http://informahealthcare.com/jas
ISSN: 0277-0903 (print), 1532-4303 (electronic)
J Asthma, 2013; 50(7): 776–782
! 2013 Informa Healthcare USA, Inc. DOI: 10.3109/02770903.2013.803116

ADHERENCE, CONTROL AND QUALITY OF LIFE

An evaluation of asthma medication utilization for risk evaluation and
mitigation strategies (REMS) in the United States: 2005–2011
Rachael L. DiSantostefano, Anne M. Yeakey, Ibrahim Raphiou, and David A. Stempel

Abstract

Keywords

Purpose: The purpose of this study was to assess drug utilization patterns of fluticasone
propionate (FP)/salmeterol (SAL) combination (FSC) and SAL over the 7-year period of 2005–
2011 in patients with asthma as part of the Risk Evaluation and Mitigation Strategies (REMS).
Methods: A descriptive, retrospective observational study utilizing national pharmacy data and
employer-based claims data to characterize drug utilization patterns. Results: For patients with
asthma, the total number of FSC and SAL dispensings and users of FSC and SAL has declined
between 2005 and 2011. During this period, FSC and SAL dispensing for asthma decreased 24%
and 76%, respectively, with a more pronounced decline between 2010 and 2011 relative to
other years. The total number of patients with asthma who were dispensed FSC has decreased
by 10% among adults and by 40% in children and adolescents. While SAL-containing
medications decreased, dispensing of FP monotherapy increased 39% during the same 7-year
period. The number of patients dispensed FP for asthma has increased 47% in children 4–11
years of age, 72% in adolescents 12–17 years of age, and 6% in adults. SAL use without a
controller was infrequent and decreasing, reported by 1.7% and 0.5% of patients with asthma in
2005 and 2011, respectively. Conclusions: In patients with asthma, use of FSC and SAL decreased
between 2005 and 2011, while the use of FP increased. Use of SAL monotherapy was infrequent
and declined during the study period. The data suggest that the substantial communication
activities have encouraged appropriate prescribing of long-acting b2-adrenergic agonist (LABA).

Drug utilization, fluticasone propionate/
salmeterol combination, salmeterol,
fluticasone propionate

Introduction
Two large clinical trials reported data on a potential
association between salmeterol (SAL; trade name
SEREVENT), a long-acting b2-adrenergic agonist (LABA),
and an increased risk of asthma-related death [1,2]. Subjects
randomized in these trials were requested to continue their
other asthma medications but were dispensed SAL or placebo.
In the Salmeterol Multicenter Asthma Research Trial, fewer
than 50% of subjects reported using a concomitant inhaled
corticosteroid (ICS). Despite the increased risk of asthma
fatality in patients with concomitant ICS þ LABA, the trial
was not prospectively designed to directly answer this
question. As a result of these studies and an extensive
literature review, the United States treatment guidelines
recommend that LABAs should only be used in combination
with ICS for the treatment of asthma in patients with
moderate or severe persistent asthma requiring more than
low- to medium-dose ICS [3]. Since 2003, GlaxoSmithKline
and the United States Food and Drug Administration (FDA)
have initiated a series of activities to better understand the
Correspondence: Rachael DiSantostefano, GlaxoSmithKline, Research
and Development, 5 Moore Drive, Research Triangle Park, North
Carolina 27 709-3398, USA. Tel.: +1 919-483-9237. Fax: +1 919-3154947. E-mail: Rachael.L.DiSantostefano@gsk.com

History
Received 21 February 2013
Revised 23 April 2013
Accepted 2 May 2013
Published online 18 July 2013

safe use of LABAs, and a series of measures were
implemented to communicate this information to prescribers.
These activities have included four FDA advisory committee
meetings; a boxed warning with several revisions and updated
labeling for LABA-containing medications; an FDA-conducted meta-analysis of safety data; two FDA-issued communications on the safe use of LABAs [4,5] and a class-level
Risk Evaluation and Mitigation Strategy (REMS) for LABAcontaining medications [6]. In preparation for the 2008
advisory committee, the FDA-conducted meta-analysis noted
an increased risk of asthma-related hospitalizations associated
with LABA use, particularly in children [7]. These activities
are discussed in more detail elsewhere [8].
The REMS for LABA-containing medications included a
communication plan with letters to current and potential
prescribers of LABAs, printed/web-based information for
healthcare providers, and letters to the leadership of relevant
professional societies [6]. Under the REMS, the FDA issued
revised prescribing information which no longer characterized
LABAs as long-term maintenance medications for asthma
treatment and encouraged discontinuing LABA use if the
patient’s asthma could be controlled with ICS monotherapy.
Additionally, the LABA prescribing information was updated
to indicate that the available data were inadequate to
determine whether concurrent use of ICS or other long-term

20
13

J Asthma Downloaded from informahealthcare.com by Jennifer Stahl on 05/12/15
For personal use only.

GlaxoSmithKline, Research Triangle Park, North Carolina, USA

 Asthma REMS drug utilization

DOI: 10.3109/02770903.2013.803116

asthma medication mitigates the observed increased risk of
asthma-related death associated with LABAs.
As part of an assessment of whether the REMS [9,10] was
meeting its goals, GlaxoSmithKline examined drug utilization
trends to evaluate prescribing patterns over time. The primary
objective was to characterize drug utilization trends for
fluticasone propionate (FP; trade name FLOVENT)/SAL
combination (FSC; trade name ADVAIR) and SAL, between
2005 and 2011 in the United States among patients with
asthma. Secondary objectives were to characterize drug
utilization class-level trends for ICS/LABA medications and
to characterize drug utilization for FP. This analysis was
submitted to the FDA in March 2012, and the FDA released
these medications from their REMS in August 2012 [6].

J Asthma Downloaded from informahealthcare.com by Jennifer Stahl on 05/12/15
For personal use only.

Methods

777

OptumÔ ImpactÔ National Managed Care Benchmark
Database [15] (OptumInsight, Eden Prairie, Minnesota, USA)
was used to obtain measures of adherence based on medication dispensing. This national database comprises claims from
more than 50 sources and covers more than 20 million lives,
yielding patient-level healthcare claims data.
Definitions
Concurrent dispensing
Concurrent use of SAL with a controller was tabulated using
Source Healthcare Analytics data. Concurrent use was defined
as controller dispensing identified within 90 days of a LABA
dispensing ( 90 days) to account for variable adherence with
respiratory medications. Categories included FSC, SAL with
ICS, SAL with other controller, and SAL monotherapy.

Study design

Length of therapy

This was a retrospective data analysis using healthcare claims
data to summarize asthma drug utilization trends from 2005
to 2011 by age categories for LABA-containing and ICS
medications. Analyses were conducted at the aggregated level
for prescription dispensing and at the patient-level for
concurrent medication use and length of therapy. This
analysis was to support the GlaxoSmithKline REMS, and
therefore GlaxoSmithKline medications including FSC, SAL
and FP were the medications of interest; however, class-level
analyses are also provided in the online appendix
(Supplemental Table S1).

To examine the length of therapy, the proportion of days
covered and the medication possession ratio in a 1-year period
were calculated using OptumÔ InsightÔ employer-based
claims data. The proportion of days covered provides the
length of time a patient is covered with the medication for a
1-year period, and was calculated as the sum of the days
supplied within the 1-year period from index prescription to
the end of the period divided by 365 days [15]. To be included
in the proportion of days covered tabulations, patients were
required to have at least one medication fill for the medication
of interest during the 1-year study period. The medication
possession ratio was calculated as the number of days supply
dispensed from the first index fill to the last fill (not including
the days supply of the last medication fill) divided by the days
between the first and last fill in a 1-year time period [16].
Patients were required to have at least two medication fills,
with the first and last dispensing at least 6 months apart in
order to measure persistence over time. By requiring at least
two prescriptions, medication possession ratio measures tend
to be skewed to more adherent patients relative to the
proportion of days covered.
The identification period for these measures was 1 January
2010 to 30 June 2011. The first prescription fill in the first 6
months of this period was selected as the index prescription.
Patients were required to have continuous health plan
coverage for 12 months from the index date and a diagnosis
of asthma in the 6 months prior to or during the 12 months
after the index date.
Assumptions required for the length of therapy measures
included: (1) Days supplied for dispensed medications that
overlap with the previous fill were added to the end of the
days supplied rather than truncating the prior dispensing. (2)
If there was a dispensing at the end of the period that went
past the 365 day interval, the coverage was truncated to
coincide with the 365th day. (3) Medication fills with missing
days supplied were deleted from the analysis due to potential
data errors.

Data sources
Drug utilization trends were determined using several data
sources. Source Healthcare Analytics (SHA; Phoenix,
Arizona, USA) prescription-level data include estimates of
prescription-level data representing 82% coverage of retail
pharmacy dispensing, and 50–80% of mail order (depending
on product) in the United States [11]. Patient-level originates
from 75 million covered lives per year, representing 75%
coverage of retail pharmacy dispensing, 55–80% of mail order
(depending on product) and 60% of the long-term care
dispensing [12]. Source Healthcare Analytics data were used
to estimate patient counts and concurrent use of LABA and
other controllers.
IMS health data (IMS health, Danbury, Connecticut, USA),
including the IMS National Prescription AuditÔ [13] and the
National Disease and Therapeutic IndexÔ [14], were used to
determine the market shares of medications within the asthma
market. The National Prescription AuditÔ measures volume
and shares of medications dispensed in the United States based
on a representative sample of the prescribing universe,
including 74% of retail pharmacies, 40–70% of non-governmental mail service pharmacy outlets, and 45–50% of longterm care facilities. The National Disease and Therapeutic
IndexÔ includes disease encounters in 4 140 office-based
practices in the continental United States, which provide
treatment dispensings by diagnosis. IMS data were combined
with Source Healthcare Analytics data to estimate the total
number of prescription dispensings by class within the asthma
market.

Statistical analysis
The number and proportion (%) of dispensing and patients
was assessed annually by age (4–11 years, 12–17 years,  18

 778

R. L. DiSantostefano

years) at the class-level and GlaxoSmithKline medication
level. The number of patients with concurrent prescription of
SAL with and without a controller was tabulated by age for
patients with asthma. No statistical testing was performed.

Results
Total dispensing

J Asthma Downloaded from informahealthcare.com by Jennifer Stahl on 05/12/15
For personal use only.

The total number of dispensings of FSC, FP and SAL for
asthma from 2005 to 2011 is presented in Figure 1 and
Supplemental Table S2. Dispensing of FSC and SAL
decreased during this time period with a 24% reduction of
FSC dispensing (12 486 637 to 9 512 223 dispensing), and a
76% reduction in SAL dispensing (counts decreasing from
847 200 to 203 149 dispensing). In contrast, dispensing of FP
increased 39% (3 449 461 to 4 791 544 dispensing) during the
same period. FSC and SAL dispensing decreased annually by

J Asthma, 2013; 50(7): 776–782

an average of 3.4% and 11%, respectively, and the decrease
was most pronounced between the years 2010–2011.
GlaxoSmithKline-level patient information
The calculated number of patients using FSC, FP and SAL for
asthma between 2005 and 2011, by age group are presented in
Figure 2 for children aged 4–11, Figure 3 for adolescents aged
12–17, and Figure 4 for adults aged  18; and in Supplemental
Table S3. The patient data follow the same trends as the
prescription dispensing data. The number of patients with
asthma dispensed FSC and SAL has decreased for all age
groups since 2005, and the greatest decrease was observed in
children and adolescents.
In this analysis, the number of patients with asthma treated
with FSC decreased 48% in children, 44% in adolescents, and
10% in adults between 2005 and 2011. In contrast, the number
of patients dispensed FP between 2005 and 2011 increased

Figure 1. –Drug utilization: total number of dispensings of GlaxoSmithKline medicines for patients with asthma during 2005–2011, by year. FP,
fluticasone propionate; FSC, fluticasone propionate/salmeterol combination; SAL, salmeterol. Sources: IMS Health, IMS National Prescription
AuditÔ, IMS National Disease and Therapeutic IndexÔ, and GlaxoSmithKline Custom Asthma/COPD Hierarchical Regimen Analysis Tool.

Figure 2. –Drug utilization in total number of patients with asthma aged 4–11 years, for GlaxoSmithKline medications during 2005–2011. Note: FSC
line is overlapped by SAL-containing line in the graph above. FP, fluticasone propionate; FSC, fluticasone propionate/salmeterol combination; SAL,
salmeterol. Sources: Source Healthcare Analytics, GlaxoSmithKline Custom Asthma/COPD Hierarchical Regimen Analysis Tool and Annual Unique
Patient Counts.

 DOI: 10.3109/02770903.2013.803116

47% in children, 72% in adolescents, and 6% in adults. SAL
monotherapy was infrequent among patients with asthma
(52%).
Concurrent salmeterol with controllers in asthma
The use of concurrent SAL with a controller as FSC or SAL
plus a controller has increased for all age groups since 2005
(98.2% in 2005 vs. 99.8% in 2011 in children; 97.5% in 2005
vs. 99.5% in 2011 in adolescents; 91.9% in 2005 vs. 97.3% in
2011 in adults). The use of SAL alone has decreased since
2005 across all age groups as presented in Table 1. When
dispensed SAL, the majority of patients with asthma were
dispensed a concurrent controller medication (e.g., ICS or
other therapy such as a leukotriene receptor antagonist).
Length of therapy

J Asthma Downloaded from informahealthcare.com by Jennifer Stahl on 05/12/15
For personal use only.

Medication persistence: proportion of days covered
Medication persistence measured as a proportion of days
covered during a 1-year period for FSC, FP and SAL is
presented in Table 2. Among the 74 070 patients treated
with FSC, the average proportion of days covered was 43%
for asthma in 1 year. Results were lower in children

Asthma REMS drug utilization

779

and adolescents. Proportion of days covered for FP was
28% for all patients with asthma (n ¼ 40 198), and similar in
all age groups. Five percent of all patients with asthma using
FP had  80% coverage during the 1-year study period.
Proportion of days covered for SAL was 54% for all patients
with asthma (n ¼ 1 522; mostly adults).
Medication adherence: medication possession ratio
Medication adherence measured as medication possession
ratio during a 1-year period for FSC, FP and SAL is presented
in Table 3. Patients were required to have more than one
prescription within the 1-year interval for this analysis,
resulting in smaller sample sizes but higher measures of
adherence than proportion of days covered.
The average medication possession ratio for patients
treated with FSC was 56%. MPR in children and adolescents
was lower: 50% and 48% respectively. Twenty-four percent of
all patients with asthma had a medication possession ratio
 80%, with 16% of children and 15% of adolescents having a
medication possession ratio of  80%.
Medication possession ratio for FP was 43% for all patients
with asthma (n ¼ 26 813), and was similar among age groups.

Figure 3. –Drug utilization in total number of patients with asthma aged 12–17 years, for GlaxoSmithKline medications during 2005–2011. Note: FSC
line is overlapped by SAL-containing line in the graph above. FP, fluticasone propionate; FSC, fluticasone propionate/salmeterol combination; SAL,
salmeterol. Sources: Source Healthcare Analytics, GlaxoSmithKline Custom Asthma/COPD Hierarchical Regimen Analysis Tool and Annual Unique
Patient Counts.

Figure 4. –Drug utilization in total number of patients with asthma  18 years old, for GlaxoSmithKline medications during 2005–2011. FP,
fluticasone propionate; FSC, fluticasone propionate/salmeterol combination; SAL, salmeterol. Sources: Source Healthcare Analytics, GlaxoSmithKline
Custom Asthma/COPD Hierarchical Regimen Analysis Tool and Annual Unique Patient Counts.

 780

R. L. DiSantostefano

J Asthma, 2013; 50(7): 776–782

Table 1. Drug utilization in patient counts for concurrent GlaxoSmithKline medicines at age group level in patients with asthma, by year.
Patient counts

J Asthma Downloaded from informahealthcare.com by Jennifer Stahl on 05/12/15
For personal use only.

Asthma 4–11
FSC
SAL alone
ICS þ SAL
SAL þ other
Total
Asthma 12–17
FSC
SAL alone
ICS þ SAL
SAL þ other
Total
Asthma 18þ
FSC
SAL alone
ICS þ SAL
SAL þ other
Total

Total patient count (%)
2005
4 457 88
864
5351
1960
453 963
2005
434 181
2218
5397
3363
445 159
2005
3 565 747
64 218
160 939
90 102
3 881 006

2006
375 352
331
3 046
859
379 588
2006
387 077
954
3223
1739
392 993
2006
3 489 736
44 485
121 295
62 548
3 718 064

2007
320 000
233
1 778
686
322 697
2007
346 532
646
2 371
1 214
350 763
2007
3 448 009
34 272
96 310
50 245
3 628 836

2008
293 092
167
1160
391
294 810
2008
313 842
460
1 717
796
316 815
2008
3 489 532
30 786
80 550
42 760
3 643 628

2009
268 970
81
832
229
270 112
2009
302 636
344
1 177
519
304 676
2009
3 421 194
23 613
63 639
33 339
3 541 785

2010
257 862
59
580
168
258 669
2010
284 174
216
939
335
285 664
2010
3 356 553
20 040
53 840
27 562
3 457 995

2011
230 514
65
321
97
230 997
2011
243 686
175
807
307
244 975
2011
3 195 260
17 962
47 387
24 387
3 284 996

2005*
98.2%
0.2%
1.2%
0.4%
100%
2005 (%)
97.5%
0.5%
1.2%
0.8%
100%
2005 (%)
91.9%
1.7%
4.1%
2.3%
100%

2011*
99.8%
0.0%
0.1%
0.0%
100%
2011 (%)
99.5%
0.1%
0.3%
0.1%
100%
2011 (%)
97.3%
0.5%
1.4%
0.7%
100%

Sources: Source Healthcare Analytics, GlaxoSmithKline Custom Asthma/COPD Hierarchical Regimen Analysis Tool and Annual Unique
Patient Counts. FSC, fluticasone propionate/salmeterol combination; ICS, inhaled corticosteroids; Other, other controller; SAL, salmeterol.
*Expressed as a % of total patients.

Table 2. Proportion of days covered in 1 year among patients with asthma.
Asthma

FSC (n)
Avg. proportion of days covered
% at  80%
% at  70%
FP (n)
Avg. proportion of days covered
% at  80%
% at  70%
SAL (n)
Avg. proportion of days covered
% at  80%
% at  70%

Ages 4–11

Ages 12–17

Ages 18þ

All patients

5 023
0.36
8
13
15 121
0.27
3
6
10
0.19
n/a
n/a

6 502
0.34
7
11
5 231
0.21
2
3
21
0.24
5
10

62 434
0.45
17
23
17 422
0.30
7
10
1 491
0.55
30
37

74 070
0.43
15
21
40 198
0.28
5
8
1 522
0.54
29
36

Source: OptumÔ ImpactÔ National Managed Care Benchmark Database. Avg., average; FP, fluticasone propionate; FSC, fluticasone propionate/
salmeterol combination; SAL, salmeterol.

Less than 20% of asthma patients using FP had  80%
coverage during the 1-year study period. Medication possession ratio for SAL was 69% for all patients with asthma
(n ¼ 1253), which was an adult population with only 17
combined children and adolescents with more than one SAL
dispensing in the database.

Discussion
This study provides information on United States drug
utilization patterns for FSC, SAL and FP, as well as the
overall class of these respiratory medications between 2005
and 2011. The results suggest that the series of communications regarding LABA safety in the United States, prior to and
including the REMS, may have contributed to decreased
SAL-containing medication use and increased use of FP
monotherapy. The use of FSC and SAL in asthma has
declined and the use of FP has increased between 2005 and
2011. SAL was used in a fixed dose combination or with

another controller for the vast majority of patients with
asthma. The measures of persistence and adherence for the
medications studied suggest that these medications were
commonly used intermittently as opposed to continuously,
suggesting patients or healthcare professionals may have
down titrated these medications when patients’ asthma was
controlled.
Frequent and concurrent use of SAL with another antiinflammatory controller has been reported previously in the
United Kingdom [17]. In patients receiving SAL prescriptions
(excluding FSC), 90% of patients initiating SAL used a
concurrent controller within 90 days of the index date. When
accounting for all SAL-containing prescriptions (SAL and
FSC), 95% of all patients, and 97% of children 4–11 years of
age, were prescribed concurrent controllers.
Others have demonstrated that LABA monotherapy is
infrequent in patients with asthma. In a Medicaid population
between 2006 and 2008, the use of LABA monotherapy was
uncommon (51% in patients with asthma), decreased with

 Asthma REMS drug utilization

DOI: 10.3109/02770903.2013.803116

781

Table 3. Medication possession ratio in 1 year among patients with asthma.
Asthma

FSC (n)
Avg. medication possession ratio
% at  80%
% at  70%
FP (n)
Avg. medication possession ratio
% at  80%
% at  70%
SAL (n)
Avg. medication possession ratio
% at  80%
% at  70%

Ages 4–11

Ages 12–17

Ages 18þ

All patients

3 862
0.50
16
24
10 759
0.40
9
14
5
0.51
20
20

4 856
0.48
15
22
3 073
0.38
9
13
12
0.47
25
33

50 757
0.57
26
34
11 252
0.47
18
23
1 236
0.69
45
52

59 555
0.56
24
32
26 813
0.43
13
18
1 253
0.69
45
52

J Asthma Downloaded from informahealthcare.com by Jennifer Stahl on 05/12/15
For personal use only.

Source: OptumÔ ImpactÔ National Managed Care Benchmark Database. Avg., average; FP, fluticasone propionate; FSC, fluticasone propionate/
salmeterol combination; SAL, salmeterol.

time, and was less frequent in children [18]. In a United
States healthcare claims dataset, LABA monotherapy
comprised 9% and 3% of LABA-containing medications in
the first half of 2003 and 2009, respectively [19]. Infrequent
use of LABA has also been observed in Scotland, where
51.5% of patients with asthma (991 of 73 486) used LABA
as monotherapy between 2005 and 2007 [20]. In the Scottish
study, use of sustained LABA monotherapy was more
frequent in patients 460 years of age, and sustained LABA
monotherapy was significantly reduced with primary care
asthma reviews. Similar to our findings with proportion of
days covered, Stempel et al. demonstrated that FSC was
filled more frequently than FP [21]. The mean number of
medication dispensings obtained using FSC during a 1-year
study (3.98; 95% confidence interval [CI] 3.78–4.18), was
nearly double that observed in the FP cohort (2.29; 95% CI
2.11–2.47). The difference in the number of dispensings and
proportion of days covered between FSC and FP likely
reflect the differences in severity or seasonality of disease
for treated patients. Consistent with the treatment guidelines,
patients dispensed FSC were more likely to have indicators
of more severe or uncontrolled asthma and require more
medication than those patients who were dispensed ICS
monotherapy.
Drug utilization analyses are useful to characterize trends
of medication use over time. However, there are limitations in
interpreting the results of these studies based on available
data. Drug utilization studies are limited by the information
recorded and available for analyses in retrospective healthcare
databases. Prescription-level trend data (as presented here)
have a relatively short lag (2–4 months) which allows us to
examine data through 2011; however, it is limited by the lack
of patient-level information that would allow for tabulation
beyond indication and age. Alternatively, healthcare encounters that require payment are well captured in most healthcare
claims and electronic medical record databases; yet, these
data may lag up to 18 months. Therefore, trends in LABAcontaining medication use following the 2010 LABA safety
communication and labeling changes have not yet been
evaluated using these sources [19]. Unfortunately, nearly all
databases are deficient in either systematically collected
measures of asthma symptoms or patient-reported level of

asthma control and cannot provide insight into the physician’s
judgment that might explain the reasons for the observed
trends in prescribing these medications (e.g., reasons for
patients maintaining, stepping up, or switching various
asthma medications).
Others have evaluated appropriate use of LABAs based on
disease severity, and considered the use of LABAs appropriate if patients had prior single controller use (ICS, leukotriene
receptor agonist), prior healthcare utilization, courses of oral
corticosteroids, or excessive short acting b2-adrenergic
agonist use in the year prior to initiation of ICS/LABA
combinations. Using healthcare claims data, the authors could
explain less than half of the LABA-containing medications as
being appropriate in the context of their criteria, and cited
similar limitations of databases regarding the absence of
symptoms and physician judgment [19,22].

Conclusions
The use of FSC and SAL for asthma has declined progressively since the dissemination of targeted information
describing the safety risks associated with the use of LABA
from 2005 to 2011, with the largest annual decline observed
between 2010 and 2011. Conversely, during the same period,
the use of FP and other ICS monotherapy have increased. In
addition, the majority of LABA use was concurrent with a
controller, typically in the same device such as FSC, or less
commonly LABA used with a separate controller. Finally,
only 15% of patients dispensed FSC had high levels of
adherence (proportion of days covered  80%) during the 1year study period, and the average proportion of days covered
was 43%, suggesting that FSC may be down titrated when
patients achieve asthma control. These data suggest that the
substantial communication about LABA safety, prior to and
including the REMS, have resulted in a reduction of FSC use
and encouraged appropriate use of LABA for the treatment of
asthma.

Acknowledgements
The authors would like to acknowledge David Hinds for
editorial support in assembling this manuscript from an
earlier draft, Julie Priest for tabulating the adherence analysis,

 782

R. L. DiSantostefano

and Ted Murphy for tabulating the annual dispensing and
patient counts.

Declaration of interest
All authors are full-time employees and shareholders of
GlaxoSmithKline. The study was sponsored by
GlaxoSmithKline. All authors were active participants in
the study design, implementation, data analysis and preparation and editing of the manuscript. The data collection was
performed by the contracted companies as described in the
paper.

J Asthma Downloaded from informahealthcare.com by Jennifer Stahl on 05/12/15
For personal use only.

References
1. Nelson HS, Weiss ST, Bleecker ER, et al. SMART Study Group.
The salmeterol multicenter asthma research trial: a comparison of
usual pharmacotherapy for asthma or usual pharmacotherapy plus
salmeterol. Chest 2006;129:15–26. Erratum in: Chest
2006;129:1393.
2. Castle W, Fuller R, Hall J, Palmer J. Serevent nationwide
surveillance study: comparison of salmeterol with salbutamol in
asthmatic patients who require regular bronchodilator treatment.
BMJ 1993;306:1034–7.
3. National Institutes of Health. National asthma education and
prevention program (NAEPP). The NAEPP expert panel report 3
(EPR-3) summary report 2007: guidelines for the diagnosis and
management of asthma. US Department of Health and Human
Services; October 2007. NIH publication no. 08–5846.
4. US Food and Drug Administration (FDA). Drug safety communication: new safety requirements for long-acting inhaled asthma
medications called long-acting beta-agonists (LABAs); 18
February, 2010. Available from: http://www.fda.gov/Drugs/
DrugSafety/PostmarketDrugSafetyInformationforPatientsand
Providers/ucm200776.htm [last accessed 15 Feb 2012].
5. US Food and Drug Administration (FDA). Drug safety communication: drug labels now contain updated recommendations on the
appropriate use of long-acting inhaled asthma medications called
long-acting beta-agonists (LABAs); 2 June, 2010. Available
from:
http://www.fda.gov/Drugs/DrugSafety/PostmarketDrug
SafetyInformationforPatientsandProviders/ucm213836.htm
[last
accessed 15 Feb 2012].
6. US Food and Drug Administration (FDA). Approved risk evaluation and mitigation strategies (REMS). Available from: http://
www.fda.gov/Drugs/DrugSafety/PostmarketDrugSafetyInformation
forPatientsandProviders/ucm111350.htm [last accessed 13 Nov
2012].

J Asthma, 2013; 50(7): 776–782

7. McMahon AW, Levenson MS, McEvoy BW, et al. Age and risks of
FDA-approved long-acting b2-adrenergic receptor agonists.
Pediatrics 2011;128:e1147–54.
8. Sears MR. The FDA-mandated trial of safety of long-acting betaagonists in asthma: finality or futility? Thorax 2012;68:195–8.
9. US Food and Drug Administration (FDA). Risk evaluation and
mitigation strategy (REMS): NDA 21-077 ADVAIR DISKUSÕ
(fluticasone propionate and salmeterol xinafoate) inhalation
powder; December 2010. Available from: http://www.fda.gov/
downloads/Drugs/DrugSafety/PostmarketDrugSafetyInformationfor
PatientsandProviders/UCM241712.pdf [last accessed 15 Feb 2012].
10. US Food and Drug Administration (FDA). Risk evaluation and
mitigation strategy (REMS): NDA 20-692 SEREVENT DISKUSÕ
(salmeterol xinafoate) inhalation powder; December 2010.
Available from: http://www.fda.gov/downloads/Drugs/DrugSafety/
PostmarketDrugSafetyInformationforPatientsandProviders/
UCM241712.pdf [last accessed 15 Feb 2012].
11. Source Healthcare Analytics. GlaxoSmithKline custom asthma/
COPD hierarchical regimen analysis tool; January 2005–December
2011.
12. Source Healthcare Analytics. GlaxoSmithKline asthma/COPD
annual unique patient counts (UPCs). January 2005–December 2011.
13. National Prescription AuditÔ (NPA), IMS Health. January 2005–
December 2011.
14. National Disease and Therapeutic IndexÔ (NDTI). IMS Health,
January 2005–December 2011.
15. OptumÔ ImpactÔ National Managed Care Benchmark Database.
(OptumInsight, Eden Prairie, MN, USA) 2012.
16. Cramer JA, Roy A, Burrell A, et al. Medication compliance and
persistence: terminology and definitions. Value Health. 2008;11:
44–7.
17. DiSantostefano RL, Davis KJ. Prescription patterns in asthma
patients initiating salmeterol in UK general practice: a retrospective
cohort study using the General Practice Research Database
(GPRD). Drug Saf. 2011;34:511–20.
18. Wasilevich EA, Clark SJ, Cohn LM, Dombkowski KJ. Long-acting
b-agonist monotherapy among children and adults with asthma. Am
J Manag Care 2011;17:e91–5.
19. Kaplan S, Zhou EH, Iyasu SI. Characterization of long-acting
beta2-adrenergic agonists utilization in asthma patients. J Asthma
2012;49:1079–85.
20. Morales DR, Jackson C, Fielding S, Guthrie B. Long-acting
b-agonist prescribing in people with asthma in primary care.
Thorax 2013;68:192–4.
21. Stempel DA, Stoloff SW, Carranza Rosenzweig JR, et al.
Adherence to asthma controller medication regimens. Respir Med
2005;99:1263–7.
22. Ye X, Gutierrez B, Zarotsky V, et al. Appropriate use of inhaled
corticosteroid and long-acting beta (2)-adrenergic agonist combination therapy among asthma patients in a US commercially insured
population. Curr Med Res Opin 2009;25:2251–8.