Original Paper
Eur Neurol 2011;65:59–67
DOI: 10.1159/000323216
Adherence to Disease-Modifying Therapies in
Spanish Patients with Relapsing Multiple
Sclerosis: Two-Year Interim Results of the Global
Adherence Project
E. Arroyo a C. Grau a C. Ramo-Tello b J. Parra a O. Sánchez-Soliño a  
on behalf of the GAP Study Group
a
  Biogen Idec Iberia SL, Madrid, and b Hospital Universitario Germans Trias i Pujol, Badalona, Spain
Key Words
Disease-modifying therapy ⴢ Patient adherence ⴢ
Multiple sclerosis ⴢ Patient education
Abstract
The post-marketing international Global Adherence Project
investigated adherence to disease-modifying therapy for re-
lapsing-remitting multiple sclerosis. We report adherence
data from the first 2 years in the Spanish subset of patients
(n = 254 at baseline). The overall adherence rate was 85.4%.
Patients taking intramuscular (IM) interferon-␤ (IFN␤)-1a
were significantly more adherent (96.4%) compared with
patients taking subcutaneous (SC) IFN␤-1a 22 ␮g (79.1%; p =
0.0064), SC IFN␤-1a 44 ␮g (79.6%; p = 0.0064) and glatiramer
acetate (82.7%; p = 0.0184). At year 1 (n = 142), the overall
adherence rate was 86.6%. Patients on IM IFN␤-1a were sig-
nificantly more adherent than patients on SC IFN␤-1a 22 ␮g
(93.9 vs. 66.7%; p = 0.0251). At year 2 (n = 131), the overall ad-
herence rate was 82% (87.5% for IM IFN␤-1a, 80.0% for SC
IFN␤-1a 22 ␮g, 77.8% for SC IFN␤-1a 44 ␮g, 85.2% for IFN␤-
1b, and 80.0% for glatiramer acetate). In conclusion, adher-
ence remained high among all disease-modifying therapies
over the first 2 years of the study and was significantly high-
er for IM IFN␤-1a, at visit 1, compared with SC IFN␤-1a.
Copyright © 2011 S. Karger AG, Basel
© 2011 S. Karger AG, Basel
0014–3022/11/0652–0059$38.00/0
Fax +41 61 306 12 34
E-Mail karger@karger.ch Accessible online at:
www.karger.com www.karger.com/ene
Received: April 2, 2010
Accepted: November 29, 2010
Published online: January 4, 2011
Introduction
Multiple sclerosis (MS) is a chronic and debilitating
disease that can partly be controlled with long-term use
of disease-modifying therapy (DMT). The DMTs on the
market on initiation of the Global Adherence Project
(GAP) were intramuscular (IM) interferon-␤ (IFN␤)-1a
(Avonex쏐), subcutaneous (SC) IFN␤-1a 22 ␮g (Rebif 쏐22),
SC IFN␤-1a 44 ␮g (Rebif 쏐 44), IFN␤-1b (Betaferon쏐),
and glatiramer acetate (Copaxone쏐). The WHO defines
treatment adherence as compliance (that is, taking the
medication according to the prescribed dosing regimen)
and persistence with dosing (taking the medication
throughout the entire indicated treatment period), and
affirms that lack of adherence contributes to reduced ef-
ficacy in the treatment of chronic diseases [1]. Between 19
and 39% of patients with MS stop treatment with IFN␤
within 3 years [2, 3]. Moreover, it has been reported that
between 10 and 20% of patients who stop treatment do so
in the first 6 months [3, 4].
In clinical trials, the patients undergo close follow-up
and regular examinations and are encouraged to contin-
ue treatment, while the healthcare professionals are mon-
itored by the sponsor. In observational studies, in con-
trast, monitoring of adherence is variable and may be in-
fluenced by several factors to which patients in clinical
Hoffmann-La Roche & Co. AG, Hauptbibliothek
72.34.133.250 - 1/5/2019 4:09:06 AM
Estefanía Arroyo
Departamento Médico, Biogen Idec Iberia
Paseo de la Castellana, 41-3a, ES–28046 Madrid (Spain)
Tel. +34 91 310 7142, Fax +34 91 310 7111
E-Mail estefania.arroyo @ biogenidec.com
Downloaded by:
trials are not exposed, thus allowing a more faithful eval-
uation of clinical practice [5–7].
Patients are often wary of starting DMT because of the
side effects and chronic administration regimens with
frequent injections [6]. It has been observed that DMT
might not be effective in those patients with a poor level
of adherence in the long term [2]. According to different
authors, the main factors that contribute to whether the
patient is adherent are related to perceived lack of effica-
cy, lack of information or too complex information, false
hopes of improvement in the disease, problems adminis-
tering treatment (such as fear of needles or self-injections)
and sociocultural factors [8, 9].
Few studies have compared adherence among DMTs
currently available for MS [6, 10]. The international GAP
study, an observational, multicenter, phase IV, post-
marketing, retrospective, cross-sectional study, included
2,648 patients with relapsing-remitting MS (RRMS) in 22
countries [5]. A global adherence rate of 75% was found,
with forgetting to inject the factor that most contributed
to lack of adherence (50.2%).
In Spain and Portugal, a 5-year follow-up study is be-
ing conducted to assess long-term adherence and the fac-
tors that influence adherence in patients who were previ-
ously enrolled in the GAP study [5]. In this article, we
report the results of the first 2 years of the study in Spain.
Patients and Methods
This is an observational, multicenter, phase IV, post-market-
ing, retrospective, cross-sectional study of patients with RRMS
who attend annual visits for 5 years. The secondary objectives in
Spain included assessment of patient satisfaction and physician
satisfaction with the new presentation of IM IFN␤-1a and the
health-related quality of life – these data will be published sepa-
rately.
Non-adherence was defined as missing an injection or dose
modification in the 4 weeks prior to completing the survey. The
patients signed an informed consent both at the start of the study
and for follow-up, and the ethics committees approved the initial
study in 18 centers and the follow-up study in 15.
The study included adult patients with RRMS in treatment for
at least 6 months prior to inclusion in the study with one of the
DMTs available at the start of the study (in accordance with the
prescribing information and the judgment of the MS Assessment
Committee in each Spanish autonomous region, if applicable).
Each center aimed to include the same number of patients in
each of the treatment groups. For each treatment group, it was
planned to include at least 3 patients and at most 6. The maxi-
mum number of patients included per center was therefore 30.
Patients were included consecutively as they attended their regu-
lar follow-up appointment and gave their consent to participate
in the study.
60 Eur Neurol 2011;65:59–67
The questionnaire data were compared with the medical
records by monitoring visits conducted by an independent com-
pany. Patients and neurologists filled out the questionnaires in-
dependently and without any interaction in a single annual visit.
The Neurologist Questionnaire comprised 13 questions that in-
cluded information on details of the place of work (infrastructure,
roles of nurses and/or other professions involved), questions on
the information provided to the patients about treatment (mech-
anism of action, adverse reactions, administration method) and
questions about the relevance of adherence and factors that might
influence it. The patient questionnaire collected information on
the point of view of the patients about the care received (health-
care staff involved, value of the different visits and education that
was given during these visits), personal view of the current MS
therapy and its complications, adherence to drugs for treating MS
in the last 4 weeks and sources of support that might influence
patient adherence. In each patient questionnaire, the healthcare
worker completed the first 10 questions about disease duration,
degree of disability, treatment, and history of relapses and treat-
ment.
Statistical Analysis
A descriptive analysis was undertaken, along with a correla-
tion analysis of the variables or factors that influence adherence.
Continuous data were described using appropriate statistics:
mean and standard deviation, or median and range. Possible
group differences were analyzed by means of a parametric test
(ANOVA) and a non-parametric one (Kruskal-Wallis test); cate-
gorical data were presented by frequency distribution and per-
centage for each group of DMT.
Adherence rates were estimated and compared between IM
IFN␤-1a and the other DMTs. The analyses were done using two-
tailed tests with a type-I error (␣-type error) of 0.05.
Frequencies and percentages for ‘good treatment adherence’
versus ‘lack of treatment adherence’ variable were analyzed, and
possible differences between treatment environments were inves-
tigated using the ␹2 test or the Fisher exact test.
Factors potentially related to treatment adherence were ana-
lyzed using a log-rank test. The dependent variable (adherence)
was analyzed according to treatment satisfaction, effectiveness of
current treatment, ease of treatment administration, treatment
tolerability, effect on delaying disease progression and improve-
ment in MS symptoms, in addition to sociodemographic vari-
ables.
Results
The Spanish part of the GAP study included 254 pa-
tients at the baseline visit, 142 after 1 year of follow-up
(visit 1) and 131 after 2 years of follow-up (visit 2). The
baseline demographic characteristics are presented in ta-
ble 1. Significant differences were only found in the gla-
tiramer-acetate-treated group (lower mean age, shorter
treatment duration and shorter disease duration), in the
SC IFN␤-1a 44-␮g-treated group (shorter treatment dura-
tion) and the IFN␤-1b-treated group (greater disease dura-
Hoffmann-La Roche & Co. AG, Hauptbibliothek
72.34.133.250 - 1/5/2019 4:09:06 AM
Arroyo et al.
Downloaded by:
Table 1. Baseline patient characteristics

IM IFN␤-1a IFN␤-1a 22 ␮g IFN␤-1a 44 ␮g IFN␤-1b GA Overall

(n = 56) (n = 43) (n = 54) (n = 49) (n = 529) (n = 254)

Mean age 8 SD, years 38.8810.1 37.289.0 38.4810.6 40.1811.0 35.188.2a 37.989.9
Sex, % female 78.2 62.8 68.5 67.3 73.1 70.4
Mean number of relapses in prior year, % patients
0 66.1 67.4 55.6 69.4 57.7 63.0
1 28.6 23.3 24.1 20.4 15.4 22.4
2 5.4 4.7 11.1 10.2 17.3 9.8
>3 0 4.7 9.3 0 9.6 4.7
Median (range)
Current treatment duration 40.5 (6–108) 40.0 (6–120) 24.0 (6–60)b 45.0 (6–124) 15.0 (6–42)c 28.0 (6–124)
Disease duration 8.0 (1–33) 7.0 (1–17) 5.0 (1–34) 7.0 (1–37)d 6.0 (1–16)e 6.0 (0–37)

GA = Glatiramer acetate; IFN = interferon; IM = intramuscular.

ap = 0.01 vs. SC IFN␤-1b. b p = 0.0009 vs. IM IFN␤-1a; p = 0.0012 vs. IFN␤-1a 22 ␮g; p = 0.0003 vs. IFN␤-1b. c p < 0.0001 vs. IM
IFN␤-1a, IFN␤-1a 22 ␮g, IFN␤-1b; p = 0.02 vs. IFN␤-1a 44 ␮g. d p = 0.05 vs. IFN␤-1a 44 ␮g. e p ≤ 0.05 vs. IM IFN␤-1a and IFN␤-1b.

Table 2. Reasons for lack of adherence at baseline: treatment comparison

Reason, % patients IM IFN␤-1a IFN␤-1a 22 ␮g IFN␤-1a 44 ␮g IFN␤-1b GA

(n = 56) (n = 43) (n = 54) (n = 49) (n = 52)

Forgotten 0 16.3 11.1 12.2 13.5

Injection-related reasons 3.6 4.7 7.4 4.1 11.6
Flu-like symptoms 0 0 0 2.0 1.9
Weakness 0 0 0 2.0 1.9
Depression 0 0 1.9 2.0 0
Fatigue 0 0 0 2.0 1.9
Medication not collected 0 0 3.7 0 0
Did not feel the need to inject 0 2.3 0 0 0
No help for administration 1.8 0 0 0 0
Not confident about the treatment benefits 0 0 0 0 1.9

GA = Glatiramer acetate; IFN = interferon; IM = intramuscular.

tion). The median time on current therapy was 28.0 months
and the median disease duration was 2.3 years (table 1).
Patient Questionnaire
Adherence Rate
The overall adherence rate was 85.4% at baseline and
82.4% after 2 years (fig. 1). At the baseline visit, after al-
most 3 years on treatment, patients receiving IM IFN␤-1a
were significantly more adherent (96.4%) than those re-
ceiving SC IFN␤-1a 22 ␮g (79.1%; p = 0.0064), SC IFN␤-
1a 44 ␮g (79.6%; p = 0.0064) and glatiramer acetate
(82.7%; p = 0.0184) (fig. 2). The reasons for lack of adher-
ence at baseline are summarized in table 2. The overall
adherence rate at visit 1 was 86.6% and patients receiving
IM IFN␤-1a were significantly more adherent than those
receiving SC IFN␤-1a 22 ␮g (93.9 vs. 66.7%; p = 0.0251).
At visit 2, the overall adherence rate was 82.4%; this was
greater for IM IFN␤-1a (87.5%) versus the remaining
DMTs (SC IFN␤-1a 22 ␮g: 80%; SC IFN␤-1a 44 ␮g
(77.8%); IFN␤-1b (85.2%) and glatiramer acetate (80%).
After 2 years, 30.3% of all patients continued with the
same treatment as at baseline; 43% continued with IM
IFN␤-1a, a significantly higher percentage (p = 0.0103)
than with the other DMTs (fig. 3).
Hoffmann-La Roche & Co. AG, Hauptbibliothek
72.34.133.250 - 1/5/2019 4:09:06 AM

Two-Year Adherence to MS Therapy: Eur Neurol 2011;65:59–67 61

The GAP Study
Downloaded by:
 50
Non-adherent a
Adherent 45 43
100
40
14.6 13.4 17.6
90
35 32.3
80
30

Patients (%)
26.7 26.7
70
25
60
Patients (%)

20 17.5
50
15
85.4 86.6 82.4
40
10
30
5
20
0

IM IFN␤-1a

GA
IFN␤-1a 22

IFN␤-1b
IFN␤-1a 44
10

0
Baseline visit Visit 1 Visit 2

Fig. 1. Overall adherence rate at baseline and visits 1 and 2. Fig. 3. Percentage of patients who continued with the same treat-
ment at 2 years. a p = 0.0103 versus other treatments (abbreviations
as in fig. 2).

Baseline Visit 1 Visit 2

100 96.4 93.9
90.9
87.5 c 87.5 87.8 d
90 a 85.2 85.4 86.6
82.7 82.4
79.1 80 79.6 80
77.6 77.8
80
b

70 66.7
Adherence (%)

60

50

40

30

20

10

0
IM IFN␤-1a IFN␤-1a 22 IFN␤-1a 44 IFN␤-1b GA All

Fig. 2. Adherence rates by treatment at baseline and visits 1 and 2. GA = Glatiramer acetate; IM = intramus-
cular; IFN = interferon. a p = 0.0064 vs. IM IFN␤-1a; b p = 0.0251 vs. IM IFN␤-1a; c p = 0.0064 vs. IM IFN␤-1a;
d p= 0184 vs. IM IFN␤-1a.
Hoffmann-La Roche & Co. AG, Hauptbibliothek
72.34.133.250 - 1/5/2019 4:09:06 AM

62 Eur Neurol 2011;65:59–67 Arroyo et al.

Downloaded by:
 Independence afforded by the
4.1
treatment

Number of times a week I have to

3.5
administer medication

My ability to self-administer
3.6
the medication

Availability of a self-injection device

3.7
for administration of my medication

Route of administration, e.g.,

3.7
intramuscular or subcutaneous

Nurse administers my treatment 2.2

Support of my nurse to teach me

self-administration of the therapy 3.5

Medication reduces MRI lesions 4

Medication has few adverse reactions 3.8

Medication is well known 3.6

Medication produces fewer

neutralizing antibodies 3.6

Medication delays disease

progression 4.6

Medication reduces my relapses 4.6

Cost of the medication 2.9

How the medication works 4.4

0 1 2 3 4 5
Least Most
Fig. 4. Considerations highlighted by the important important
patients at visit 2.

Factors Related to Lack of Adherence
Treatment-Related Factors. At baseline, the most com-
mon reason cited for lack of adherence was forgetting
to inject (70.3%), followed by injection-related reactions
(43.2%) which encompassed the following aspects: tired
of self-injection, skin reactions, needle phobia, injection-
site pain, not feeling the need to inject and nobody avail-
able to administer the injections. At years 1 and 2, the
most common reason for lack of adherence was injection-
related factors (89.5 and 72%, respectively), followed by
forgetting to dose (42.1 and 32%, respectively). After 2
years of treatment, among the injection-related reasons,
being tired of injecting (28%) stands out.
Hoffmann-La Roche & Co. AG, Hauptbibliothek
72.34.133.250 - 1/5/2019 4:09:06 AM

Two-Year Adherence to MS Therapy: Eur Neurol 2011;65:59–67 63

The GAP Study
Downloaded by:
 Tired of taking
injections

New relapses

Injection
frequency

Fatigue

Flu-like
symptoms Overall

IM IFN␤-1a

Pain at IFN␤-1a 22
injection site
IFN␤-1a 44

IFN␤-1b
Needle phobia
GA

Skin reaction

0 20 40 60 80 100
Fig. 5. Considerations highlighted by the Patients (%)
neurologist at visit 2 (abbreviations as in
fig. 2).

The considerations with the highest score for the pa-
tients at 2 years when the MS treatment is chosen, on a
scale of 0–5 were: medication delays disease progression
(4.61), medication reduces relapses (4.5), how the medica-
tion works (4.3) and the independence afforded by the
treatment (4.1) (fig. 4).
Factors Related to the Healthcare Provider. The neu-
rologists and nurses of adherent patients saw the patients
more often and more regularly both at the baseline visit
(89.1 and 25.7%) and at 2 years (99 and 31%), respectively.
Treatment decisions on the type of DMT that the pa-
tient was to receive was an activity shared between the
neurologist and the patient; however, at the start of the
study, the neurologists’ decision had greater weight (in
13% of the cases, the patient did not take any decision
about treatment), whereas in the second year, the deci-
sion was shared equally.
Sociocultural Factors. At the baseline visit, 36% of the
patients worked full-time and 16% were retired or receiv-
ing assistance for MS. At 2 years, 32.2% of the patients
were working full-time versus 26.3% who were retired or
receiving assistance for MS at the baseline visit. Most pa-
tients were living with their spouse/partner or family
(parents or children) both at the baseline visit and at 2
years.
The main sources of patient support at the start and
after 2 years of the study were, respectively: family (87
and 85.8%), spouse/partner (84.8 and 83.6%), physician
Hoffmann-La Roche & Co. AG, Hauptbibliothek
72.34.133.250 - 1/5/2019 4:09:06 AM

64 Eur Neurol 2011;65:59–67 Arroyo et al.

Downloaded by:
Table 3. Time dedicated to the consultation with an MS patient at Among the support staff for the neurologist in clinical
baseline and visits 1 and 2 practice at the baseline visit and at 2 years, of note were
the nurses (83.3 and 85.7%), neuropsychologist (44.4 and
Diagnosis of Start of Routine
the disease treatment follow-up visits 71.4%), the physiotherapeutist (27.8 and 50.0%), and the
collaborating physician (50.0 and 42.9%).
Neurologist
Baseline visit 49.17% 35.83% 21.67%
Visit 2 50.71% 35.36% 20% Discussion
Nurse
Baseline visit 30.59% 49.12% 12.85%
Visit 2 12.65% 30.71% 12.14% The international GAP study [5] is being extended in
Spain for 5 years, and an interim analysis has been per-
formed at 2 years. At 2 years, an inflexion point in adher-
ence was observed, with a decrease across all treatments.
The adherence rate has remained high among all DMTs
for the first 2 years of the study, and was significantly
and nurse (80.6 and 77.4%), friends (69.6 and 69.2%), oth- higher with IM IFN␤-1a compared to other DMTs after
er healthcare workers (50.2 and 54.8%), and religious be- a median of more than 3 years of treatment. The number
liefs (49.4 and 48.4%). of patients who continued with the same treatment after
2 years was significantly higher with IM IFN␤-1a versus
Neurologist Questionnaire the other DMTs.
Adherence Rate The results of the baseline visit of the GAP study in
The estimated overall adherence according to the neu- Spain are consistent with the results from the interna-
rologist at baseline and visit 2 was 87.3 and 79.7% (SC tional study [5], with an overall adherence rate more than
IFN␤-1a: 93.9 and 90.9%; SC IFN␤-1a 22 ␮g: 88.5 and 10% higher (75% in the international GAP study [5] vs.
81.7%; SC IFN␤-1a 44 ␮g: 85.0 and 80.0%; IFN␤-1b: 82.8 85.4% in the Spanish GAP study). The distribution of the
and 75.4%, and glatiramer acetate: 86.5 and 71.1%, re- adherence rates in all treatments is also greater in the
spectively). Spanish GAP study. At the baseline visit, among factors
inherent to the medication, forgetting to inject was the
Factors Related to Lack of Adherence most common in both studies. Differences were found
Treatment-Related Factors. After 2 years of treatment, between the healthcare-provider-related and sociocul-
among the reasons related to lack of adherence, the neu- tural factors. At the baseline visit, unlike the internation-
rologists highlighted the development of new relapses al GAP study [5], no significant differences were found
(66.8%) (fig. 5). regarding sex and level of education of the patients. The
For the neurologist, the 5 attributes that were most involvement of the nurse in the follow-up of the patients
important for discussing with the patient when offering showed no changes between the baseline visit and visit 2,
treatment for MS at the baseline visit and visit 2 were ef- remaining steady at 12% (table  3). In contrast, after 2
fectiveness at preventing relapses (94.4 and 92.9%, re- years, an increase in multidisciplinary follow-up of the
spectively) and disease progression (77.8 and 92.9%, re- patients in the form of incorporation of neuropsycholo-
spectively), factors related to adverse effects of treatment gists and physiotherapists was detected compared to the
(88.9 and 92.9%, respectively), frequency of administra- baseline visit. In view of the decrease in overall adherence
tion (44.4 and 42.9%, respectively), administration skill rate at visit 2, it might be necessary to review the role of
(44.4 and 50.0%, respectively) and disease activity ac- nursing staff in follow-up. Of note is the neurologists’
cording to magnetic resonance imaging (44.4 and 42.9%, perception of adherence, which is almost 3 points lower
respectively). than the real patient adherence. This perceived adherence
Factors Related to the Healthcare Provider. At the base- is only higher in the case of IM IFN␤-1a and SC IFN␤-1a
line visit and visit 2, the time dedicated by the neurologist 44 ␮g, suggesting that the neurologists overestimate real
to diagnosis of the disease, starting treatment, and rou- adherence in their patients with these treatments and
tine follow-up visits was similar, whereas the nurse dedi- dedicate them less time. After 2 years of follow-up, the
cated more time to the patients at the baseline visit com- partner, family members and religious beliefs continue to
pared to follow-up visits (table 3). play an important role.
Hoffmann-La Roche & Co. AG, Hauptbibliothek
72.34.133.250 - 1/5/2019 4:09:06 AM

Two-Year Adherence to MS Therapy: Eur Neurol 2011;65:59–67 65

The GAP Study
Downloaded by:
 One of the limitations of this study is the way adher-
ence is measured. Observational studies usually have to
rely on measures susceptible to bias such as patient recall
of missed doses in the past 4 weeks, as is the case in the
present study. Given that patients with MS can experi-
ence progressive cognitive decline [11], recall could be in-
fluenced by memory decline. In English-speaking coun-
tries of the international GAP study [5], the MS Neuro-
psychological Screening Questionnaire was administered
to patients; unfortunately, these data were not available in
our subpopulation. It is not clear how a decline in cogni-
tive function would affect recall of adherence, that is,
whether patients with cognitive decline are more likely to
overestimate or underestimate adherence. We do note,
however, a greater apparent difference in the neurolo-
gist’s assessment of adherence (87.3% at baseline vs. 79.7%
after 2 years) as compared to the patient assessment (85.4
vs. 82.4%). Cognitive decline may also have a direct influ-
ence on adherence [12].
Another of the limitations of this study is the loss to
follow-up (48.4%) either due to personal decision of the
neurologist (in 1 center) or due to the decision of the cen-
ters themselves (in 2 centers). This resulted in a loss of
statistical power making a ␤-type error possible. The
groups were fairly homogeneous at baseline. Certain sig-
nificant differences were only found in the glatiramer ac-
etate group, the SC IFN␤-1a 44 ␮g group and the IFN␤-
1b group. These differences can be explained by the
shorter time on the market of glatiramer acetate and SC
IFN␤-1a 44 ␮g, and the longer time on the market in the
case of IFN␤-1b. However, at the different visits, adjust-
ments between groups were not made for baseline values,
essentially because of the obvious decrease in sample size.
Other factors that might complicate analysis of the study
include the lack of randomization, differences in treat-
ment duration and withdrawals or changes in treatment,
which are typical occurrences in observational studies.
The present study has shown that the therapeutic op-
tion may directly affect adherence of an MS patient to
DMT.
The most common reasons for lack of adherence were
forgotten doses and injection-related factors, with such
factors predominating during follow-up (particularly
getting tired of injection), probably due to the treatment
duration and chronic nature of the disease.
Frequency of drug administration, side effects of the
medication and the patients’ perception of treatment ef-
ficacy were treatment-related factors found to affect lack
of adherence. These factors should be considered when
making therapeutic choices. Healthcare staff should deal
66 Eur Neurol 2011;65:59–67
with these factors that influence adherence, maintaining
fluid dialogue with the patient and monitoring aspects
related to the safety of the drug. Although these are dy-
namic factors, if they are managed effectively, there
would be greater emphasis on patient education and long-
term adherence would be enhanced.
The results of the Spanish GAP study, along with the
studies conducted until present [6, 10], confirm the need
to be aware of the importance of adherence to MS therapy
with DMTs. The use of questionnaires enabled assessment
of the views of the neurologists and patients on factors
that may influence adherence. This shows that in general,
adherence in Spain is higher than in other countries and
remains high after 2 years of follow-up, remaining above
the international baseline level despite a decrease of 3%.
Follow-up of the GAP study in Spain has enabled us to
verify the importance of certain health and sociocultural
factors described previously, as well as aspects of quality
of life and patient and neurologist satisfaction, which will
be published separately. In Spain, of particular note is the
support of the partner, family and healthcare staff (neu-
rologist, nursing staff) as factors that have a positive influ-
ence on treatment adherence. After 2 years of follow-up,
multidisciplinary staff such as neuropsychologists and
physiotherapists played an increasing role in the care of
patients with MS. Neurologists were also found to remain
involved in patient follow-up. We believe that these factors
contribute positively to the high adherence to DMTs seen
in this country. The positive correlation of IM IFN␤-1a
with adherence is due, we believe, to the lower frequency
of injection compared to other DMTs. We wish to high-
light that in 2 years of follow-up, forgetting to inject went
from being the principal reason for poor adherence to sec-
ond place (after injection-related factors), with a 38.3% de-
crease. We therefore believe that the greater awareness
among healthcare staff and patients in the study is having
a positive effect on remembering to inject the medication.
To date, as far as we are aware, this is the only obser-
vational study of adherence to DMTs that is being con-
ducted in MS patients in Spain. Observational studies
with long-term follow-up have the added value of being
able to compare the results of clinical trials with daily
clinical practice and analyze aspects relating to effective-
ness, which are not studied in clinical trials. This has
prompted us to draw up adherence questionnaires which
will be published separately. We consider it necessary to
develop tools that help healthcare staff assess adherence
to DMTs in order to anticipate any problems that patients
might have with specific therapeutic regimens and DMTs
in general. This would help optimize the outcomes of
Hoffmann-La Roche & Co. AG, Hauptbibliothek
72.34.133.250 - 1/5/2019 4:09:06 AM
Arroyo et al.
Downloaded by:
current DMTs. These questionnaires would also provide
healthcare staff with the most appropriate measures for
improving follow-up of therapy in MS patients, particu-
larly in those at high risk of lack of adherence and in those
with longer disease and treatment durations, in whom
adherence has been found to be poorer.
In conclusion, in the Spanish GAP study, the overall
rate of adherence remained high over the 2 years of fol-
low-up reported here, and higher than the baseline ad-
herence recorded in the international GAP study [5]. It
was highest with IM IFN␤-1a; this difference was signif-
icant at the baseline visit (after nearly 3 years on therapy)
compared with other DMTs. It would be appropriate to
strengthen the role of nursing staff in the follow-up of
patients receiving DMTs. Treatment adherence is also
improved if both patients and healthcare providers are
aware of its importance.
Appendix 1
GAP Investigators
Ramo C., Grau L. (H. Universitario Germans Trias i Pujol,
Badalona); Roquer J., Munteis E., Martínez E. (H. Del Mar, Bar-
celona); Ramió i Torrento L.L., Merchán M. (H. Universitario de
Girona Dr. Josep Trueta, Girona); Brieva L. (H. Universitario Ar-
nau de Vilanova, Lleida); Lacruz F., Bujanda M. (H. De Navarra,
Pamplona); de Castro P. (Clínica Univ. De Navarra (Pamplona);
Marzo E. (Complejo H. San Millán-San Pedro, Logroño); Hernán-
dez M.A., González M. (H. Ntra. Sra. de Candelaria, Tenerife); de
Andrés C., Martínez M. (H. Universitario Gregorio Marañón,
Madrid); Izquierdo G. (Complejo Hospitalario Virgen Macarena,
Sevilla); Sánchez F., Agüera E., Sánchez V. (H. Reina Sofía de Cór-
doba, Córdoba); Guerrero M. (H. Univer. San Cecilio, Granada);
Arnal C. ( H. Universitario Virgen de las Nieves, Granada); Bowa-
kin D., Renedo F. (Hospital Universitario ‘Del Río Hortega’, Val-
ladolid); Cacho J. (Complejo Asistencial de Salamanca, Salaman-
ca); Yusta A. (H. Univer. De Guadalajara, Guadalajara); Pérez A.
(H. General Universitario de Alicante, Alicante); Belenguer A. (H.
General de Castellón, Castellón).
Acknowledgments
The GAP study was sponsored by Biogen Idec. Four of the au-
thors (E.A., C.G., J.P., O.S.S.) worked for Biogen Idec. C.R.-T. has
received lecture fees and grant support from Bayer-Schering, Bio-
gen Idec, Merck-Serono, and Teva Neuroscience. We are indebted
to the GAP investigators who are listed in Appendix I. Editorial
support was provided by Dr. Gregory Morley (funded by Biogen
Idec Iberia). The authors would like to thank all the investigators
and patients who participated in the study.

References
1 World Health Organization Report: Adher-
ence to Long-Term Therapies. Geneva,
WHO, 2003.
2 Tremlett HL, Ogel J: Interrupted therapy:
stopping and switching of the ␤-interferons
prescribed for MS. Neurology 2003; 61: 551–
554.
3 Río J, Porcel J, Téllez N, Sánchez-Betancourt
A, Tintoré M, Arévalo MJ, Nos C, Montalban
X: Factors related with treatment adherence
to interferon-␤ and glatiramer acetate ther-
apy in multiple sclerosis. Mult Scler 2005;11:
306–309.
4 Taus C, Taffi R, Morelli M, Fie A, Angeleri
VA, Sgolastra M: Adherence to IFN-␤ thera-
py in MS: an Italian clinical setting. Mult
Scler 2001;7(suppl 1):S65.
5 Devonshire V, Lapierre Y, Macdonell R, et al:
The Global Adherence Project (GAP): a mul-
ticenter observational study on adherence to
disease-modifying therapies in patients with
relapsing-remitting multiple sclerosis. Eur J
Neurol 2010 (in press).
6 Frohman E, Phillips T, Kokel K, Van Pelt J,
O’Leary S, Gross S, Hawker K, Racke M: Dis-
ease-modifying therapy in multiple sclero-
sis: strategies for optimizing management.
Neurologist 2002;8:227–236.
7 Treadaway KD, Brannon KD, Morrison A,
Remmington GM: Factors that influence ad-
herence with disease-modifying therapy
in multiple sclerosis (abstract). Neurology
2006;66(suppl 2):A30.
8 Holland N, Wiesel P, Cavallo P, Edwards C,
Halper J, Kalb R, Morgante L, Namey M,
O’Leary M, Smith-Williamson L: Adherence
to disease-modifying therapy in multiple
sclerosis: part I. Rehabil Nurs 2001; 26: 172–
176.
9 Holland N, Wiesel P, Cavallo P, Edwards C,
Halper J, Kalb R, Morgante L, Namey M,
O’Leary M, Smith-Williamson L: Adherence
to disease-modifying therapy in multiple
sclerosis: part II. Rehabil Nurs 2001;26:221–
226.
10 Fiore AP, Fragoso YD: Tolerability, adverse
events and compliance to glatiramer acetate
in 28 patients with multiple sclerosis using
the drug continuously for at least six months.
Arq Neuropsiquiatr 2005;63:738–740.
11 Duque B, Sepulcre J, Bejarano B, et al: Mem-
ory decline evolves independently of disease
activity in MS. Mult Scler 2008;14:947–953.
12 Bruce JM, Hancock LM, Arnett P, Lynch S:
Treatment adherence in multiple sclerosis:
association with emotional status, personal-
ity, and cognition. J Behav Med 2010;33:219–
227.
Hoffmann-La Roche & Co. AG, Hauptbibliothek
72.34.133.250 - 1/5/2019 4:09:06 AM

Two-Year Adherence to MS Therapy: Eur Neurol 2011;65:59–67 67

The GAP Study
Downloaded by: