd] at 08:17 16 February 2016
determine the data available and limit the generalizability
of results to managed care patients. As the majority of MS
patients are diagnosed between 20–50 years of age, how-
ever, this data source is representative of the MS patient
population covered by managed care organizations.
Administrative claims allow for the identification of
changes in treatment, but do not allow for an examination
of the reasons why treatment changes occur. In this study,
it is unknown whether changes in treatment are due to a
lack of response to therapy, tolerability issues or side-
effects, form of administration, or health plan structure
such as copayments.
In this study, there were differences in the proportion of
patients who had no observable change in DMT depend-
ing on whether the investigation was done using a retro-
spective or prospective approach. Retrospectively, nearly
half to two-thirds of the patients had no observable DMT
change, while, prospectively, it appears that about one-
third of patients had no change in therapy over the subse-
quent 2 years. This is probably an artifact of the study
design with a longer period of insurance coverage and,
thus, database eligibility required looking forward from
the anchor DMT prescription than was required looking
backward.
Because recent and pending entries of oral medications
into the MS armamentarium were outside of the timeframe
of this study their patterns of use could not be examined.
There is a need for future research to examine the place of
oral therapy in the treatment of MS.
Finally, the patient population was identified for this
study based on a filled prescription for an MS DMT.
Patients were not stratified by type of MS, by disease sever-
ity, or by being newly diagnosed or newly treated. As noted
by Margolis et al.11, who investigated only newly-diag-
nosed MS patients, the proportion of therapy changes
among a sub-set of only the newly diagnosed would be
lower than for the combination of newly-diagnosed and
continuously-diagnosed patients examined for this study.
Changing patterns and choice of treatment may vary by
these characteristics.
! 2013 Informa UK Ltd www.informahealthcare.com/jme
Journal of Medical Economics Volume 16, Number 3 March 2013
Conclusion
In general, switching rates between drugs is an under-
served topic with more published data focusing on therapy
adherence. The level of switching from the anchor DMT
prescriptions found in our study is consistent with other
published studies on treatment patterns. However, this is
the first study to describe a dynamic pattern of switching
over a relatively long time interval inclusive of the time
two new drugs (IFN -1a SC and NZ) entered the market.
Changes in therapy prior to the anchor DMT prescriptions
of 2003 and 2008 were reflective of the addition of IFN -
1a SC to the market in 2002 and the re-entry to the market
of NZ in 2006. Following the anchor prescriptions of 2003
and 2008 there were differences in patterns of treatment
with more IFN -1a IM users being changed to IFN -1a SC
after the 2003 anchor DMT, while more of each of the
interferons and NZ were changed to GA following the
2008 anchor DMT. Now that the first oral DMTs have
been introduced into the MS market there is a need to
evaluate more recent changes in the treatment patterns
of persons with MS.
Transparency
Declaration of funding
Support for this study was provided by Teva Pharmaceuticals.
Declaration of financial/other relationships
MOB is a former employee of Teva Pharmaceuticals. AC and ES
are employees of OptumInsight, which was contracted by Teva
Pharmaceuticals to conduct the study.
Acknowledgments
Michelle Sotak provided medical writing assistance and is an
employee of OptumInsight.
References
1. Compston A, Coles A. Multiple sclerosis. Lancet 2002;359:1221-31
2. National Multiple Sclerosis Society. Epidemiology of MS [online]. National
Multiple Sclerosis Society, 2011. http://www.nationalmssociety.org/about-
multiple-sclerosis/what-we-know-about-ms/who-gets-ms/epidemiology-of-
ms/index.aspx. Accessed September 5, 2011.
3. Derwenskus J. Current disease-modifying treatment of multiple sclerosis. Mt
Sinai J Med 2011;78:161-75
4. Expert opinion paper: disease management consensus statement [online].
National Clinical Advisory Board of the National Multiple Sclerosis Society,
2008, New York, NY. http://www.nationalmssociety.org/search-results/
index.aspx?q¼ConsensusþStatement&start¼0&num¼20&x¼22&y¼10.
Accessed September 5, 2011.
5. O’Rourke KE, Hutchinson M. Stopping beta-interferon therapy in multiple scle-
rosis: an analysis of stopping patterns. Mult Scler 2005;11:46-50
6. Rı́o J, Porcel J, Téllez N, et al. Factors related with treatment adherence to
interferon beta and glatiramer acetate therapy in multiple sclerosis. Mult Scler
2005;11:306-9
7. Tremlett HL, Oger J. Interrupted therapy: stopping and switching of the beta-
interferons prescribed for MS. Neurology 2003;61:551-4
Multiple sclerosis treatment patterns Oleen-Burkey et al. 405
 Journal of Medical Economics Volume 16, Number 3 March 2013
8. Twork S, Nippert I, Scherer P, et al. Immunomodulating drugs in multiple
sclerosis: compliance, satisfaction and adverse effects evaluation in a German
multiple sclerosis population. Curr Med Res Opin 2007;23:1209-15
9. Reynolds MW, Stephen R, Seaman C, et al. Healthcare resource utilization
following switch or discontinuation in multiple sclerosis patients on disease
modifying drugs. J Med Econ 2010;13:90-8
10. Reynolds MW, Stephen R, Seaman C, et al. Persistence and adherence to
disease modifying drugs among patients with multiple sclerosis. Curr Med
Res Opin 2010;26:663-74
11. Margolis JM, Fowler R, Johnson BH, et al. Disease-modifying drug initiation
patterns in commercially insured multiple sclerosis patients: a retrospective
cohort study. BMC Neurol 2011;11:122-31. http://www.biomedcentral.com/
1471-2377/11/122. Accessed November 20, 2012.
12. Avasarala JR, O’Donovan CA, Roach SE, et al. Analysis of NAMCS data for
multiple sclerosis, 1998–2004. BMC Med 2007;5.6 doi:10.1186/1741-7015-
5-6 Accessed Jan 2013 http://www/biomedcentral.com/1741-7015/5/6
13. Vollmer TL, Hadjimichael O, Preiningerova J, et al. Disability and treatment
patterns of multiple sclerosis patients in United States: a comparison of vet-
Downloaded by [TEVA Pharmaceutical Ltd] at 08:17 16 February 2016
erans and nonveterans. J Rehabil Res Dev 2002;39:163-74
406 Multiple sclerosis treatment patterns Oleen-Burkey et al.
14. Lo AC, Hadjimichael O, Vollmer TL. Treatment patterns of multiple sclerosis
patients: a comparison of veterans and non-veterans using the NARCOMS
registry. Mult Scler 2005;11:33-40
15. 104th Congress. Health Insurance Portability and Accountability Act of 1996.
Public Law 104-191 ed. 1996.
16. Patti F. Optimizing the benefit of multiple sclerosis therapy: the importance of
treatment adherence. Patient Prefer Adherence 2010;4:1-9.
17. Mikol DD, Barkhof F, Coyle PK, et al. Comparison of subcutaneous interferon
beta-1a with glatiramer acetate in patients with relapsing multiple sclerosis
(the REbif vs. Glatiramer Acetate in Relapsing MS Disease [REGARD] Study): a
multicentre, randomised, parallel, open-label trial. Lancet Neurol 2008;7:
903-14
18. O’Connor P, Filippi M, Arnason B, et al. 250 mg or 500 mg interferon beta-1b
versus 20 mg glatiramer acetate in relapsing-remitting multiple sclerosis:
a prospective, randomised, multicentre study. Lancet Neurol 2009;8:
889-97.
19. Gajofatto A, Bachetti P, Grimes B, et al. Switching first-line disease-modifying
therapy after failure: impact on the course of relapsing-remitting multiple
sclerosis. Mult Scler 2009;15:50-8
www.informahealthcare.com/jme ! 2013 Informa UK Ltd