Pulmonary Pharmacology & Therapeutics 24 (2011) 647e653

Contents lists available at SciVerse ScienceDirect

Pulmonary Pharmacology & Therapeutics

journal homepage: www.elsevier.com/locate/ypupt

Association between prescribing patterns of anti-asthmatic drugs and clinically

uncontrolled asthma: A cross-sectional study
Jesper Rmhild Davidsena, c, *, Jesper Hallasb, Jens Sndergaarda, Ren dePont Christensena, b,
Hans Christian Sierstedc, Malene Plejdrup Hansena, Thomas Bllingtoft Knudsena, Jesper Lykkegaarda,
Morten Andersena, d
a

Research Unit of General Practice, Institute of Public Health, University of Southern Denmark, J.B. Winslws Vej 9A, DK-5000 Odense C, Denmark
Research Unit of Clinical Pharmacology, Institute of Public Health, University of Southern Denmark, J.B. Winslws Vej 19, DK-5000 Odense C, Denmark
Department of Respiratory Medicine, Odense University Hospital, Sdr. Boulevard 29, DK-5000 Odense C, Denmark
d
Centre for Pharmacoepidemiology, Karolinska Institutet, Karolinska University Hospital, SE-171 76 Stockholm, Sweden
b
c

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 23 December 2010
Received in revised form
11 August 2011
Accepted 9 September 2011

Objective: Studies based on prescription data have shown that many asthmatics tend to use large
quantities of inhaled beta-2-agonists, suggesting poorly controlled disease. The aim of the present study
was to investigate the association between clinically uncontrolled asthma and prescribing patterns of
anti-asthmatic drugs with a primary focus on short-acting beta-2-agonists (SABA).
Methods: In a cross-sectional study 357 subjects, selected by their prescriptions of inhaled beta-2agonists in Odense Pharmaco-Epidemiological Database, underwent individual clinical assessment
including the Asthma Control Questionnaire (ACQ) and spirometry. The associations between uncontrolled asthma (ACQ score
1.50) and individual anti-asthmatic prescribing were analysed by means of
logistic regression.
Results: Clinically uncontrolled asthma was positively associated with SABA use, the association
becoming stronger with higher annual quantity of SABA use, odds ratio (OR) 11.1 (95% CI 4.4e28.0) for

400 DDD/year. This trend persisted after stratifying for gender, age, and controller treatment. Although
subjects using
450 DDD/year were all uncontrolled, there was substantial overlap in SABA use between
controlled and uncontrolled subjects below this limit. We found no effect modication by age and
gender. Use of inhaled corticosteroids protected against uncontrolled asthma, OR 0.51 (95% CI 0.27
e0.95).
Conclusion: Asthmatics with a high use of SABA frequently have problems with uncontrolled asthma, and
users of ICS are protected against uncontrolled asthma. The associations we found were, however, to
weak too allow rm conclusions about asthma control for most individual asthma patients.
2011 Elsevier Ltd. All rights reserved.

Keywords:
ACQ
Asthma control
Cross-sectional study
Pharmacoepidemiology
Treatment
Short acting beta-2-agonists

1. Introduction
Prescription data have been extensively used in quality indicators for different treatments [1,2]. In the case of asthma, studies
based on prescription data have shown that some asthmatics have
a massive consumption of inhaled beta-2-agonists (IBA) [3,4].
Among those with an IBA consumption corresponding to a high
daily use, 20e35% did not receive controller therapy with inhaled

* Corresponding author. University of Southern Denmark, Faculty of Health

Sciences, Institute of Public Health, Research Unit of General Practice, J.B. Winslws
Vej 9A, DK-5000 Odense C, Denmark. Tel.: 45 6550 3968; fax: 45 6550 3980.
E-mail address: jrd@dadlnet.dk (J.R. Davidsen).
1094-5539/$ e see front matter 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.pupt.2011.09.001

corticosteroids (ICS) [5,6]. These ndings have been regarded as

indicative of suboptimal asthma treatment. However, using
prescribing data as surrogate markers of suboptimal asthma
treatment in individual patients may be misleading and should be
interpreted with caution [7e9].
According to guidelines, the goal of successful asthma treatment
is to achieve and maintain clinical control of the disease [10,11].
Asthma can be classied into controlled, partly controlled, or
uncontrolled by a composite measure of individual disease
manifestations [10,11]. One of the key elements in assessing asthma
control is the use of reliever therapy, primarily short-acting beta-2agonists (SABA), as these can mask inammation in the asthmatic
airways, leading patients to ignore symptoms of impending deterioration and thus inducing poorly controlled asthma [12,13].

648

J.R. Davidsen et al. / Pulmonary Pharmacology & Therapeutics 24 (2011) 647e653

Even though prescribing data may be inappropriate to predict

the quality of asthma treatment on an individual level [8], a relation
between prescribed SABA and clinical asthma control may, very
likely, exist. Such knowledge would have a substantial impact for
identication of poorly controlled asthmatics on the basis of the
quantity of SABA prescribed. Hence, we designed a cross-sectional
study aiming to investigate associations between prescribing
patterns of anti-asthmatic drug use, in particular SABA, and clinically uncontrolled asthma. We used population-based prescription
data linked to clinical patient data.
2. Methods and materials
2.1. Setting and design
We performed a cross-sectional study on young adult asthmatics resident in the municipality of Odense, County of Funen,
Denmark, by use of two different data sources: Odense PharmacoEpidemiological Database (OPED), and data from a clinical assessment, including the Asthma Control Questionnaire (ACQ). The
study was conducted during 30 November 2008 and 7 July 2009.
2.2. Data sources
2.2.1. Odense Pharmaco-Epidemiological Database (OPED)
Information on reimbursed drug dispensing in the County of
Funen (population 470 000) has been recorded in OPED since 1990
and in the Region of Southern Denmark (population 1.2 million)
since the 1 January 2007 [14,15]. Each prescription record includes
a person identier, the date of dispensing, the brand, quantity, and
form of the drug. The substances and quantities are registered
according to WHOs anatomical-therapeutic-chemical (ATC) system
and dened daily doses (DDD) methodology [16]. The indication for
treatment and the dosing instruction are not recorded. Drugs not
reimbursed and therefore not recorded in the database are over-thecounter drugs and some non-reimbursed prescription drugs, mainly
oral contraceptives, hypnotics, sedatives, some antibiotics, but also
some intranasal drugs for rhinitis [17].
2.2.2. The Asthma Control Questionnaire (ACQ)
Asthma control is classied into three levels: Controlled,
partly controlled, or uncontrolled according to a composite
measure of different characteristics (day and night symptoms,
exacerbations, activity limitation, FEV1% predicted, and need of
reliever therapy) [10,11]. The ACQ is a validated seven-item questionnaire developed to measure asthma control during the past
week in clinical research studies as well as in clinical practice [18].
Items 1e5 cover questions on asthma symptoms (nighttime
waking, symptoms on waking, activity limitation, shortness of
breath, and wheezing), item 6 the use of short-acting beta-2agonists (SABA), and item 7 is a measure of FEV1% predicted.
Each answer is scored on a 7-point Likert scale from 0 to 6
(0 totally controlled, 6 severely uncontrolled) with the total
ACQ score being the average of the answers. An ACQ score 0.75
and
1.50 was dened as having controlled and uncontrolled
asthma, respectively [19]. An ACQ score of 0.75e1.50 has previously
been dened as having partly controlled asthma [12]. We used the
Danish version of ACQ, already translated by the MAPI Research
Institute, with permission from the copyright owner.
2.3. Study population
2.3.1. Identifying subjects
From OPED we retrieved data on all subjects who had received
drugs with ATC-code R03 (drugs for obstructive airway diseases)

between 1 December 2007 and 30 November 2008. We used the

following inclusion criteria: 1) age 18e44 years, 2) at least two
redeemed prescriptions on inhaled beta-2-agonists (IBA) 365 days
before 30 November 2008, 3) current IBA use dened as at least one
redeemed prescription on IBA 180 days before the index date, and
4) residency in the municipality of Odense, Denmark. Similar
criteria have been used in previous publications [6,20]. IBA referred
to inhaled short-acting beta-2-agonists (SABA) (ATC R03AC02,
R03AC03, R03AC04, and R03AC05), inhaled long-acting beta-2agonists (LABA) (ATC R03AC12 and R03AC13), xed combinations
of SABA with anticholinergics (ATC R03AK03 and R03AK04), or
LABA with ICS (ATC R03AK06 and R03AK07).
2.3.2. Recruitment and clinical assessment
Names and addresses of the identied subjects (N 1435) as
well as address of their general practitioners (GPs) were obtained
from the regional patient administrative system [15]. Prior to
inviting subjects for clinical assessment (during 28 January to 10
February 2009), their GPs were sent a list of subjects registered
with their practice in order to exclude subjects that they did not
nd appropriate for any further contact (e.g. language problems
and psychiatric diagnoses). Subjects not excluded by their GP
(N 1308) were invited during 13 February and 22 April 2009.
Non-responders were sent postal reminders 3e4 weeks after
receiving the invitation letter. Responders had to answer two
questions: Have you ever had asthma? and was asthma diagnosed by a doctor? Only responders answering yes to both
questions were subsequently contacted by telephone to schedule
time for an appointment (N 421). Finally, all responders who
attended examination were clinically assessed (N 365). Clinical
assessment was performed by two research nurses and the rst
author (JRD) at the Research Unit of Respiratory Medicine at Odense
University Hospital, Denmark, between 1 April and 7 July 2009.
Spirometry was performed on an IntraMedic, Micro Loop MK8
spirometer, in accordance with the European Respiratory Society
(ERS) and American Thoracic Society (ATS) standards [21].
The project was approved by the Regional Ethics Committee of
Southern Denmark (Project ID S-20080025), the Multi Practice
Committee under the Danish College of General Practitioners
(Multipraksisudvalget), and the Danish Data Protection Agency. All
participants gave informed consent.
2.4. Drug use variables
Updated information on subjects anti-asthmatic drug use was
obtained from OPED. The index date was dened as the date of
clinical assessment. Prevalent use of a particular drug was dened
as at least one redeemed prescription during the year preceding the
index date. SABA referred to inhaled short-acting beta-2-agonists
(SABA) (ATC R03AC02, R03AC03, R03AC04, and R03AC05), and
combinations of SABA with anticholinergics (ATC R03AK03 and
R03AK04). Subjects were categorised according to their annual
cumulative SABA use in DDD (0e99, 100e199, 200e399, and

400 DDD/year). The DDD is not a dose recommendation, but
a measure to aggregate drugs that have different potency, and
represents the assumed average maintenance dose per day for
a drug used on its main indication in adults [16]. Long-acting beta2-agonists (LABA) referred to LABA in non-combination inhalers
(ATC R03AC12 and R03AC13), and xed combinations with LABA
and ICS (ATC R03AK06 and R03AK07). ICS referred to ICS in noncombination inhalers (ATC R03BA), and xed combinations with
LABA and ICS (ATC R03AK06 and R03AK07). Thus, xed-dose
combinations (e.g. LABA and ICS, or SABA and anticholinergics)
were counted in two drug categories, corresponding to their individual constituents. Other drugs included were leukotriene

J.R. Davidsen et al. / Pulmonary Pharmacology & Therapeutics 24 (2011) 647e653

receptor antagonists (ATC R03DC), systemic glucocorticosteroids

(ATC H02AB), oral long-acting beta-2-agonists (ATC R03CC), oral
methylxanthines (ATC R03DA), inhaled long-acting anticholinergics (ATC R03BB04), inhaled short-acting anticholinergics (ATC
R03BB01 and R03BB02), oral chromones (ATC R03BC), intranasal
drugs for rhinitis (ATC R01A), and drugs for specic immune
therapy (ATC V01). We did not include omalizumab (ATC R03DX) as
this drug category has restricted delivery to hospitals.
2.5. Data analysis
We analysed associations between use of anti-asthmatic drugs
and the outcome, clinical uncontrolled asthma dened as an individual ACQ score
1.50 [19].

INITIAL COHORT
IDENTIFIED IN OPED

649

Our primary independent variable was the amount of SABA used

in the year preceding the date of clinical assessment, categorised in
the groups mentioned above. Other independent variables
considered were gender, age, and use of other anti-asthmatic drugs.
We stratied age into three categories: 18e24 years, 25e34 years,
and 35e44 years.
We rst estimated crude odds ratios (ORs) with 95% condence
intervals (CIs) by means of a logistic regression model. Secondly, to
examine whether a potential association was consistent, we performed similar analyses stratied on gender, age category, and
concomitant use of LABA, and ICS, respectively.
Lastly, to evaluate the consistency of our ndings we explored
the relation between mean scores of individual ACQ items and the
categories of annual SABA use.

EXCLUDED
BY GP
N = 127

N= 1435

STUDY
COHORT
N=1308

NO

LETTER OF ACCEPTANCE
OF PARTICIPATION a
N=421

N = 106
NON-RESPONSE
N = 781

NO

TELEPHONE ACCEPTANCE
OF PARTICIPATION
N=386

CLINICAL
ATTENDANCE
N=365

N=7
NO CONTACT
N = 28

NO
N = 21

INCLUSION c
NO

N=357

N=8

OMITTED
N = 951

Fig. 1. Patient ow. aLetter acceptance of participation and answering yes to both diagnostic questions on asthma. bEleven patients (10.4%) of the total 106 patients with nonacceptance of participation did not answer yes to both diagnostic questions on asthma. cSufciently performed ACQ completion including spirometry. dInsufciently performed ACQ completion including spirometry (N 6) and language difculties (N 2). Abbreviations: ACQ Asthma Control Questionnaire.

650

J.R. Davidsen et al. / Pulmonary Pharmacology & Therapeutics 24 (2011) 647e653

All analyses were performed using Stata Release 10.1 (StataCorp,

College Station, TX, USA). A p-value <0.05 was considered statistically signicant.
3. Results
3.1. Baseline data
Flow chart for recruitment and clinical assessment is presented
in Fig. 1. A total of 357 subjects (56.6% women) completed the
clinical assessment with ACQ and spirometry. There was
a substantial overlap in levels of SABA use among controlled and
uncontrolled subjects, and those using
450 DDD/year were all
uncontrolled (Fig. 2). Overall, 96 subjects (26.9%) were classied as
having uncontrolled asthma, 54.2% of whom were women. No use
was observed for inhaled short-acting anticholinergics, oral chromones, intranasal drugs for rhinitis, and drugs for specic immune
therapy, why these variables were excluded from further analyses.
Other baseline characteristics are presented in Table 1.
Using 1 June 2009 as index date, the distribution of annual SABA
use among non-participants (N 951) was 54.1% for 0e99 DDD/
year, 20.1% for 100e199 DDD/year, 14.3% for 200e399 DDD/year,
and 11.5% for
400 DDD/year. Thus, non-participants had an overall
higher annual SABA use than participants.
3.2. Associations with uncontrolled asthma
Table 2 shows determinants of uncontrolled asthma. We found
a clear association between increasing annual consumption of
SABA and clinically uncontrolled asthma reected by higher ORs for
subjects in high SABA use categories compared to subjects in the
lowest SABA use category. Men had similar odds of uncontrolled
asthma compared to women (OR 1.14 (95% CI 0.71e1.83), p 0.577).
Use of LABA, ICS, and leukotriene receptor antagonists was found to
protect against uncontrolled asthma, but was only statistically
signicant for ICS use (OR 0.51, 95% CI 0.27e0.95). Oral glucocorticosteroids, long-acting anticholinergics, and oral beta-2agonists, all showed a positive association with uncontrolled
asthma, but all statistically insignicant due to a low number of
users.
Table 3 shows ORs for the different SABA use categories associated with uncontrolled asthma stratied on gender, age category,

Table 1
Baseline characteristics.
Number (%)a
Total number of subjects
(mean age  SD, years)
Women
(mean age  SD, years)
Men
(mean age  SD, years)
Age
(years)
ACQ score

SABA use (DDD/year)

LABA
ICS
Leukotriene receptor
antagonists
Oral glucocorticosteroids
Long-acting anticholinergics
Oral beta-2-agonists
Methylxanthines

18e24
25e34
35e44
Well controlled
Partly controlled
Uncontrolled
0e99
100e199
200e399

400

357 (100.0)
(32.2  7.3)
202 (56.6)
(31.8  7.6)
155 (43.4)
(32.7  7.1)
70 (19.6)
138 (38.7)
149 (41.7)
146 (40.9)
115 (32.2)
96 (26.9)
245 (68.6)
52 (14.6)
34 (9.5)
26 (7.3)
256 (71.7)
309 (86.6)
34 (9.5)
45 (12.6)
8 (2.2)
4 (1.1)
1 (0.3)

Abbreviations: ACQ Asthma Control Questionnaire. SD standard deviation.

SABA short-acting beta-2-agonists. LABA long-acting beta-2-agonists.
ICS inhaled corticosteroids. DDD dened daily dose.
a
Number and percentage for the different anti-asthmatic drugs refer to prevalent
use (dened by
1 redeemed prescription) of the particular drug the year preceding
date of clinical assessment. There was no prevalent use of inhaled short-acting
anticholinergics, oral chromones, intranasal drugs for rhinitis, and drugs for
specic immune therapy.

LABA use, and ICS use. As observed in Table 2, a stronger association

between uncontrolled asthma and increasing annual consumption
of SABA was shown for all stratifying variables. However, some
strata were small, leading to statistically insignicant estimates.
Female gender and age category 25e34 years were the strata in
which the strongest associations for all SABA categories were
observed. Concomitant LABA and ICS use again tended to lower the
odds for being uncontrolled compared to no concomitant use of
LABA and ICS use.
The mean score of ACQ items 1e5 showed an overall increase
with increasing annual SABA use (Table 4). There was also a clear
positive correlation between the mean score of item 6, reecting
the patients account of recent SABA use, and prescription information on annual SABA use.
4. Discussion

Fig. 2. Distribution of ACQ scores according to individual annual use of SABA in DDD.
Area of circles proportional to the number of subjects for each data point.
ACQ Asthma Control Questionnaire. SABA short-acting beta-2-agonists.
DDD dened daily dose.

The main nding of our study was that clinically uncontrolled

asthma was positively associated with high SABA use. This was
found across different genders and ages, and irrespective of
controller treatment with LABA and ICS. Despite this consistent
nding, the association does not allow conclusions on an individual
level due to a substantial overlap in levels of SABA use among
controlled and uncontrolled subjects (Fig. 2).
In all analyses ICS use was found to protect against uncontrolled
asthma. This nding agrees with the fact that ICS forms the core of
controller treatment with indication for any level of persistent or
uncontrolled asthma [22]. This tendency also applied to LABA and
leukotriene receptor antagonists concordant with their role as
controllers.
Other studies have also found that a large need of reliever
therapy is associated with poor asthma control [12,23]. Their
methodological approach differed, however, from the one we used.

J.R. Davidsen et al. / Pulmonary Pharmacology & Therapeutics 24 (2011) 647e653

Table 2
Associations between uncontrolled asthma, age, gender, and anti-asthmatic drug
use.

Gender
Age
(Years)
SABA use
(DDD/year)

LABA
ICS
Leukotriene receptor
antagonists
Oral glucocorticosteroids
Long-acting
anticholinergs
Oral beta-2-agonists
Methylxanthines

Women
Men
18e24
25e34
35e44
0e99
100e199
200e399

400
No
Yes
No
Yes
No
Yes
No
Yes
No
Yes
No
Yes
No
Yes

Uncontrolled
asthma N (%)a

Crude OR
(95% CI)b

52
44
14
36
46
48
19
10
19
33
63
19
77
89
7
79
17
92
4
94
2
95
1

1.00
1.14 (0.71e1.83)
1.00
1.41 (0.70e2.84)
1.79 (0.90e3.53)
1.00
2.36 (1.24e4.51)
1.71 (0.77e3.81)
11.14 (4.43e28.02)
1.00
0.67 (0.41e1.11)
1.00
0.51 (0.27e0.95)
1.00
0.68 (0.29e1.62)
1.00
1.79 (0.93e3.45)
1.00
2.79 (0.68e11.40)
1.00
2.76 (0.38e19.83)

(25.7)
(28.4)
(20.0)
(26.1)
(30.9)
(19.6)
(36.5)
(29.4)
(73.1)
(32.7)
(24.6)
(39.6)
(24.9)
(27.6)
(20.6)
(25.3)
(37.8)
(26.4)
(50.0)
(26.6)
(50.0)
(26.9)
(100.0)

ec

Abbreviations: ACQ Asthma Control Questionnaire. IBA inhaled beta-2-agonists.

SABA short-acting beta-2-agonists. LABA long-acting beta-2-agonists.
ICS inhaled corticosteroids. DDD dened daily dose. CI condence interval.
a
Number and percentages of uncontrolled subjects in each stratum.
b
Female gender, lowest age category, lowest SABA use category, and non-use of
specic anti-asthmatic drug categories were considered as references.
c
Not applicable.

In the Inspire study the ACQ was also used to judge asthma control,
but the amount of SABA use was based on self reporting [12]. On the
contrary Lynd et al. used prescription records to estimate SABA
consumption and to dene controlled subjects by their number of
SABA canisters used in a year [23]. We used elements from both of
these studies, as information on anti-asthmatic drug use was
retrieved from prescriptions records, and the ACQ was the tool used
to assess clinical asthma control. To our knowledge, only one study
with a similar set-up has examined clinical asthma control in
relation to redeemed anti-asthmatic medication. In this study,
Laforest et al. recruited 1282 asthma patients among regular
customers of prescribed anti-asthmatic drugs at community pharmacies [24]. The level of asthma control was measured by the
Asthma Control Test Questionnaire to be completed at home, and
drug use determinants associated with uncontrolled asthma were
investigated by use of pharmacy records. Nearly 56% were uncontrolled, and the patients being prescribed reliever medication had
the highest proportion of inadequate asthma control. The fact that
the level of uncontrolled asthma was almost twice as high as in our
nding may be explained by a lower prevalent use of ICS (49.1%). In
that study, however, prevalent drug use was dened by at least one
prescription during the past six months.
There are good reasons to believe that the subjects included in
this study do have asthma. We used redeemed anti-asthmatic
prescriptions as a proxy for the asthma diagnosis to identify the
study population, a method which has previously been tested valid
[25,26]. Also, we restricted to subjects aged 18e44 years to avoid
inclusion of subjects with use of anti-asthmatics on indications
other than asthma (e.g. COPD). Among the subjects identied we,
furthermore, attempted to specify the diagnosis by only inviting
subjects answering yes to the following questions in the invitation letter: Have you ever had asthma? and was asthma

651

diagnosed by a doctor? In population-based studies the use of

questionnaire responses as the only diagnostic criterion has been
shown to be reliable with respect to identifying subjects with
asthma [27,28].
To measure the degree of clinical asthma control, we used the
seven-item ACQ which has been widely validated and has shown
strong evaluative and discriminative properties [18,19].
The rst 5 items of the ACQ concern the patients asthma
symptoms, item 6 the recent use of SABA, and item 7 the FEV1%
predicted. To evaluate how much items 6 and 7 of the ACQ
contributed to the denition of uncontrolled asthma, we recalculated the ACQ score based on only items 1e5. We found a statistically signicant but only trivial difference of 0.075 (95% CI
0.047e0.103) higher mean score for the ve-item ACQ compared to
the seven-item ACQ. Thus, the clinical symptoms have the largest
impact on the total ACQ mean score, in line with the ndings in
a previous study [29]. Furthermore, the consistent relation between
the mean scores of individual ACQ items 1-5 and the categories of
annual SABA use also corroborate our results (Table 4).
In order to examine associations we used a cross-sectional
design. However, this design does not allow studying causality,

Table 3
Associations between uncontrolled asthma and SABA use categories stratied on
gender, age category, LABA use, and ICS use.

Women

Men

Age 18e24
years

Age 25e34
years

Age 35e44
years

Concomitant
use of LABA

No concomitant
use of LABA

Concomitant
use of ICS

No concomitant
use of ICS

SABA use
(DDD/year)

Total Na

Uncontrolled
asthma
N (%)b

Crude OR
(95% CI)c

0e99
100e199
200e399

400
0e99
100e199
200e399

400
0e99
100e199
200e399

400
0e99
100e199
200e399

400
0e99
100e199
200e399

400
0e99
100e199
200e399

400
0e99
100e199
200e399

400
0e99
100e199
200e399

400
0e99
100e199
200e399

400

151
25
17
9
94
27
17
17
55
8
5
2
93
20
13
12
97
24
16
12
208
25
13
10
37
27
21
16
224
40
25
20
21
12
9
6

30
10
5
7
18
9
5
12
7
2
3
2
16
8
3
9
25
9
4
8
43
8
5
7
5
11
5
12
43
13
7
14
5
6
3
5

1.00
2.69 (1.10e6.58)
1.68 (0.55e5.14)
14.12 (2.79e71.44)
1.00
2.11 (0.82e5.46)
1.76 (0.55e5.63)
10.13 (3.17e32.42)
1.00
2.29 (0.38e13.64)
10.29 (1.45e72.81)
ed
1.00
3.21 (1.13e9.11)
1.44 (0.36e5.84)
14.44 (3.51e59.33)
1.00
1.73 (0.67e4.44)
0.96 (0.28e3.25)
5.76 (1.60e20.79)
1.00
1.81 (0.73e4.46)
2.40 (0.75e7.70)
8.95 (2.22e36.07)
1.00
4.40 (1.30e14.84)
2.00 (0.50e7.93)
19.20 (4.40e83.73)
1.00
2.03 (0.97e4.25)
1.64 (0.64e4.17)
9.82 (3.57e27.03)
1.00
3.20 (0.70e14.53)
1.40 (0.29e8.86)
16.00 (1.50e171.20)

(19.9)
(40.0)
(29.4)
(77.8)
(19.1)
(33.3)
(29.4)
(70.6)
(12.7)
(25.0)
(60.0)
(100.0)
(17.2)
(40.0)
(23.1)
(75.0)
(26.0)
(37.5)
(25.0)
(66.7)
(20.7)
(32.0)
(38.5)
(70.0)
(13.5)
(40.7)
(23.8)
(75.0)
(19.2)
(32.5)
(28.0)
(56.0)
(23.8)
(50.0)
(33.3)
(83.3)

Abbreviations: ACQ Asthma Control Questionnaire. IBA inhaled beta-2-agonists.

SABA short-acting beta-2-agonists. LABA long-acting beta-2-agonists.
ICS inhaled corticosteroids. DDD dened daily dose. CI condence interval.
a
Number of subjects in each stratum.
b
Number and percentages of uncontrolled subjects in each stratum.
c
Lowest SABA use category was considered as reference.
d
Not applicable.

652

J.R. Davidsen et al. / Pulmonary Pharmacology & Therapeutics 24 (2011) 647e653

Table 4
ACQ item mean score stratied by SABA use category.
ACQ items

1. Nighttime waking
2. Symptoms on waking
3. Activity limitation
4. Shortness of breath
5. Wheeze
6. Reliever SABA use
7. FEV1% of predicted

Item mean score (95% CI) by SABA use in DDD/year

0e99 (N 245)

100e199 (N 52)

200e399 (N 34)

400 (N 26)

0.52 (0.39e0.65)
1.11 (0.96e1.26)
1.21 (1.07e1.36)
1.62 (1.46e1.78)
0.84 (0.70e0.98)
0.53 (0.44e0.62)
0.90 (0.75e1.06)

0.77 (0.46e1.08)
1.29 (0.95e1.63)
1.60 (1.31e1.88)
1.83 (1.53e2.13)
1.15 (0.82e1.49)
1.04 (0.75e1.32)
1.66 (1.20e2.11)

0.94 (0.47e1.41)
1.38 (0.95e1.81)
1.26 (0.85e1.68)
1.79 (1.33e2.25)
1.06 (0.58e1.55)
1.61 (1.22e1.99)
0.97 (0.48e1.46)

1.38 (0.85e1.92)
2.04 (1.57e2.50)
2.08 (1.52e2.64)
2.46 (1.99e2.94)
1.85 (1.29e2.40)
2.42 (1.96e2.88)
0.96 (0.46e1.46)

Abbreviations: ACQ Asthma Control Questionnaire. SABA short-acting beta-2-agonists. DDD dened daily dose. CI condence interval.

e.g. whether the lack of control is due to over reliance of SABA or

due to inherent disease activity which would require more SABA.
Different types of bias due to the identication and recruitment
procedure should be considered. Healthy volunteer bias needs to be
considered, as the percentage of high annual SABA consumption
was higher among non-participants compared to participants. This
is consistent with previously reported ndings on survey data from
asthmatics [30]. Despite this possible bias, we found a large variation in both ACQ scores and SABA use among participants and were
able to demonstrate a consistent association between SABA use and
uncontrolled asthma.
Asthma control refers to the level of disease activity at a single
point in time and reects the effectiveness of current treatment
[31]. ACQ referred to asthma control during the week before clinical
assessment, while our assessment of anti-asthmatic drug use was
based on redeemed prescriptions during the past year. This may
have led to non-differential misclassication of asthma drug
exposure and we may thus have underestimated the associations
between uncontrolled asthma and use of anti-asthmatic drugs [32].
Indeed, appropriate asthma control can be very complex to
achieve due to interactions between many factors such as asthma
education, the doctor/patient relationship, the frequency of followup, weight, comorbidity, diet, exercise, and smoking. Furthermore,
asthma control can change rapidly due to exposure of triggers (e.g.
irritants, allergens) or treatment [31]. A major drawback of our
study was that the data sources did not account for any of these
potential factors inuencing asthma control. Moreover, as we used
redeemed prescriptions we were not able to distinguish between
physician behavior and patient non-adherence, i.e. not purchasing
a prescribed drug. Also, as our study population was limited to
subjects who had special prescription patterns of IBA use, the study
ndings may not be generalisable to the entire population of
asthma patients.
Our results have some clinical implications. Although we did
nd a consistent link between the level of SABA use and poor
asthma control, this association was not completely applicable to
the individual asthma patient. There was a substantial overlap in
levels of SABA use among controlled and uncontrolled subjects.
However, those using 450 DDD/year or more were all uncontrolled.
This corresponds to 5 inhalations daily with a 0.5 mg terbutaline
inhalator device [33], a level found only in small minority of
subjects. Thus, a surrogate marker of uncontrolled asthma based on
individual-based dispensed SABA could be invalid, unless the level
of SABA use is very high. This aspect has previously been emphasized in relation to suboptimal treatment [7e9], and demonstrates
that prescription data alone can rarely identify the individual
patient with suboptimal asthma control. Future studies should
investigate whether a combination of prescription and clinical
information (e.g. peak expiratory ow measurement, asthma diaries, and questionnaires on asthma symptoms) could be useful for
assessing and optimizing asthma control.

In conclusion, by using a combination of prescription data and

clinical information on asthma control, our results demonstrate
a clear association between clinically uncontrolled asthma and
increasing quantities of SABA use, and an inverse association between
uncontrolled asthma and ICS use. Hence, asthmatics with a high use of
SABA frequently have problems with uncontrolled asthma, and users
of ICS are protected against uncontrolled asthma. The associations we
found were, however, too weak to allow rm conclusions about
asthma control for most individual asthma patients.
Funding
This study was supported by a research fellowship to JRD from
the University of Southern Denmark, and by a research grant from
MSD Denmark. None of the funders played a role in the study
design, analysis, interpretation of data, writing of the article, or
decision to submit the article for publication.
Conicts of interest
All authors have completed the International Committee of
Medical Journal Editors Unied Uniform Disclosure Form for
Potential Conicts of Interest (available on request from the corresponding author) and declare that: JRD has received a fee for
organizing education and travel funding to the ERS congress in
2009 from AstraZeneca. JHs research institution has received
support from MSD for the submitted work, and JH has received fees
for teaching from AstraZeneca, Nycomed, Pzer and the Danish
Association of the Pharmaceutical Industry, and research grants
from Nycomed, Pzer, MSD, and Alkabello. JS has participated in
a study funded by a research grant from AstraZeneca. HCS has
received payment for research from Asthmatx Inc. MA has received
fees for teaching from the Danish Association of the Pharmaceutical
Industry, and has participated in research projects funded by
AstraZeneca, Nycomed, and Novartis.
Author contributions
JRD, MA, JH, JS and designed the study. JRD wrote the rst protocol
draft, validated data, performed the analysis, interpreted data and
wrote the rst article draft. JH, JS, and HCS provided input to the
protocol and article, helped with interpretation of data, and made
critical revision of the manuscript for important intellectual content.
RdC provided help to data validation, analysis and interpretation of
data. MPH, TBK, and JL provided input to the article and made critical
revision of the manuscript for important intellectual content. MA
provided input to the protocol and article, helped with data validation,
analysis and interpretation of data and made critical revision of the
manuscript for important intellectual content. All authors have
approved the nal version. JRD is the guarantor. All authors, external
and internal, had full access to all of the data (including statistical
reports and tables) in the study and can assume responsibility for the
integrity of the data and the accuracy of the data analysis.

J.R. Davidsen et al. / Pulmonary Pharmacology & Therapeutics 24 (2011) 647e653

Acknowledgements
The authors wish to thank Lise Keller Stark for proofreading the
manuscript, and Susanne Dssing Berntsen and Sidsel Bindzus
Hauge for managing data. A special thanks to Research Nurses
Bettina Dalsgaard and Maibritt Rvdal Christensen for their valuable assistance with patient assessment.
References
[1] Hoven JL, Haaijer-Ruskamp FM, Vander Stichele RH. Indicators of prescribing
quality in drug utilisation research: report of a European meeting (DURQUIM,
13e15 May 2004. Eur J Clin Pharmacol 2005;60:831e4.
[2] Martirosyan L, Arah OA, Haaijer-Ruskamp FM, Braspenning J, Denig P.
Methods to identify the target population: implications for prescribing quality
indicators. BMC Health Serv Res 2010;10:137.
[3] Hallas J, Hansen NC. Individual utilization of anti-asthma medication by
young adults: a prescription database analysis. J Intern Med 1993;234:
65e70.
[4] Gaist D, Hallas J, Hansen NC, Gram LF. Are young adults with asthma treated
sufciently with inhaled steroids? a population-based study of prescription
data from 1991 and 1994. Br J Clin Pharmacol 1996;41:285e9.
[5] Klomp H, Lawson JA, Cockcroft DW, Chan BT, Cascagnette P, Gander L, et al.
Examining asthma quality of care using a population-based approach. CMAJ
2008;8(178):1013e21.
[6] Davidsen JR, Sndergaard J, Hallas J, Siersted HC, Lykkegaard J, Andersen M.
Increased use of inhaled corticosteroids among young Danish adult asthmatics: an observational study. Respir Med 2010;104:1817e24.
[7] Himmel W, Hummers-Pradier E, Schumann H, Kochen MM. The predictive
value of asthma medications to identify individuals with asthmaea study in
German general practices. Br J Gen Pract 2001;51:879e83.
[8] Pont LG, Denig P, van der Molen T, van der Veen JV, Haaijer-Ruskamp FM.
Validity of performance indicators for assessing prescribing quality: the case
of asthma. Eur J Clin Pharmacol 2004;59:833e40.
[9] Andersen M. Is it possible to measure prescribing quality using only
prescription data? Basic Clin Pharmacol Toxicol 2006;98:314e9.
[10] Global Initiative for Asthma (GINA). Global strategy for asthma management
and prevention. GINA. Available from: http://www.ginasthma.org; 2009
[accessed 01.08.11].
[11] National Asthma Education and Prevention Programme. Expert panel report
3: guidelines for the diagnosis and management of asthma. Bethesda MD:
National Heart, Lung and Blood Institute. Full Report 2007. NIH Publication
No. 07e4051. Available from: http://www.nhlbi.nih.gov/guidelines/asthma/
asthgdln.pdf; August 2007 [accessed 01.08.11].
[12] Partridge MR, van der MT, Myrseth SE, Busse WW. Attitudes and actions of
asthma patients on regular maintenance therapy: the INSPIRE study. BMC
Pulm Med 2006;6:13.

653

[13] Kaplan A, Ryan D. The role of budesonide/formoterol for maintenance and

relief in the management of asthma. Pulm Pharmacol Ther 2010;23:88e96.
[14] Statistics Denmark. Available at: http://www.dst.dk [accessed 01.08.11].
[15] Hallas J, Dall M, Andries A, Andersen BS, Aalykke C, Hansen JM, et al. Use of
single and combined antithrombotic therapy and risk of serious upper
gastrointestinal bleeding: population based case-control study. BMJ 2006;
333:726.
[16] WHO Collaborating Centre for Drug Statistics Methodology. ATC index with
DDDs and guidelines for ATC classication and DDD assignment. Oslo:
Norwegian Institute of Public Health; 2006.
[17] Gaist D, Sorensen HT, Hallas J. The Danish prescription registries. Dan Med
Bull 1997;44:445e8.
[18] Juniper EF, OByrne PM, Guyatt GH, Ferrie PJ, King DR. Development and validation
of a questionnaire to measure asthma control. Eur Respir J 1999;14:902e7.
[19] Juniper EF, Bousquet J, Abetz L, Bateman ED. Identifying well-controlled and
not well-controlled asthma using the asthma control questionnaire. Respir
Med 2006;100:616e21.
[20] Davidsen JR, Sndergaard J, Hallas J, Siersted HC, Knudsen TB, Lykkegaard J,
et al. Impact of socioeconomic status on the use of inhaled corticosteroids in
young adult asthmatics. Respir Med 2011;105:683e90.
[21] Miller MR, Hankinson J, Brusasco V, Burgos F, Casaburi R, Coates A, et al.
Standardisation of spirometry. Eur Respir J 2005;26:319e38.
[22] Fanta CH. Asthma N Engl J Med 2009;5(360):1002e14.
[23] Lynd LD, Guh DP, Pare PD, Anis AH. Patterns of inhaled asthma medication
use: a 3-year longitudinal analysis of prescription claims data from British
Columbia, Canada. Chest 2002;122:1973e81.
[24] Laforest L, Van GE, Devouassoux G, Bousquet J, Chretin S, Bauguil G, et al.
Inuence of patients characteristics and disease management on asthma
control. J Allergy Clin Immunol 2006;117:1404e10.
[25] Pont LG, van der Werf GT, Denig P, Haaijer-Ruskamp FM. Identifying general
practice patients diagnosed with asthma and their exacerbation episodes
from prescribing data. Eur J Clin Pharmacol 2002;57:819e25.
[26] Moth G, Vedsted P, Schiotz P. Identication of asthmatic children using
prescription data and diagnosis. Eur J Clin Pharmacol 2007;63:605e11.
[27] Toren K, Brisman J, Jarvholm B. Asthma and asthma-like symptoms in adults
assessed by questionnaires. A literature review. Chest 1993;104:600e8.
[28] Thomsen SF, Ulrik CS, Kyvik KO, Larsen K, Skadhauge LR, Steffensen I, et al.
The incidence of asthma in young adults. Chest 2005;127:1928e34.
[29] Juniper EF, Svensson K, Mork AC, Stahl E. Measurement properties and
interpretation of three shortened versions of the asthma control questionnaire. Respir Med 2005;99:553e8.
[30] Ronmark E, Lundqvist A, Lundback B, Nystrom L. Non-responders to a postal
questionnaire on respiratory symptoms and diseases. Eur J Epidemiol 1999;
15:293e9.
[31] Fuhlbrigge AL. Asthma severity and asthma control: symptoms, pulmonary
function, and inammatory markers. Curr Opin Pulm Med 2004;10:1e6.
[32] Rothman KJ, Greenland S, Lash TL. Modern epidemiology. 3rd ed. Philadelphia: Lippincott Williams & Wilkins; 2008.
[33] Danish Medicines Agency. Available at: http://www.medstat.dk/ [accessed
01.08.11].