- Email: [email protected]
Inhaled corticosteroids vs. leukotriene-receptor antagonists and asthma exacerbations in children
Respiratory Medicine (2011) 105, 846e855 available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/rmed
Inhaled corticosteroids vs. leukotriene-receptor antagonists and asthma exacerbations in children `re a,b, Lucie Blais a,b,c,*, Fatima-Zohra Kettani b, Catherine Lemie Marie-France Beauchesne a,b,c, Sylvie Perreault a, Naoual Elftouh b, Francine M. Ducharme d a
Universite´ de Montre´al, Faculte´ de Pharmacie, C.P. 6128, Succursale Centre-ville, Montre´al, Que´bec, Canada H3C 3J7 Research Center, Hoˆpital du Sacre´-Cœur de Montre´al, Montre´al, Que´bec, Canada c Endowment Pharmaceutical Chair AstraZeneca in Respiratory Health, Canada d Research Center, CHU Sainte-Justine and Department of Pediatrics, University of Montreal, Montreal, Quebec, Canada b
Received 24 September 2010; accepted 8 December 2010 Available online 31 December 2010
KEYWORDS Asthma; Child; Leukotriene receptor antagonists; Inhaled corticosteroids; Adherence; Exacerbations
Summary Background: Compared to inhaled corticosteroids (ICS), better use of leukotriene-receptor antagonists (LTRA) may lead to a greater reduction in exacerbations among asthmatic children in real-life settings. Methods: To test this hypothesis, we used the Quebec administrative databases and identified a cohort of 27,355 asthmatic children aged 5e15 years in whom ICS or LTRA monotherapy was initiated in 1998e2005. The primary outcome was the rate of moderate-or-severe asthma exacerbations (emergency department visit or hospitalization for asthma or a dispensed prescription of oral corticosteroids) over the subsequent year. The adjusted rate ratios (RR) of asthma exacerbations were estimated with Poisson regression models. To minimize confounding by indication, all analyses were stratified by the presence or not of an asthma exacerbation in the year before treatment initiation. We also measured the proportion of days with supply prescribed and patient’s adherence with the Proportion of Prescribed Days Covered (PPDC). Results: The risk of exacerbations was significantly higher in the ICS than the LTRA group among children with no previous exacerbation (RR Z 2.3; 95%CI:1.3e4.0), but not in those with 1 exacerbations (RR Z 1.6; 0.8e3.1). The PPDC was similar between the groups (66%) but the proportion of days with supply prescribed was significantly higher in the LTRA than the ICS group (52% vs. 34%), resulting in higher use.
* Corresponding author. Universite ´ de Montre ´al, Faculte ´ de Pharmacie, C.P. 6128, Succursale Centre-ville, Montre ´al, Que ´bec, Canada H3C 3J7. Tel.: þ1 514 343 6111x3786; fax: þ1 514 343 6120. E-mail address: [email protected] (L. Blais). 0954-6111/$ - see front matter ª 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.rmed.2010.12.007
Inhaled corticosteroids vs. leukotriene-receptor antagonists
847
Conclusions: While confounding by indication cannot be firmly ruled out, ICS appears to be more frequently prescribed as an intermittent than a daily controller therapy resulting in less use, which may contribute to the apparent lower effectiveness compared to LTRA. ª 2010 Elsevier Ltd. All rights reserved.
Introduction Regular daily therapy with inhaled corticosteroids (ICS) or leukotriene-receptor antagonists (LTRA) is effective in reducing asthma exacerbations in both adults and children.1e4 According to current guidelines on asthma management and prevention, ICS remain, however, the preferred controller monotherapy for persistent asthma.5e7 This recommendation is further supported by a 2004 Cochrane Review of 27 trials including adults and children reporting a 65% higher risk of exacerbations requiring systemic steroids in patients treated with LTRA compared to ICS.8 Subsequent to this review, several recent randomized clinical trials also demonstrated the superiority of ICS over LTRA in children for a broad range of clinical endpoints, including asthma control.9e12 Whether the superiority of ICS over LTRA persists in real life practice remains to be proven. Indeed, the use of ICS is notoriously low, frequently hovering around 45% of the time when filled prescriptions are examined.13 Although the use of LTRA is also not optimal, it generally appears superior to that of ICS, being reported to be between 60 and 70%,13,14 perhaps because of its oral administration, which makes it easier and more convenient to take than inhaled medications, particularly in children.15,16 Few observational studies have compared the relative effectiveness of ICS and LTRA in children,17,18 yielding conflicting results. While one study concluded that users of LTRA were more likely to experience treatment failure and to be admitted to the hospital for asthma,17 the other reported no group difference in the rate of asthma-related resource use, hospitalizations, or rescue medication use.18 None of these studies compared the proportion of days with supply prescribed, patient’s adherence to the prescribed therapy and the usage of ICS and LTRA, which are important elements to consider when examining drug effectiveness. We undertook a population-based study to investigate the relative effectiveness of ICS and LTRA for preventing moderate and severe asthma exacerbations among asthmatic children aged 5e15 years. We also aimed to explore determinants of effectiveness, namely the proportion of days with supply prescribed, adherence to, and use of the therapy, and prior asthma severity. Our hypothesis was that better adherence to LTRA than ICS would lead to a greater reduction in the rate of moderate or severe asthma exacerbations among patients treated with LTRA.
Methods Source of data Data for this study were retrieved from two administrative databases from the Canadian province of Que ´bec: the Re ´gie de l’Assurance Maladie du Que ´bec (RAMQ) and the MED-
ECHO databases that are linked by a patient unique encrypted identifier. The RAMQ provides medical coverage to all Quebec residents and drug coverage to about 42% of the population,19 including the elderly (65 years), persons receiving social assistance, and workers and their families (called the adherents) who do not have access to a private drug insurance plan from their workplace. Children of parents insured by the RAMQ Drug Insurance Plan are automatically insured and prescribed medications are free of charge for them. The RAMQ database includes three types of files. The Beneficiary file gives information on patients’ characteristics, such as age, sex, area of residence, social aid status and periods of coverage for the medication insurance. The Medical Services file contains all the information relative to medical services received e.g. the type, date, and location of medical service (outpatient clinic, emergency department (ED), hospitalization), diagnostic codes, as well as the specialty of the treating physician. The Prescription Drugs file contains data on prescribed medications dispensed in community pharmacies and includes notably the name of the medication, dosage, amount of medication given to the patient, the dispensing date, the type of prescription (new or refill), the number of refills allowed with a new prescription, the prescription duration, and the specialty of the prescribing physician. The MED-ECHO database contains information on all acute care hospitalizations occurring in the province, e.g. the date of admission and length of stay as well as primary and secondary diagnoses. These databases are routinely used for epidemiological asthma research.20e24 Data recorded in the RAMQ Prescription Drugs file and the medical diagnosis for asthma recorded in the RAMQ Medical Services file have been formally evaluated and validated.25,26 This study has been approved by the Ethics Committee of the Hoˆpital du Sacre´-Coeur de Montre´al.
Study design From the RAMQ database, we reconstructed a cohort of asthmatic children that initiated a treatment in monotherapy with either ICS [beclomethasone, fluticasone or budesonide] or LTRA [montelukast or zafirlukast]. Patients were eligible if they fulfilled the following inclusion criteria: (1) first prescription of monotherapy with ICS or LTRA between January 1st 1998 and August 31st, 2005; (2) no prior asthma controller therapy, that is, no dispensed ICS and LTRA prescriptions in the year preceding to the treatment initiation; (3) at least one asthma diagnosis (ICD-9 493) recorded in the RAMQ database in the year prior to treatment initiation; (4) age between 5 and 15 years at treatment initiation; (5) covered by the RAMQ Drug Insurance Plan for at least one year prior to treatment initiation; and (6) a follow-up of at least 4 months after treatment initiation. Patients were excluded if they had at least one diagnosis of chronic bronchitis, cystic fibrosis, dyspnea, or
848 laryngotracheitis or a dispensed medication to treat these conditions in the year prior to treatment initiation. Patients were excluded if they had regular therapy with oral corticosteroids in the year prior to treatment initiation (182 supply days or more) or in the first 3 months following treatment initiation (45 supply days or more). Patients were followed for a minimum of 4 months and a maximum of 12 months after treatment initiation. A minimum of 4 months of follow-up was required in order to obtain a precise measure of adherence to ICS and LTRA. Follow-up was terminated before the 12th month (i.e. data were censored) if one of the following events occurred: a dispensed prescription of another controller therapy (i.e. a class switch or a class addition), end of drug coverage by the RAMQ Drug Insurance Plan, 16th birthday, or death. Data were obtained from the RAMQ between January 1st, 1997 and December 31st, 2005. The cohort selected from the RAMQ was then linked to the MED-ECHO database with a unique patient identifier to obtain data on acute care hospitalizations that occurred during the same period.
Primary and secondary outcomes The primary outcome was the rate of moderate or severe asthma exacerbations defined as an emergency department (ED) visit for asthma, a hospital admission for asthma, or a dispensed short-course (14 days) prescription of oral corticosteroids. All the aforementioned events occurring during a 15-day interval were considered as a single exacerbation. The secondary outcome was a dispensed shortcourse prescription of oral corticosteroids, a marker of a moderate or severe asthma exacerbation27 more specific than an acute care visit where as many as 50% could be made for mild exacerbation.28
Compared therapy Children with asthma initiating a treatment with ICS as monotherapy were compared to those initiating a monotherapy with LTRA.
Confounding and stratification variables Potential confounders measured in the year before treatment initiation included average number of doses of inhaled short-acting beta2-agonists (SABA) per week (3, >3e10, >10), oral SABA (0, 1 filled prescription) and inhaled long-acting beta2-agonists (LABA) (0, 1 filled prescription). Potential confounders measured at treatment initiation were age (5e11, 12e15 years), gender, child living with parents receiving social assistance (yes/no), area of residence (urban/rural), specialty of the prescribing physician (family physician, paediatrician, respiratory specialist, others), and initial daily dose prescribed. To estimate the daily dose of ICS prescribed (expressed in beclomethasone-equivalent doses), we developed an algorithm based on the name of the medications and equivalencies between the different ICS products as recognized by the Canadian Asthma Consensus guidelines,29 the content of active ingredient per dosage unit and the quantity of
L. Blais et al. drug prescribed, the date and duration of the prescription, and the rate of renewals of the prescription. A second algorithm was developed in order to obtain an estimate of the average number of inhaled SABA filled per week, based on the name of the medications and equivalencies between the different SABA agents, as established by a pharmacist (M-F.B), date of filling, prescription renewals, and length of time between renewals.30 The average number of doses per week was divided into three categories (3, >3e10, >10 doses per week), according to the Canadian Asthma Consensus guidelines.29 In order to minimize potential confounding by indication due to the severity of asthma, we performed all analyses stratified on the occurrence or not of a moderate or severe asthma exacerbation in the year before treatment initiation.
Prescription pattern, treatment adherence and usage of medications Prescription pattern was assessed by the proportion of days with supply prescribed. This measure is obtained by dividing the total days’ supply prescribed (i.e. the sum of the duration of all prescriptionsdnew and allowed refills from all physicians consulted during the follow-up period) over the number of days of follow-up. Of note, the duration of allowed refills were summed whether or not the patient went to the pharmacy to get the refills. Treatment adherence was measured by the “Proportion of prescribed Days Covered” (PPDC), a new measure of patient’s adherence that has been recently developed by our group.31 The PPDC is defined as the total days’ supply dispensed during the follow-up over the total days’ supply prescribed.31 Finally, the usage of medications was summarized as the “number of prescription dispensed” and “the proportion of days covered” (PDC). The PDC is defined as the total number of days with supply dispensed during the follow-up over the number of days of follow-up.32 The PDC reflects the behaviour of both the patient and the prescribing physician and consequently the usage of medications, while the PPDC tends to measure more specifically the behaviour of the patient as it measures patient adherence to the therapy that was actually prescribed, and not to the therapy that should have been prescribed according to clinical guidelines, i.e. chronic daily use in the case of ICS and LTRA.
Statistical analysis All analyses, including the description of patient characteristics, were stratified by the presence or not of a moderate or severe asthma exacerbation (as defined in the primary outcome) in the year before treatment initiation with ICS or LTRA. We calculated the crude rate (per person per year) of moderate or severe asthma exacerbations and use of short-course of oral corticosteroids during follow-up, separately for patients treated with ICS and LTRA. The crude and the adjusted rate ratios (RR) of the outcomes, comparing patients treated with ICS to LTRA, were calculated using Poisson regression models with
Inhaled corticosteroids vs. leukotriene-receptor antagonists a dispersion parameter and reported with 95% confidence intervals (CIs). All potential confounding variables were entered in the models.33 We also calculated the mean (standard deviation) proportion of days with supply prescribed, PPDC, PDC and number of filled prescriptions of ICS and LTRA and compared these measures between ICS and LTRA patients using t-tests. All reported P-values are two-sided, with a significance level of 0.05. All analyses were performed using SAS software, version 9.1 (SAS Institute Inc., Cary, NC, USA).
Results Patients’ characteristics The cohort consisted of 27,355 children, including 7494 (27%) children with at least one exacerbation and 19,861 (73%) without any exacerbation in the year prior to treatment initiation. A flow chart with the number of patients excluded and the number of patients in which follow-up was censored is available in the online supplement. For note, among children who had suffered at least one exacerbation in the year prior treatment initiation, 74% experienced it in the 15 days preceding treatment initiation. Children’s characteristics are depicted in Table 1. Within strata defined by the presence or not of previous exacerbation, the two groups were comparable for sociodemographic characteristics, but we observed differences in variables related to asthma. Among children who had at least one exacerbation in the year preceding treatment initiation, a greater proportion of children treated with LTRA than ICS were prescribed their initial controller medication by a paediatrician, had used high doses of SABA, filled 2 prescriptions of oral corticosteroids, had made outpatient (non ED) visits for asthma, and had 2 moderate or severe exacerbations in the year prior to the initiation of the therapy. However, children treated with ICS were more likely to have had an ED visit and a hospitalization for asthma in the year prior to the initiation of the therapy. Similar trends were observed among children who had no exacerbation prior to treatment initiation. Among children who had at least one exacerbation in the year prior to treatment initiation, the proportion of children who received their initial prescription during an ED visit (prescription filled at most 15 days after the ED visit) was significantly higher in the ICS group than in the LTRA group (55% vs. 13%; p < 0.0001). The median starting daily dose of ICS prescribed was high (about 1000 mg/day), whether children had had an asthma exacerbation or not in the year before treatment initiation.
Crude rate of asthma exacerbations In the strata with no exacerbation in the preceding year, the crude rate of moderate to severe exacerbations, rescue oral corticosteroids, and ED visits was significantly higher among those treated with ICS than among those treated with LTRA; no group difference was found in the other strata (Table 2).
849
Adjusted rates of asthma exacerbations The adjusted rate ratios confirmed these above findings. In the strata with no previous exacerbation, the risk of exacerbations was two-fold higher in the ICS compared to the LTRA groups (adjusted RR Z 2.3; 95% CI; 1.3e4.0) (Table 3). A similar trend favouring LTRA but not reaching statistical significance was observed in the other strata (adjusted RR Z 1.6; 95% CI: 0.8e3.1). Comparable rate ratios favouring LTRA were seen regarding the use of oral corticosteroids in both strata, again statistically significant RR Z 2.3; 1.2e4.6 only among children with no previous exacerbation (RR Z 2.0; 0.8e5.3 among children with previous exacerbation). In both strata, an increased risk of asthma exacerbations with ICS compared to LTRA was associated with prescriptions by paediatricians at treatment initiation and with an increased number of doses of inhaled SABA per week in the year prior to treatment initiation. Oral SABA in the preceding year was also associated with an increased risk of exacerbations in the ICS group among children with a history of asthma exacerbations.
Treatment adherence to ICS and LTRA in the year following treatment initiation There was no group difference in patients’ adherence measured by the PPDC, which hovered around two-thirds for both groups (Table 4). Yet, the mean PDC, that is the proportion of days with a controller medication, was twofold higher in the LTRA group than the ICS group (p < 0.001), in both strata. Similarly, the proportion of days with supply prescribed and the number of prescriptions filled were significantly higher among children treated with LTRA than ICS.
Discussion In this observational study among Canadian children with asthma, initiation of monotherapy with ICS was associated with a two-fold higher rate of exacerbations compared to LTRA, but only among children with no exacerbation requiring a health care utilisation in the year before treatment initiation; there was no significant group difference among those with exacerbations in the preceding year. In both strata, patients’ adherence rates were similar between ICS and LTRA groups, hovering around two thirds of the prescribed doses. However, there was two-fold higher daily usage of LTRA than ICS, resulting from a combination of a higher proportion of days with a supply prescribed by the physician and a higher number of prescription dispensed per year in the LTRA than the ICS groups. The observed relative risk of exacerbations in our study contradicts those of previous randomized clinical trials demonstrating better outcomes with ICS compared with LTRA.912 The present investigation was based on data from actual practice, resulting from a combination of the physician prescription patterns and duration, and the patient adherence to filling the prescribed medications. While the results from clinical trials have high levels of
850
L. Blais et al.
Table 1
Children’s Characteristics. 1 exacerbation in the year prior to treatment initiation
No exacerbation in the year prior to treatment initiation
ICSa
LTRAb
ICSa
LTRAb
Number of patients Mean s.d. follow-up time (days)
7427 351 45
67 297 86
19439 350 46
422 321 74
At treatment initiation Age (years); n (%) 5e<12 12e15 Male gender, n (%) Receipt of social assistance; n (%)
5888 1539 4370 2772
49 18 34 28
(73.1) (26.9) (50.8) (41.8)
15017 (77.3) 4422 (22.7) 10642 (54.8) 7266 (37.4)
317 105 232 124
Area of residence; n (%)c Urban Rural Missing
5521 (74.3) 1905 (25.7) 1
55 (82.1) 12 (17.9) 0
15402 (79.3) 4033 (20.7) 4
327 (77.5) 95 (22.5) 0
Specialty of prescribing physician; n (%) Family physician Pediatrician Respiratory physician Other specialties Median daily dose prescribed
5130 (69.0) 2062 (27.8) 44 (0.6) 191 (2.6) 1071 mg
33 (49.2) 29 (43.3) 0 5 (7.5) 5 mg
13339 (68.6) 5428 (27.9) 265 (1.4) 407 (2.1) 1000 mg
202 (47.9) 181 (42.9) 12 (2.8) 27 (6.4) 5 mg
In the year preceding treatment initiation Average number of doses of inhaled SABAd per week; n (%) 3 6629 (89.3) >3e10 677 (9.1) >10 121 (1.6) Oral SABAd (1 filled prescription); n (%) 418 (5.6) 5 (0.1) LABAe (1 filled prescription); n (%) Theophylline (1 filled prescription); n (%) 8 (0.1)
55 (82.1) 10 (14.9) 2 (3.0) 5 (7.5) 1 (1.5) 0
18033 (92.8) 1189 (6.1) 217 (1.1) 895 (4.6) 26 (0.1) 35 (0.2)
379 (89.8) 37 (8.8) 6 (1.4) 29 (6.9) 4 (1.0) 1 (0.2)
Short-course filled prescription of oral corticosteroids; n (%) 0 3351 (45.1) 1 3601 (48.5) 2 475 (6.4) Outpatient medical visit for asthma (1); n (%) 2996 (40.3)
27 30 10 53
19439 (100.0)
422 (100.0)
19439 (100.0)
422(100.0)
EDf visit for asthma (1); n (%) Hospitalization for asthma (1); n (%)
5574 (75.1) 657 (8.9)
34(50.8) 4(6.0)
0 0
0 0
Asthma exacerbations; n (%) 0 1 2
0 6614 (89.0) 813 (11.0)
0 54 (80.6) 13 (19.4)
19439 (100.0) 0 0
422 (100.0) 0 0
a b c d e f
(79.3) (20.7) (58.8) (37.3)
(40.3) (44.8) (14.9) (79.1)
(75.1) (24.6) (55.0) (29.4)
ICS: Inhaled corticosteroids. LTRA: Leukotriene receptor antagonists. Measured in the 15 days preceding the index date. SABA: Short-acting beta-agonists. Long-acting beta-agonists. ED: Emergency department.
internal validity, their generalizability to the clinical practice setting is often questionable. Indeed, efficacy clinical trials are performed under ideal settings, with carefully selected patients, strict treatment protocols, and close
monitoring by the research team; poor adherence to drug or follow-up visit is a common exclusion criteria. Consequently, efficacy clinical trials minimize variability due to physician prescription patterns and patient adherence to
Inhaled corticosteroids vs. leukotriene-receptor antagonists Table 2
851
Crude Rate of Exacerbations for Children Treated with ICS or LTRA During Follow-up. 1 exacerbation in the year prior to treatment initiation
P-value No exacerbation in the year prior to treatment initiation
P-value
ICSa
LTRAb
ICSa
LTRAb
Number of patients
7427
67
19439
422
Exacerbations 1 exacerbation; n (%) Number of exacerbations Rate of exacerbation/patient/year
1671 (22.5) 2252 0.32
10 (14.9) 12 0.22
0.128
1535 (7.7) 1769 0.09
14 (3.3) 16 0.04
0.001
Oral corticosteroids 1 filled prescription; n (%) Number of filled prescriptions Rate of oral corticosteroids/patient/year
1222 (16.5) 1691 0.24
5 (7.4) 7 0.13
0.068
1222 (6.3) 1464 0.08
10 (2.4) 13 0.04
0.003
EDc visit for asthma 1 EDc visit for asthma; n (%) Number EDc visit for asthma Rate of EDc visit for asthma/patient/year
1158 (15.6) 1664 0.23
7 (10.5) 12 0.22
0.389
708 (3.6) 844 0.05
7 (1.7) 8 0.02
0.037
3 (4.5) 3 0.06
0.175
85 (0.4) 88 0.005
1 (0.2) 1 0.003
0.340
Hospitalizations for asthma 1 hospitalizations for asthma; n (%) 167 (2.2) Number of hospitalizations for asthma 191 Rate of hospitalizations for asthma/patient/year 0.03 a b c
ICS: Inhaled corticosteroids. LTRA: Leukotriene-receptor antagonists. ED: Emergency department.
the tested drug regimens. The possible role of prescribing patterns and patient adherence on the health outcomes is thus best explored using data from a real-world setting. Contrary to our initial hypothesis, patient adherence, measured by the proportion of prescribed days covered, was comparable between groups hovering around 66%. Of interest, the proportion of days with supply prescribed, reflecting prescription patterns and duration, is nearly twice higher for LTRA compared to ICS, suggesting that physician prescribe enough LTRA supply to last twice as long as the ICS supply. The unusually high dose of ICS hovering around 1000 mg/day and the higher rate of treatment initiation during an exacerbation, suggest that ICS is more frequently prescribed as a curative intermittent treatment during an exacerbation than as a daily controller medication, a pattern not found with LTRA. These findings would suggest that a large proportion of the non-adherence seen with the commonly employed “proportion of days covered (PDC)” measure may not be due to patients’ nonadherence to the prescribed medication, but rather to the pattern of ICS prescription. This is in line with the findings of recent US surveys which showed that only half of the primary care physicians reported concordance with the guidelines recommendations for the prescription of ICS therapy for children with persistent asthma.34,35 The resulting higher daily use of LTRA than ICS medications could therefore explain, at least in part, the greater reduction in exacerbations among children with asthma treated by LTRA compared to ICS.
The two-fold increased rate of asthma exacerbation associated with ICS observed in this study also differ from that reported in two similar other observational studies, which themselves yielded inconsistent results.17,18 Stempel and colleagues reported that children 4e17 years of age treated with montelukast were 2.5 times more likely to experience treatment failure (defined as asthma-related hospitalizations, ED visits, oral corticosteroids use, or switch/augmentation of index therapy) than those treated with fluticasone (OR Z 2.55; 95% CI: 2.19e2.96).17 This study did not include incident users of ICS or LTRA as we did, but specifically required active ICS therapy in the year in both groups before the index prescription; perhaps this criteria led to the selection of patients that were particularly responsive to their respective controlled medication. Conversely, Allen-Ramey et al. showed that among children 2e14 years of age who were new users of ICS or LTRA, there was no statistical difference between the montelukast and fluticasone cohorts in the risk of rescue oral corticosteroids (RR Z 1.02; 0.88e1.19). This result persisted in the subgroup of children aged 6e14 years, more comparable to our age group.18 No significant difference between the two groups in the risk of asthma-related ED visits or hospitalizations were neither noted in the 6e14 years age group. Those findings differ from ours, maybe because of differences in the baseline severity of patients between the two studies. Indeed, in the Allen-Ramey study, there was a significantly higher pre-index rate of oral corticosteroid use among those who initiated fluticasone than among
852
L. Blais et al.
Table 3 Crude and Adjusted Rate Ratios of Moderate or severe Asthma Exacerbations Comparing Children Treated with ICS and LTRA During Follow-up. 1 exacerbation in the year prior to treatment initiation
No exacerbation in the year prior to treatment initiation
Crude RRa
Adjusted RRa
95% CI
Crude RRa
Adjusted RRa
95% CI
1.4
1.6
0.8e3.1
2.2
2.3
1.3e4.0
At treatment initiation Age (<12 years vs.12e15 years) Male vs. female Area of residence (rural vs. urban) Social assistance (yes vs. no)
1.2 1.0 0.9 1.1
1.2 1.0 0.9 1.1
1.0e1.4 0.9e1.1 0.8e1.1 1.0e1.2
1.1 1.0 1.0 1.1
1.1 1.0 1.1 1.1
1.0e1.2 0.9e1.1 0.9e1.2 1.0e1.2
Specialty of prescribing physician Family physician Pediatrician Respiratory physician Other specialties
Reference 1.3 0.9 1.5
1.2 0.8 1.4
1.1e1.4 0.4e1.7 1.1e1.8
Reference 1.2 0.5 0.9
1.2 0.5 1.0
1.1e1.3 0.3e1.0 0.7e1.4
In the year prior to treatment initiation Number of doses of SABAd/week 0e3 Reference >3-10 1.8 >10 2.6 1.4 Oral SABAd (yes vs. no) LABAe (yes vs. no) 0.6
1.8 2.7 1.3 0.5
1.5e2.0 2.1e3.5 1.1e1.6 0.1e5.6
Reference 1.7 2.4 1.1 0.4
1.7 2.6 1.1 0.4
1.4e2.1 1.8e3.6 0.8e1.4 0.0e3.8
b
ICS vs. LTRA
c
a RR: Rate ratio; adjusted for age, sex, area of residence, social assistance, specialty of prescribing physician, average number of doses of inhaled short-acting beta2-agonists (SABA) per week, oral SABA and inhaled long-acting beta2-agonists (LABA). b ICS: Inhaled corticosteroids. c LTRA: Leukotriene-receptor antagonists. d SABA: Short-acting beta-agonists. e LABA: Long-acting beta-agonists.
those who initiated montelukast, contrarily to our study. Of interest, none of these studies compared the proportion of days covered by the prescribed therapy, patient’s adherence to the prescribed therapy and the usage of ICS and LTRA, which are important elements to consider when examining drug effectiveness. Our study has several strengths. First, it is a populationbased effectiveness study, representative of real-world clinical practice. Second, our data were obtained from a validated healthcare database providing detailed information on prescribed medications recorded prospectively, thus excluding the possibility of recall bias. Third, we used established and objective markers of moderate and severe asthma exacerbations.36e38 Finally, the large sample size provided precise estimates of association. We acknowledge the following limitations. First and foremost, as for any non-randomized population-based study, it is not possible to control treatment assignment which could potentially lead to systematic differences between the two study groups, a phenomenon called confounding by indication. In other words, ICS may have been used for children with more severe asthma than those treated with LTRA. We attempted to minimize such bias by stratifying all the analyses for the occurrence or not of
a moderate or severe asthma exacerbation in the year before treatment initiation. Although the stratification led to comparable treatment groups for most variables in the strata with no prior exacerbation, we cannot rule out imbalances in other unmeasured variables known to be associated with asthma morbidity, such as ethnicity or environmental factors (aeroallergens).39,40 In the strata with prior exacerbations, several group imbalances remained that could have masked important group differences in health outcomes. Other clinical markers of disease severity at the baseline such as lung function or asthma symptoms were not provided by administrative databases and thus could not be used for a more precise assessment of asthma severity. The data retrieved from the Medical and Prescription databases were validated in adults and not in children. Adherence to study medications was based on dispensed prescriptions and not the actual use of medications; while one cannot prove intake from dispensed medication, intake cannot take place without first dispensing. The use of administrative databases does not allow determining whether patients have filled their new prescription, which may have resulted in an underestimation of the number of new prescriptions and allowed refills as only prescriptions that reach the pharmacy are
Inhaled corticosteroids vs. leukotriene-receptor antagonists Table 4
853
Treatment Adherence to ICS and LTRA in the Year Following Treatment Initiation. 1 exacerbation prior to treatment initiation
LTRAb
67
19439
422
Proportion of days with supply prescribed Mean percentage (s d) 32.8 (30.1)
55.9 (35.9)c
34.0 (30.9)
51.9 (35.1)c
Proportion of Prescribed Days Covered (PPDC ) Mean percentage (s d) 68.0 (33.0)
64.7 (29.3)d
63.9 (34.4)
62.7 (33.5)d
Proportion of Days Covered (PDC ) Mean percentage (s d) 15.1 (12.2)
33.3 (28.3)c
14.2 (11.6)
29.4 (28.7)c
Number of prescriptions dispensed per year Mean number (s d) 4.4 (5.7)
7.9 (11.3)c
4.4 (5.2)
6.8 (8.1)c
Number of patients
7427
LTRA
b
No exacerbation prior to treatment initiation ICSa
ICS
a
DEFINITIONS:Proportion of days with supply prescribed Z Total days’ supply prescribed (i.e. the sum of the duration of all new prescriptions plus allowed refills over the follow-up period)/Duration of follow-up: reflects prescription patterns.Proportion of Prescribed Days Covered (PPDC) Z Total days’ supply dispensed over the follow-up period/Total days’ supply prescribed: reflects patients’ adherence to prescribed therapy.Proportion of Days Covered (PDC) Z Total days’ supply dispensed during follow-up period/ Duration of Follow-up: reflects usage of medications (both patient and physician behaviour). a ICS: Inhaled corticosteroids. b LTRA: Leukotriene receptor antagonists. c All p-values < 0.001 for the ICS/LTRA comparison with a t-test (PDSP, PDC and number of prescriptions per year). d P-value Z 0.184 for the ICS/LTRA comparison with a t-test (PPDC).
accounted for in the database. Our new measure of patient adherence was based on the total days’ supply prescribed, i.e. the sum of the duration of all prescriptions e new and allowed refills from all physicians consulted during the follow-up period. Thus, it doesn’t take into account the variation in the doses that may have occurred between two refills based on action plans for example. So, it is possible that adherence would be underestimated particularly in the case of ICS. Finally, the absence of a significant result in the group of patients with a previous exacerbation might be due to reduced power driven by the small number of patients treated with LTRA. With regards to generalizability, our population is mostly representative of children whose parents are from the lower to middle class population of Quebec. However, the free healthcare and access to medication in children of Quebec may make the impact of social class less important than in other health care systems. In conclusion, this effectiveness study of real life drug utilisation patterns in the Canadian setting revealed that the risk of asthma exacerbations was two-fold higher in children prescribed ICS compared to LTRA in children without exacerbations requiring health care resources in the year prior to treatment initiation; with no difference in children with prior exacerbations. Higher proportion of days with a supply prescribed and of days of use of LTRA compared to ICS in light of similar patient’s adherence, suggest that the prescribing of ICS as an intermittent rather than a daily controller explain in part our findings. Yet, confounding by indication cannot be firmly ruled out. Our study demonstrates the importance of considering the prescribing patterns while measuring patients’ adherence with prescription refills recorded in administrative databases.
Ethics statement This study was approved by the Scientific and Ethics Committee of the Ho ˆpital du Sacre ´-Coeur de Montre ´al, Quebec, Canada.
Funding Lucie Blais, Catherine Lemie `re, and Sylvie Perreault are the recipients of a Salary Award from the Fonds de la recherche en sante´ du Que´bec (FRSQ). This study was funded through a grant received from the Fonds de la recherche en sante´ du Que´bec (FRSQ) and the Conseil du me´dicament du Que´bec.
Conflict of interest statement Lucie Blais received research grants from Sanofi-Aventis, AstraZeneca and GlaxoSmithKline and co-chairs the AstraZeneca Endowment Pharmaceutical Chair in Respiratory Health. Marie-France Beauchesne has received grant support from AstraZeneca Canada Inc and GlaxoSmithKline Canada Inc for the conduct of her research projects. She has also received honoraria for education programs given to pharmacists from GlaxoSmithKline Canada Inc, Boehringer Ingelheim Canada Inc., AstraZeneca Inc, and Pfizer Canada Inc. She co-chairs the AstraZeneca Endowment Pharmaceutical Chair in Respiratory Health. Francine Ducharme received research grants from GlaxoSmithKline and Merck. She has also received travel support for meeting attendance,
854 research funds, fees for speaking and/or consulting fees from AstraZeneca, Merck Frosst Inc, GlaxoSmithKline and Novartis. Naoual Elftouh, Fatima-Zohra Kettani, Catherine Lemie `re and Sylvie Perreault have no competing interests to declare.
Supplementary material The supplementary data associated with this article can be found in online version at doi:10.1016/j.rmed. 2010.12.007.
References 1. Suissa S, Ernst P, Kezouh A. Regular use of inhaled corticosteroids and the long term prevention of hospitalisation for asthma. Thorax 2002 Oct;57(10):880e4. 2. Barnes NC, Miller CJ. Effect of leukotriene receptor antagonist therapy on the risk of asthma exacerbations in patients with mild to moderate asthma: an integrated analysis of zafirlukast trials. Thorax 2000 Jun;55(6):478e83. 3. The Childhood Asthma Management Program Research Group. Long-term effects of budesonide or nedocromil in children with asthma. N Engl J Med 2000 Oct 12;343(15):1054e63. 4. Garcia Garcia ML, Wahn U, Gilles L, Swern A, Tozzi CA, Polos P. Montelukast, compared with fluticasone, for control of asthma among 6- to 14-year-old patients with mild asthma: the MOSAIC study. Pediatrics 2005 Aug;116(2):360e9. 5. Becker A, Berube D, Chad Z, Dolovich M, Ducharme F, D’Urzo T, et al. Canadian pediatric asthma consensus guidelines, 2003 (updated To December 2004): introduction. CMAJ 2005 Sep 13; 173(6 Suppl):S12e4. 6. Bateman ED, Hurd SS, Barnes PJ, Bousquet J, Drazen JM, FitzGerald M, et al. Global strategy for asthma management and prevention: GINA executive summary. Eur Respir J 2008 Jan;31(1):143e78. 7. National Asthma Education and Prevention Program. Expert Panel Report: guidelines for the diagnosis and management of asthma Update on selected Topicse2002. J Allergy Clin Immunol 2002 Nov;110(5 Suppl):S141e219. 8. Ducharme FM, Di Salvio F. Anti-leukotriene agents compared to inhaled corticosteroids in the management of recurrent and/or chronic asthma in adults and children. Cochrane Database Syst Rev 2004;(1). CD002314. 9. Szefler SJ, Phillips BR, Martinez FD, Chinchilli VM, Lemanske RF, Strunk RC, et al. Characterization of withinsubject responses to fluticasone and montelukast in childhood asthma. J Allergy Clin Immunol 2005 Feb;115(2): 233e42. 10. Ostrom NK, Decotiis BA, Lincourt WR, Edwards LD, Hanson KM, Carranza Jr R, et al. Comparative efficacy and safety of lowdose fluticasone propionate and montelukast in children with persistent asthma. J Pediatr 2005 Aug;147(2):213e20. 11. Szefler SJ, Baker JW, Uryniak T, Goldman M, Silkoff PE. Comparative study of budesonide inhalation suspension and montelukast in young children with mild persistent asthma. J Allergy Clin Immunol 2007 Nov;120(5):1043e50. 12. Sorkness CA, Lemanske Jr RF, Mauger DT, Boehmer SJ, Chinchilli VM, Martinez FD, et al. Long-term comparison of 3 controller regimens for mild-moderate persistent childhood asthma: the Pediatric Asthma Controller Trial. J Allergy Clin Immunol 2007 Jan;119(1):64e72.
L. Blais et al. 13. Sherman J, Patel P, Hutson A, Chesrown S, Hendeles L. Adherence to oral montelukast and inhaled fluticasone in children with persistent asthma. Pharmacotherapy 2001 Dec; 21(12):1464e7. 14. Carter ER, Ananthakrishnan M. Adherence to montelukast versus inhaled corticosteroids in children with asthma. Pediatr Pulmonology 2003 Oct;36(4):301e4. 15. Kelloway JS, Wyatt RA, Adlis SA. Comparison of patients’ compliance with prescribed oral and inhaled asthma medications. Arch Intern Med 1994 Jun 27;154(12):1349e52. 16. Weinberg EG, Naya I. Treatment preferences of adolescent patients with asthma. Pediatr Allergy Immunol 2000 Feb;11 (1):49e55. 17. Stempel DA, Kruzikas DT, Manjunath R. Comparative efficacy and cost of asthma care in children with asthma treated with fluticasone propionate and montelukast. J Pediatr 2007 Feb; 150(2):162e7. 18. Allen-Ramey FC, Markson LE, Riedel AA, Sajjan S, Weiss KB. Patterns of asthma-related health care resource use in children treated with montelukast or fluticasone. Curr Med Res Opin 2006 Aug;22(8):1453e61. 19. Re ´gie de l’assurance maladie du Que ´bec. Statistiques annuelles; 2006 Jan 1. 20. Blais L, Beauchesne MF. Use of inhaled corticosteroids following discharge from an emergency department for an acute exacerbation of asthma. Thorax 2004 Nov;59(11):943e7. 21. Blais L, Beauchesne MF, Rey E, Malo JL, Forget A. Use of inhaled corticosteroids during the first trimester of pregnancy and the risk of congenital malformations among women with asthma. Thorax 2007 Apr;62(4):320e8. 22. Marceau C, Lemie `re C, Berbiche D, Perreault S, Blais L. Persistence, adherence, and effectiveness of combination therapy among adult patients with asthma. J Allergy Clin Immunol 2006;118:574e81. ´, Beauchesne M-F, Perreault S, Lefebvre G, 23. Martel MJ, Rey E Forget A, et al. Use of inhaled corticosteroids during pregnancy and risk of pregnancy induced hypertension: nested casecontrol study. BMJ 2005;330:230e3. 24. Martel MJ, Rey E, Beauchesne MF, Perreault S, Forget A, Maghni K, et al. Use of short-acting beta2-agonists during pregnancy and the risk of pregnancy-induced hypertension. J Allergy Clin Immunol 2007 Mar;119(3):576e82. 25. Blais L, Lemie `re C, Menzies D, Berbiche D. Validity of asthma diagnoses recorded in the medical services database of Quebec. Pharmacoepidemiol Drug Safety 2005 Dec 23;15:245e52. 26. Tamblyn R, Lavoie G, Petrella L, Monette J. The use of prescription claims databases in pharmacoepidemiological research: the accuracy and comprehensiveness of the prescription claims database in Que ´bec. J Clin Epidemiol 1995; 48(8):999e1009. 27. Global Initiative for Asthma. Global strategy for asthma management and prevention; 2008. 28. Canny GJ, Reisman J, Healy R, Schwartz C, Petrou C, Rebuck AS, et al. Acute asthma: observations regarding the management of a pediatric emergency room. Pediatrics 1989 Apr;83(4):507e12. 29. Lemiere C, Bai T, Balter M, Bayliff C, Becker A, Boulet LP, et al. Adult asthma consensus guidelines update 2003. Can Respir J 2004 May;11(Suppl A):9Ae18A. 30. Blais L, Beauchesne M-F, Le ´vesque S. Socioeconomic status and medication prescription partterns in pediatric asthma in Canada. J Adolesc Health 2006;38(5). 607e9e607e16. 31. Pando S, Lemie `re C, Beauchesne MF, Perreault S, Forget A, Blais L. Suboptimal use of inhaled corticosteroids in children with persistent asthma: inadequate prescription, poor drug adherence, or both? Pharmacotherapy 2010;30(11):1109e16. 32. Hess LM, Raebel MA, Conner DA, Malone DC. Measurement of adherence in pharmacy administrative databases: a proposal
Inhaled corticosteroids vs. leukotriene-receptor antagonists
33. 34.
35.
36.
for standard definitions and preferred measures. Ann Pharmacother 2006 Jul;40(7e8):1280e8. Greenland S. Modeling and variable selection in epidemiologic analysis. Am J Pub Health 1989;79(3):340e9. Cabana MD, Rand CS, Becher OJ, Rubin HR. Reasons for pediatrician nonadherence to asthma guidelines. Arch Pediatr Adolesc Med 2001 Sep;155(9):1057e62. Cabana MD, Abu-Isa H, Thyne SM, Yawn B. Specialty differences in prescribing inhaled corticosteroids for children. Clin Pediatr (Phila) 2007 Oct;46(8):698e705. Bateman ED, Boushey HA, Bousquet J, Busse WW, Clark TJ, Pauwels RA, et al. Can guideline-defined asthma control be achieved? The gaining optimal asthma control study. Am J Respir Crit Care Med 2004 Oct 15;170(8):836e44.
855 37. Rabe KF, Atienza T, Magyar P, Larsson P, Jorup C, Lalloo UG. Effect of budesonide in combination with formoterol for reliever therapy in asthma exacerbations: a randomised controlled, double-blind study. Lancet 2006 Aug;368(9537):744e53. 38. Dombrowski MP, Schatz M, Wise R, Thom EA, Landon M, Mabie W, et al. Randomized trial of inhaled beclomethasone dipropionate versus theophylline for moderate asthma during pregnancy. Am J Obstetrics Gynaecol 2004 Mar;190(3):737e44. 39. Akinbami LJ, Schoendorf KC. Trends in childhood asthma: prevalence, health care utilization, and mortality. Pediatrics 2002 Aug;110(2 Pt 1):315e22. 40. Dales RE, Cakmak S, Judek S, Dann T, Coates F, Brook JR, et al. Influence of outdoor aeroallergens on hospitalization for asthma in Canada. J Allergy Clin Immunol 2004 Feb;113(2):303e6.
journal homepage: www.elsevier.com/locate/rmed
Inhaled corticosteroids vs. leukotriene-receptor antagonists and asthma exacerbations in children `re a,b, Lucie Blais a,b,c,*, Fatima-Zohra Kettani b, Catherine Lemie Marie-France Beauchesne a,b,c, Sylvie Perreault a, Naoual Elftouh b, Francine M. Ducharme d a
Universite´ de Montre´al, Faculte´ de Pharmacie, C.P. 6128, Succursale Centre-ville, Montre´al, Que´bec, Canada H3C 3J7 Research Center, Hoˆpital du Sacre´-Cœur de Montre´al, Montre´al, Que´bec, Canada c Endowment Pharmaceutical Chair AstraZeneca in Respiratory Health, Canada d Research Center, CHU Sainte-Justine and Department of Pediatrics, University of Montreal, Montreal, Quebec, Canada b
Received 24 September 2010; accepted 8 December 2010 Available online 31 December 2010
KEYWORDS Asthma; Child; Leukotriene receptor antagonists; Inhaled corticosteroids; Adherence; Exacerbations
Summary Background: Compared to inhaled corticosteroids (ICS), better use of leukotriene-receptor antagonists (LTRA) may lead to a greater reduction in exacerbations among asthmatic children in real-life settings. Methods: To test this hypothesis, we used the Quebec administrative databases and identified a cohort of 27,355 asthmatic children aged 5e15 years in whom ICS or LTRA monotherapy was initiated in 1998e2005. The primary outcome was the rate of moderate-or-severe asthma exacerbations (emergency department visit or hospitalization for asthma or a dispensed prescription of oral corticosteroids) over the subsequent year. The adjusted rate ratios (RR) of asthma exacerbations were estimated with Poisson regression models. To minimize confounding by indication, all analyses were stratified by the presence or not of an asthma exacerbation in the year before treatment initiation. We also measured the proportion of days with supply prescribed and patient’s adherence with the Proportion of Prescribed Days Covered (PPDC). Results: The risk of exacerbations was significantly higher in the ICS than the LTRA group among children with no previous exacerbation (RR Z 2.3; 95%CI:1.3e4.0), but not in those with 1 exacerbations (RR Z 1.6; 0.8e3.1). The PPDC was similar between the groups (66%) but the proportion of days with supply prescribed was significantly higher in the LTRA than the ICS group (52% vs. 34%), resulting in higher use.
* Corresponding author. Universite ´ de Montre ´al, Faculte ´ de Pharmacie, C.P. 6128, Succursale Centre-ville, Montre ´al, Que ´bec, Canada H3C 3J7. Tel.: þ1 514 343 6111x3786; fax: þ1 514 343 6120. E-mail address: [email protected] (L. Blais). 0954-6111/$ - see front matter ª 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.rmed.2010.12.007
Inhaled corticosteroids vs. leukotriene-receptor antagonists
847
Conclusions: While confounding by indication cannot be firmly ruled out, ICS appears to be more frequently prescribed as an intermittent than a daily controller therapy resulting in less use, which may contribute to the apparent lower effectiveness compared to LTRA. ª 2010 Elsevier Ltd. All rights reserved.
Introduction Regular daily therapy with inhaled corticosteroids (ICS) or leukotriene-receptor antagonists (LTRA) is effective in reducing asthma exacerbations in both adults and children.1e4 According to current guidelines on asthma management and prevention, ICS remain, however, the preferred controller monotherapy for persistent asthma.5e7 This recommendation is further supported by a 2004 Cochrane Review of 27 trials including adults and children reporting a 65% higher risk of exacerbations requiring systemic steroids in patients treated with LTRA compared to ICS.8 Subsequent to this review, several recent randomized clinical trials also demonstrated the superiority of ICS over LTRA in children for a broad range of clinical endpoints, including asthma control.9e12 Whether the superiority of ICS over LTRA persists in real life practice remains to be proven. Indeed, the use of ICS is notoriously low, frequently hovering around 45% of the time when filled prescriptions are examined.13 Although the use of LTRA is also not optimal, it generally appears superior to that of ICS, being reported to be between 60 and 70%,13,14 perhaps because of its oral administration, which makes it easier and more convenient to take than inhaled medications, particularly in children.15,16 Few observational studies have compared the relative effectiveness of ICS and LTRA in children,17,18 yielding conflicting results. While one study concluded that users of LTRA were more likely to experience treatment failure and to be admitted to the hospital for asthma,17 the other reported no group difference in the rate of asthma-related resource use, hospitalizations, or rescue medication use.18 None of these studies compared the proportion of days with supply prescribed, patient’s adherence to the prescribed therapy and the usage of ICS and LTRA, which are important elements to consider when examining drug effectiveness. We undertook a population-based study to investigate the relative effectiveness of ICS and LTRA for preventing moderate and severe asthma exacerbations among asthmatic children aged 5e15 years. We also aimed to explore determinants of effectiveness, namely the proportion of days with supply prescribed, adherence to, and use of the therapy, and prior asthma severity. Our hypothesis was that better adherence to LTRA than ICS would lead to a greater reduction in the rate of moderate or severe asthma exacerbations among patients treated with LTRA.
Methods Source of data Data for this study were retrieved from two administrative databases from the Canadian province of Que ´bec: the Re ´gie de l’Assurance Maladie du Que ´bec (RAMQ) and the MED-
ECHO databases that are linked by a patient unique encrypted identifier. The RAMQ provides medical coverage to all Quebec residents and drug coverage to about 42% of the population,19 including the elderly (65 years), persons receiving social assistance, and workers and their families (called the adherents) who do not have access to a private drug insurance plan from their workplace. Children of parents insured by the RAMQ Drug Insurance Plan are automatically insured and prescribed medications are free of charge for them. The RAMQ database includes three types of files. The Beneficiary file gives information on patients’ characteristics, such as age, sex, area of residence, social aid status and periods of coverage for the medication insurance. The Medical Services file contains all the information relative to medical services received e.g. the type, date, and location of medical service (outpatient clinic, emergency department (ED), hospitalization), diagnostic codes, as well as the specialty of the treating physician. The Prescription Drugs file contains data on prescribed medications dispensed in community pharmacies and includes notably the name of the medication, dosage, amount of medication given to the patient, the dispensing date, the type of prescription (new or refill), the number of refills allowed with a new prescription, the prescription duration, and the specialty of the prescribing physician. The MED-ECHO database contains information on all acute care hospitalizations occurring in the province, e.g. the date of admission and length of stay as well as primary and secondary diagnoses. These databases are routinely used for epidemiological asthma research.20e24 Data recorded in the RAMQ Prescription Drugs file and the medical diagnosis for asthma recorded in the RAMQ Medical Services file have been formally evaluated and validated.25,26 This study has been approved by the Ethics Committee of the Hoˆpital du Sacre´-Coeur de Montre´al.
Study design From the RAMQ database, we reconstructed a cohort of asthmatic children that initiated a treatment in monotherapy with either ICS [beclomethasone, fluticasone or budesonide] or LTRA [montelukast or zafirlukast]. Patients were eligible if they fulfilled the following inclusion criteria: (1) first prescription of monotherapy with ICS or LTRA between January 1st 1998 and August 31st, 2005; (2) no prior asthma controller therapy, that is, no dispensed ICS and LTRA prescriptions in the year preceding to the treatment initiation; (3) at least one asthma diagnosis (ICD-9 493) recorded in the RAMQ database in the year prior to treatment initiation; (4) age between 5 and 15 years at treatment initiation; (5) covered by the RAMQ Drug Insurance Plan for at least one year prior to treatment initiation; and (6) a follow-up of at least 4 months after treatment initiation. Patients were excluded if they had at least one diagnosis of chronic bronchitis, cystic fibrosis, dyspnea, or
848 laryngotracheitis or a dispensed medication to treat these conditions in the year prior to treatment initiation. Patients were excluded if they had regular therapy with oral corticosteroids in the year prior to treatment initiation (182 supply days or more) or in the first 3 months following treatment initiation (45 supply days or more). Patients were followed for a minimum of 4 months and a maximum of 12 months after treatment initiation. A minimum of 4 months of follow-up was required in order to obtain a precise measure of adherence to ICS and LTRA. Follow-up was terminated before the 12th month (i.e. data were censored) if one of the following events occurred: a dispensed prescription of another controller therapy (i.e. a class switch or a class addition), end of drug coverage by the RAMQ Drug Insurance Plan, 16th birthday, or death. Data were obtained from the RAMQ between January 1st, 1997 and December 31st, 2005. The cohort selected from the RAMQ was then linked to the MED-ECHO database with a unique patient identifier to obtain data on acute care hospitalizations that occurred during the same period.
Primary and secondary outcomes The primary outcome was the rate of moderate or severe asthma exacerbations defined as an emergency department (ED) visit for asthma, a hospital admission for asthma, or a dispensed short-course (14 days) prescription of oral corticosteroids. All the aforementioned events occurring during a 15-day interval were considered as a single exacerbation. The secondary outcome was a dispensed shortcourse prescription of oral corticosteroids, a marker of a moderate or severe asthma exacerbation27 more specific than an acute care visit where as many as 50% could be made for mild exacerbation.28
Compared therapy Children with asthma initiating a treatment with ICS as monotherapy were compared to those initiating a monotherapy with LTRA.
Confounding and stratification variables Potential confounders measured in the year before treatment initiation included average number of doses of inhaled short-acting beta2-agonists (SABA) per week (3, >3e10, >10), oral SABA (0, 1 filled prescription) and inhaled long-acting beta2-agonists (LABA) (0, 1 filled prescription). Potential confounders measured at treatment initiation were age (5e11, 12e15 years), gender, child living with parents receiving social assistance (yes/no), area of residence (urban/rural), specialty of the prescribing physician (family physician, paediatrician, respiratory specialist, others), and initial daily dose prescribed. To estimate the daily dose of ICS prescribed (expressed in beclomethasone-equivalent doses), we developed an algorithm based on the name of the medications and equivalencies between the different ICS products as recognized by the Canadian Asthma Consensus guidelines,29 the content of active ingredient per dosage unit and the quantity of
L. Blais et al. drug prescribed, the date and duration of the prescription, and the rate of renewals of the prescription. A second algorithm was developed in order to obtain an estimate of the average number of inhaled SABA filled per week, based on the name of the medications and equivalencies between the different SABA agents, as established by a pharmacist (M-F.B), date of filling, prescription renewals, and length of time between renewals.30 The average number of doses per week was divided into three categories (3, >3e10, >10 doses per week), according to the Canadian Asthma Consensus guidelines.29 In order to minimize potential confounding by indication due to the severity of asthma, we performed all analyses stratified on the occurrence or not of a moderate or severe asthma exacerbation in the year before treatment initiation.
Prescription pattern, treatment adherence and usage of medications Prescription pattern was assessed by the proportion of days with supply prescribed. This measure is obtained by dividing the total days’ supply prescribed (i.e. the sum of the duration of all prescriptionsdnew and allowed refills from all physicians consulted during the follow-up period) over the number of days of follow-up. Of note, the duration of allowed refills were summed whether or not the patient went to the pharmacy to get the refills. Treatment adherence was measured by the “Proportion of prescribed Days Covered” (PPDC), a new measure of patient’s adherence that has been recently developed by our group.31 The PPDC is defined as the total days’ supply dispensed during the follow-up over the total days’ supply prescribed.31 Finally, the usage of medications was summarized as the “number of prescription dispensed” and “the proportion of days covered” (PDC). The PDC is defined as the total number of days with supply dispensed during the follow-up over the number of days of follow-up.32 The PDC reflects the behaviour of both the patient and the prescribing physician and consequently the usage of medications, while the PPDC tends to measure more specifically the behaviour of the patient as it measures patient adherence to the therapy that was actually prescribed, and not to the therapy that should have been prescribed according to clinical guidelines, i.e. chronic daily use in the case of ICS and LTRA.
Statistical analysis All analyses, including the description of patient characteristics, were stratified by the presence or not of a moderate or severe asthma exacerbation (as defined in the primary outcome) in the year before treatment initiation with ICS or LTRA. We calculated the crude rate (per person per year) of moderate or severe asthma exacerbations and use of short-course of oral corticosteroids during follow-up, separately for patients treated with ICS and LTRA. The crude and the adjusted rate ratios (RR) of the outcomes, comparing patients treated with ICS to LTRA, were calculated using Poisson regression models with
Inhaled corticosteroids vs. leukotriene-receptor antagonists a dispersion parameter and reported with 95% confidence intervals (CIs). All potential confounding variables were entered in the models.33 We also calculated the mean (standard deviation) proportion of days with supply prescribed, PPDC, PDC and number of filled prescriptions of ICS and LTRA and compared these measures between ICS and LTRA patients using t-tests. All reported P-values are two-sided, with a significance level of 0.05. All analyses were performed using SAS software, version 9.1 (SAS Institute Inc., Cary, NC, USA).
Results Patients’ characteristics The cohort consisted of 27,355 children, including 7494 (27%) children with at least one exacerbation and 19,861 (73%) without any exacerbation in the year prior to treatment initiation. A flow chart with the number of patients excluded and the number of patients in which follow-up was censored is available in the online supplement. For note, among children who had suffered at least one exacerbation in the year prior treatment initiation, 74% experienced it in the 15 days preceding treatment initiation. Children’s characteristics are depicted in Table 1. Within strata defined by the presence or not of previous exacerbation, the two groups were comparable for sociodemographic characteristics, but we observed differences in variables related to asthma. Among children who had at least one exacerbation in the year preceding treatment initiation, a greater proportion of children treated with LTRA than ICS were prescribed their initial controller medication by a paediatrician, had used high doses of SABA, filled 2 prescriptions of oral corticosteroids, had made outpatient (non ED) visits for asthma, and had 2 moderate or severe exacerbations in the year prior to the initiation of the therapy. However, children treated with ICS were more likely to have had an ED visit and a hospitalization for asthma in the year prior to the initiation of the therapy. Similar trends were observed among children who had no exacerbation prior to treatment initiation. Among children who had at least one exacerbation in the year prior to treatment initiation, the proportion of children who received their initial prescription during an ED visit (prescription filled at most 15 days after the ED visit) was significantly higher in the ICS group than in the LTRA group (55% vs. 13%; p < 0.0001). The median starting daily dose of ICS prescribed was high (about 1000 mg/day), whether children had had an asthma exacerbation or not in the year before treatment initiation.
Crude rate of asthma exacerbations In the strata with no exacerbation in the preceding year, the crude rate of moderate to severe exacerbations, rescue oral corticosteroids, and ED visits was significantly higher among those treated with ICS than among those treated with LTRA; no group difference was found in the other strata (Table 2).
849
Adjusted rates of asthma exacerbations The adjusted rate ratios confirmed these above findings. In the strata with no previous exacerbation, the risk of exacerbations was two-fold higher in the ICS compared to the LTRA groups (adjusted RR Z 2.3; 95% CI; 1.3e4.0) (Table 3). A similar trend favouring LTRA but not reaching statistical significance was observed in the other strata (adjusted RR Z 1.6; 95% CI: 0.8e3.1). Comparable rate ratios favouring LTRA were seen regarding the use of oral corticosteroids in both strata, again statistically significant RR Z 2.3; 1.2e4.6 only among children with no previous exacerbation (RR Z 2.0; 0.8e5.3 among children with previous exacerbation). In both strata, an increased risk of asthma exacerbations with ICS compared to LTRA was associated with prescriptions by paediatricians at treatment initiation and with an increased number of doses of inhaled SABA per week in the year prior to treatment initiation. Oral SABA in the preceding year was also associated with an increased risk of exacerbations in the ICS group among children with a history of asthma exacerbations.
Treatment adherence to ICS and LTRA in the year following treatment initiation There was no group difference in patients’ adherence measured by the PPDC, which hovered around two-thirds for both groups (Table 4). Yet, the mean PDC, that is the proportion of days with a controller medication, was twofold higher in the LTRA group than the ICS group (p < 0.001), in both strata. Similarly, the proportion of days with supply prescribed and the number of prescriptions filled were significantly higher among children treated with LTRA than ICS.
Discussion In this observational study among Canadian children with asthma, initiation of monotherapy with ICS was associated with a two-fold higher rate of exacerbations compared to LTRA, but only among children with no exacerbation requiring a health care utilisation in the year before treatment initiation; there was no significant group difference among those with exacerbations in the preceding year. In both strata, patients’ adherence rates were similar between ICS and LTRA groups, hovering around two thirds of the prescribed doses. However, there was two-fold higher daily usage of LTRA than ICS, resulting from a combination of a higher proportion of days with a supply prescribed by the physician and a higher number of prescription dispensed per year in the LTRA than the ICS groups. The observed relative risk of exacerbations in our study contradicts those of previous randomized clinical trials demonstrating better outcomes with ICS compared with LTRA.912 The present investigation was based on data from actual practice, resulting from a combination of the physician prescription patterns and duration, and the patient adherence to filling the prescribed medications. While the results from clinical trials have high levels of
850
L. Blais et al.
Table 1
Children’s Characteristics. 1 exacerbation in the year prior to treatment initiation
No exacerbation in the year prior to treatment initiation
ICSa
LTRAb
ICSa
LTRAb
Number of patients Mean s.d. follow-up time (days)
7427 351 45
67 297 86
19439 350 46
422 321 74
At treatment initiation Age (years); n (%) 5e<12 12e15 Male gender, n (%) Receipt of social assistance; n (%)
5888 1539 4370 2772
49 18 34 28
(73.1) (26.9) (50.8) (41.8)
15017 (77.3) 4422 (22.7) 10642 (54.8) 7266 (37.4)
317 105 232 124
Area of residence; n (%)c Urban Rural Missing
5521 (74.3) 1905 (25.7) 1
55 (82.1) 12 (17.9) 0
15402 (79.3) 4033 (20.7) 4
327 (77.5) 95 (22.5) 0
Specialty of prescribing physician; n (%) Family physician Pediatrician Respiratory physician Other specialties Median daily dose prescribed
5130 (69.0) 2062 (27.8) 44 (0.6) 191 (2.6) 1071 mg
33 (49.2) 29 (43.3) 0 5 (7.5) 5 mg
13339 (68.6) 5428 (27.9) 265 (1.4) 407 (2.1) 1000 mg
202 (47.9) 181 (42.9) 12 (2.8) 27 (6.4) 5 mg
In the year preceding treatment initiation Average number of doses of inhaled SABAd per week; n (%) 3 6629 (89.3) >3e10 677 (9.1) >10 121 (1.6) Oral SABAd (1 filled prescription); n (%) 418 (5.6) 5 (0.1) LABAe (1 filled prescription); n (%) Theophylline (1 filled prescription); n (%) 8 (0.1)
55 (82.1) 10 (14.9) 2 (3.0) 5 (7.5) 1 (1.5) 0
18033 (92.8) 1189 (6.1) 217 (1.1) 895 (4.6) 26 (0.1) 35 (0.2)
379 (89.8) 37 (8.8) 6 (1.4) 29 (6.9) 4 (1.0) 1 (0.2)
Short-course filled prescription of oral corticosteroids; n (%) 0 3351 (45.1) 1 3601 (48.5) 2 475 (6.4) Outpatient medical visit for asthma (1); n (%) 2996 (40.3)
27 30 10 53
19439 (100.0)
422 (100.0)
19439 (100.0)
422(100.0)
EDf visit for asthma (1); n (%) Hospitalization for asthma (1); n (%)
5574 (75.1) 657 (8.9)
34(50.8) 4(6.0)
0 0
0 0
Asthma exacerbations; n (%) 0 1 2
0 6614 (89.0) 813 (11.0)
0 54 (80.6) 13 (19.4)
19439 (100.0) 0 0
422 (100.0) 0 0
a b c d e f
(79.3) (20.7) (58.8) (37.3)
(40.3) (44.8) (14.9) (79.1)
(75.1) (24.6) (55.0) (29.4)
ICS: Inhaled corticosteroids. LTRA: Leukotriene receptor antagonists. Measured in the 15 days preceding the index date. SABA: Short-acting beta-agonists. Long-acting beta-agonists. ED: Emergency department.
internal validity, their generalizability to the clinical practice setting is often questionable. Indeed, efficacy clinical trials are performed under ideal settings, with carefully selected patients, strict treatment protocols, and close
monitoring by the research team; poor adherence to drug or follow-up visit is a common exclusion criteria. Consequently, efficacy clinical trials minimize variability due to physician prescription patterns and patient adherence to
Inhaled corticosteroids vs. leukotriene-receptor antagonists Table 2
851
Crude Rate of Exacerbations for Children Treated with ICS or LTRA During Follow-up. 1 exacerbation in the year prior to treatment initiation
P-value No exacerbation in the year prior to treatment initiation
P-value
ICSa
LTRAb
ICSa
LTRAb
Number of patients
7427
67
19439
422
Exacerbations 1 exacerbation; n (%) Number of exacerbations Rate of exacerbation/patient/year
1671 (22.5) 2252 0.32
10 (14.9) 12 0.22
0.128
1535 (7.7) 1769 0.09
14 (3.3) 16 0.04
0.001
Oral corticosteroids 1 filled prescription; n (%) Number of filled prescriptions Rate of oral corticosteroids/patient/year
1222 (16.5) 1691 0.24
5 (7.4) 7 0.13
0.068
1222 (6.3) 1464 0.08
10 (2.4) 13 0.04
0.003
EDc visit for asthma 1 EDc visit for asthma; n (%) Number EDc visit for asthma Rate of EDc visit for asthma/patient/year
1158 (15.6) 1664 0.23
7 (10.5) 12 0.22
0.389
708 (3.6) 844 0.05
7 (1.7) 8 0.02
0.037
3 (4.5) 3 0.06
0.175
85 (0.4) 88 0.005
1 (0.2) 1 0.003
0.340
Hospitalizations for asthma 1 hospitalizations for asthma; n (%) 167 (2.2) Number of hospitalizations for asthma 191 Rate of hospitalizations for asthma/patient/year 0.03 a b c
ICS: Inhaled corticosteroids. LTRA: Leukotriene-receptor antagonists. ED: Emergency department.
the tested drug regimens. The possible role of prescribing patterns and patient adherence on the health outcomes is thus best explored using data from a real-world setting. Contrary to our initial hypothesis, patient adherence, measured by the proportion of prescribed days covered, was comparable between groups hovering around 66%. Of interest, the proportion of days with supply prescribed, reflecting prescription patterns and duration, is nearly twice higher for LTRA compared to ICS, suggesting that physician prescribe enough LTRA supply to last twice as long as the ICS supply. The unusually high dose of ICS hovering around 1000 mg/day and the higher rate of treatment initiation during an exacerbation, suggest that ICS is more frequently prescribed as a curative intermittent treatment during an exacerbation than as a daily controller medication, a pattern not found with LTRA. These findings would suggest that a large proportion of the non-adherence seen with the commonly employed “proportion of days covered (PDC)” measure may not be due to patients’ nonadherence to the prescribed medication, but rather to the pattern of ICS prescription. This is in line with the findings of recent US surveys which showed that only half of the primary care physicians reported concordance with the guidelines recommendations for the prescription of ICS therapy for children with persistent asthma.34,35 The resulting higher daily use of LTRA than ICS medications could therefore explain, at least in part, the greater reduction in exacerbations among children with asthma treated by LTRA compared to ICS.
The two-fold increased rate of asthma exacerbation associated with ICS observed in this study also differ from that reported in two similar other observational studies, which themselves yielded inconsistent results.17,18 Stempel and colleagues reported that children 4e17 years of age treated with montelukast were 2.5 times more likely to experience treatment failure (defined as asthma-related hospitalizations, ED visits, oral corticosteroids use, or switch/augmentation of index therapy) than those treated with fluticasone (OR Z 2.55; 95% CI: 2.19e2.96).17 This study did not include incident users of ICS or LTRA as we did, but specifically required active ICS therapy in the year in both groups before the index prescription; perhaps this criteria led to the selection of patients that were particularly responsive to their respective controlled medication. Conversely, Allen-Ramey et al. showed that among children 2e14 years of age who were new users of ICS or LTRA, there was no statistical difference between the montelukast and fluticasone cohorts in the risk of rescue oral corticosteroids (RR Z 1.02; 0.88e1.19). This result persisted in the subgroup of children aged 6e14 years, more comparable to our age group.18 No significant difference between the two groups in the risk of asthma-related ED visits or hospitalizations were neither noted in the 6e14 years age group. Those findings differ from ours, maybe because of differences in the baseline severity of patients between the two studies. Indeed, in the Allen-Ramey study, there was a significantly higher pre-index rate of oral corticosteroid use among those who initiated fluticasone than among
852
L. Blais et al.
Table 3 Crude and Adjusted Rate Ratios of Moderate or severe Asthma Exacerbations Comparing Children Treated with ICS and LTRA During Follow-up. 1 exacerbation in the year prior to treatment initiation
No exacerbation in the year prior to treatment initiation
Crude RRa
Adjusted RRa
95% CI
Crude RRa
Adjusted RRa
95% CI
1.4
1.6
0.8e3.1
2.2
2.3
1.3e4.0
At treatment initiation Age (<12 years vs.12e15 years) Male vs. female Area of residence (rural vs. urban) Social assistance (yes vs. no)
1.2 1.0 0.9 1.1
1.2 1.0 0.9 1.1
1.0e1.4 0.9e1.1 0.8e1.1 1.0e1.2
1.1 1.0 1.0 1.1
1.1 1.0 1.1 1.1
1.0e1.2 0.9e1.1 0.9e1.2 1.0e1.2
Specialty of prescribing physician Family physician Pediatrician Respiratory physician Other specialties
Reference 1.3 0.9 1.5
1.2 0.8 1.4
1.1e1.4 0.4e1.7 1.1e1.8
Reference 1.2 0.5 0.9
1.2 0.5 1.0
1.1e1.3 0.3e1.0 0.7e1.4
In the year prior to treatment initiation Number of doses of SABAd/week 0e3 Reference >3-10 1.8 >10 2.6 1.4 Oral SABAd (yes vs. no) LABAe (yes vs. no) 0.6
1.8 2.7 1.3 0.5
1.5e2.0 2.1e3.5 1.1e1.6 0.1e5.6
Reference 1.7 2.4 1.1 0.4
1.7 2.6 1.1 0.4
1.4e2.1 1.8e3.6 0.8e1.4 0.0e3.8
b
ICS vs. LTRA
c
a RR: Rate ratio; adjusted for age, sex, area of residence, social assistance, specialty of prescribing physician, average number of doses of inhaled short-acting beta2-agonists (SABA) per week, oral SABA and inhaled long-acting beta2-agonists (LABA). b ICS: Inhaled corticosteroids. c LTRA: Leukotriene-receptor antagonists. d SABA: Short-acting beta-agonists. e LABA: Long-acting beta-agonists.
those who initiated montelukast, contrarily to our study. Of interest, none of these studies compared the proportion of days covered by the prescribed therapy, patient’s adherence to the prescribed therapy and the usage of ICS and LTRA, which are important elements to consider when examining drug effectiveness. Our study has several strengths. First, it is a populationbased effectiveness study, representative of real-world clinical practice. Second, our data were obtained from a validated healthcare database providing detailed information on prescribed medications recorded prospectively, thus excluding the possibility of recall bias. Third, we used established and objective markers of moderate and severe asthma exacerbations.36e38 Finally, the large sample size provided precise estimates of association. We acknowledge the following limitations. First and foremost, as for any non-randomized population-based study, it is not possible to control treatment assignment which could potentially lead to systematic differences between the two study groups, a phenomenon called confounding by indication. In other words, ICS may have been used for children with more severe asthma than those treated with LTRA. We attempted to minimize such bias by stratifying all the analyses for the occurrence or not of
a moderate or severe asthma exacerbation in the year before treatment initiation. Although the stratification led to comparable treatment groups for most variables in the strata with no prior exacerbation, we cannot rule out imbalances in other unmeasured variables known to be associated with asthma morbidity, such as ethnicity or environmental factors (aeroallergens).39,40 In the strata with prior exacerbations, several group imbalances remained that could have masked important group differences in health outcomes. Other clinical markers of disease severity at the baseline such as lung function or asthma symptoms were not provided by administrative databases and thus could not be used for a more precise assessment of asthma severity. The data retrieved from the Medical and Prescription databases were validated in adults and not in children. Adherence to study medications was based on dispensed prescriptions and not the actual use of medications; while one cannot prove intake from dispensed medication, intake cannot take place without first dispensing. The use of administrative databases does not allow determining whether patients have filled their new prescription, which may have resulted in an underestimation of the number of new prescriptions and allowed refills as only prescriptions that reach the pharmacy are
Inhaled corticosteroids vs. leukotriene-receptor antagonists Table 4
853
Treatment Adherence to ICS and LTRA in the Year Following Treatment Initiation. 1 exacerbation prior to treatment initiation
LTRAb
67
19439
422
Proportion of days with supply prescribed Mean percentage (s d) 32.8 (30.1)
55.9 (35.9)c
34.0 (30.9)
51.9 (35.1)c
Proportion of Prescribed Days Covered (PPDC ) Mean percentage (s d) 68.0 (33.0)
64.7 (29.3)d
63.9 (34.4)
62.7 (33.5)d
Proportion of Days Covered (PDC ) Mean percentage (s d) 15.1 (12.2)
33.3 (28.3)c
14.2 (11.6)
29.4 (28.7)c
Number of prescriptions dispensed per year Mean number (s d) 4.4 (5.7)
7.9 (11.3)c
4.4 (5.2)
6.8 (8.1)c
Number of patients
7427
LTRA
b
No exacerbation prior to treatment initiation ICSa
ICS
a
DEFINITIONS:Proportion of days with supply prescribed Z Total days’ supply prescribed (i.e. the sum of the duration of all new prescriptions plus allowed refills over the follow-up period)/Duration of follow-up: reflects prescription patterns.Proportion of Prescribed Days Covered (PPDC) Z Total days’ supply dispensed over the follow-up period/Total days’ supply prescribed: reflects patients’ adherence to prescribed therapy.Proportion of Days Covered (PDC) Z Total days’ supply dispensed during follow-up period/ Duration of Follow-up: reflects usage of medications (both patient and physician behaviour). a ICS: Inhaled corticosteroids. b LTRA: Leukotriene receptor antagonists. c All p-values < 0.001 for the ICS/LTRA comparison with a t-test (PDSP, PDC and number of prescriptions per year). d P-value Z 0.184 for the ICS/LTRA comparison with a t-test (PPDC).
accounted for in the database. Our new measure of patient adherence was based on the total days’ supply prescribed, i.e. the sum of the duration of all prescriptions e new and allowed refills from all physicians consulted during the follow-up period. Thus, it doesn’t take into account the variation in the doses that may have occurred between two refills based on action plans for example. So, it is possible that adherence would be underestimated particularly in the case of ICS. Finally, the absence of a significant result in the group of patients with a previous exacerbation might be due to reduced power driven by the small number of patients treated with LTRA. With regards to generalizability, our population is mostly representative of children whose parents are from the lower to middle class population of Quebec. However, the free healthcare and access to medication in children of Quebec may make the impact of social class less important than in other health care systems. In conclusion, this effectiveness study of real life drug utilisation patterns in the Canadian setting revealed that the risk of asthma exacerbations was two-fold higher in children prescribed ICS compared to LTRA in children without exacerbations requiring health care resources in the year prior to treatment initiation; with no difference in children with prior exacerbations. Higher proportion of days with a supply prescribed and of days of use of LTRA compared to ICS in light of similar patient’s adherence, suggest that the prescribing of ICS as an intermittent rather than a daily controller explain in part our findings. Yet, confounding by indication cannot be firmly ruled out. Our study demonstrates the importance of considering the prescribing patterns while measuring patients’ adherence with prescription refills recorded in administrative databases.
Ethics statement This study was approved by the Scientific and Ethics Committee of the Ho ˆpital du Sacre ´-Coeur de Montre ´al, Quebec, Canada.
Funding Lucie Blais, Catherine Lemie `re, and Sylvie Perreault are the recipients of a Salary Award from the Fonds de la recherche en sante´ du Que´bec (FRSQ). This study was funded through a grant received from the Fonds de la recherche en sante´ du Que´bec (FRSQ) and the Conseil du me´dicament du Que´bec.
Conflict of interest statement Lucie Blais received research grants from Sanofi-Aventis, AstraZeneca and GlaxoSmithKline and co-chairs the AstraZeneca Endowment Pharmaceutical Chair in Respiratory Health. Marie-France Beauchesne has received grant support from AstraZeneca Canada Inc and GlaxoSmithKline Canada Inc for the conduct of her research projects. She has also received honoraria for education programs given to pharmacists from GlaxoSmithKline Canada Inc, Boehringer Ingelheim Canada Inc., AstraZeneca Inc, and Pfizer Canada Inc. She co-chairs the AstraZeneca Endowment Pharmaceutical Chair in Respiratory Health. Francine Ducharme received research grants from GlaxoSmithKline and Merck. She has also received travel support for meeting attendance,
854 research funds, fees for speaking and/or consulting fees from AstraZeneca, Merck Frosst Inc, GlaxoSmithKline and Novartis. Naoual Elftouh, Fatima-Zohra Kettani, Catherine Lemie `re and Sylvie Perreault have no competing interests to declare.
Supplementary material The supplementary data associated with this article can be found in online version at doi:10.1016/j.rmed. 2010.12.007.
References 1. Suissa S, Ernst P, Kezouh A. Regular use of inhaled corticosteroids and the long term prevention of hospitalisation for asthma. Thorax 2002 Oct;57(10):880e4. 2. Barnes NC, Miller CJ. Effect of leukotriene receptor antagonist therapy on the risk of asthma exacerbations in patients with mild to moderate asthma: an integrated analysis of zafirlukast trials. Thorax 2000 Jun;55(6):478e83. 3. The Childhood Asthma Management Program Research Group. Long-term effects of budesonide or nedocromil in children with asthma. N Engl J Med 2000 Oct 12;343(15):1054e63. 4. Garcia Garcia ML, Wahn U, Gilles L, Swern A, Tozzi CA, Polos P. Montelukast, compared with fluticasone, for control of asthma among 6- to 14-year-old patients with mild asthma: the MOSAIC study. Pediatrics 2005 Aug;116(2):360e9. 5. Becker A, Berube D, Chad Z, Dolovich M, Ducharme F, D’Urzo T, et al. Canadian pediatric asthma consensus guidelines, 2003 (updated To December 2004): introduction. CMAJ 2005 Sep 13; 173(6 Suppl):S12e4. 6. Bateman ED, Hurd SS, Barnes PJ, Bousquet J, Drazen JM, FitzGerald M, et al. Global strategy for asthma management and prevention: GINA executive summary. Eur Respir J 2008 Jan;31(1):143e78. 7. National Asthma Education and Prevention Program. Expert Panel Report: guidelines for the diagnosis and management of asthma Update on selected Topicse2002. J Allergy Clin Immunol 2002 Nov;110(5 Suppl):S141e219. 8. Ducharme FM, Di Salvio F. Anti-leukotriene agents compared to inhaled corticosteroids in the management of recurrent and/or chronic asthma in adults and children. Cochrane Database Syst Rev 2004;(1). CD002314. 9. Szefler SJ, Phillips BR, Martinez FD, Chinchilli VM, Lemanske RF, Strunk RC, et al. Characterization of withinsubject responses to fluticasone and montelukast in childhood asthma. J Allergy Clin Immunol 2005 Feb;115(2): 233e42. 10. Ostrom NK, Decotiis BA, Lincourt WR, Edwards LD, Hanson KM, Carranza Jr R, et al. Comparative efficacy and safety of lowdose fluticasone propionate and montelukast in children with persistent asthma. J Pediatr 2005 Aug;147(2):213e20. 11. Szefler SJ, Baker JW, Uryniak T, Goldman M, Silkoff PE. Comparative study of budesonide inhalation suspension and montelukast in young children with mild persistent asthma. J Allergy Clin Immunol 2007 Nov;120(5):1043e50. 12. Sorkness CA, Lemanske Jr RF, Mauger DT, Boehmer SJ, Chinchilli VM, Martinez FD, et al. Long-term comparison of 3 controller regimens for mild-moderate persistent childhood asthma: the Pediatric Asthma Controller Trial. J Allergy Clin Immunol 2007 Jan;119(1):64e72.
L. Blais et al. 13. Sherman J, Patel P, Hutson A, Chesrown S, Hendeles L. Adherence to oral montelukast and inhaled fluticasone in children with persistent asthma. Pharmacotherapy 2001 Dec; 21(12):1464e7. 14. Carter ER, Ananthakrishnan M. Adherence to montelukast versus inhaled corticosteroids in children with asthma. Pediatr Pulmonology 2003 Oct;36(4):301e4. 15. Kelloway JS, Wyatt RA, Adlis SA. Comparison of patients’ compliance with prescribed oral and inhaled asthma medications. Arch Intern Med 1994 Jun 27;154(12):1349e52. 16. Weinberg EG, Naya I. Treatment preferences of adolescent patients with asthma. Pediatr Allergy Immunol 2000 Feb;11 (1):49e55. 17. Stempel DA, Kruzikas DT, Manjunath R. Comparative efficacy and cost of asthma care in children with asthma treated with fluticasone propionate and montelukast. J Pediatr 2007 Feb; 150(2):162e7. 18. Allen-Ramey FC, Markson LE, Riedel AA, Sajjan S, Weiss KB. Patterns of asthma-related health care resource use in children treated with montelukast or fluticasone. Curr Med Res Opin 2006 Aug;22(8):1453e61. 19. Re ´gie de l’assurance maladie du Que ´bec. Statistiques annuelles; 2006 Jan 1. 20. Blais L, Beauchesne MF. Use of inhaled corticosteroids following discharge from an emergency department for an acute exacerbation of asthma. Thorax 2004 Nov;59(11):943e7. 21. Blais L, Beauchesne MF, Rey E, Malo JL, Forget A. Use of inhaled corticosteroids during the first trimester of pregnancy and the risk of congenital malformations among women with asthma. Thorax 2007 Apr;62(4):320e8. 22. Marceau C, Lemie `re C, Berbiche D, Perreault S, Blais L. Persistence, adherence, and effectiveness of combination therapy among adult patients with asthma. J Allergy Clin Immunol 2006;118:574e81. ´, Beauchesne M-F, Perreault S, Lefebvre G, 23. Martel MJ, Rey E Forget A, et al. Use of inhaled corticosteroids during pregnancy and risk of pregnancy induced hypertension: nested casecontrol study. BMJ 2005;330:230e3. 24. Martel MJ, Rey E, Beauchesne MF, Perreault S, Forget A, Maghni K, et al. Use of short-acting beta2-agonists during pregnancy and the risk of pregnancy-induced hypertension. J Allergy Clin Immunol 2007 Mar;119(3):576e82. 25. Blais L, Lemie `re C, Menzies D, Berbiche D. Validity of asthma diagnoses recorded in the medical services database of Quebec. Pharmacoepidemiol Drug Safety 2005 Dec 23;15:245e52. 26. Tamblyn R, Lavoie G, Petrella L, Monette J. The use of prescription claims databases in pharmacoepidemiological research: the accuracy and comprehensiveness of the prescription claims database in Que ´bec. J Clin Epidemiol 1995; 48(8):999e1009. 27. Global Initiative for Asthma. Global strategy for asthma management and prevention; 2008. 28. Canny GJ, Reisman J, Healy R, Schwartz C, Petrou C, Rebuck AS, et al. Acute asthma: observations regarding the management of a pediatric emergency room. Pediatrics 1989 Apr;83(4):507e12. 29. Lemiere C, Bai T, Balter M, Bayliff C, Becker A, Boulet LP, et al. Adult asthma consensus guidelines update 2003. Can Respir J 2004 May;11(Suppl A):9Ae18A. 30. Blais L, Beauchesne M-F, Le ´vesque S. Socioeconomic status and medication prescription partterns in pediatric asthma in Canada. J Adolesc Health 2006;38(5). 607e9e607e16. 31. Pando S, Lemie `re C, Beauchesne MF, Perreault S, Forget A, Blais L. Suboptimal use of inhaled corticosteroids in children with persistent asthma: inadequate prescription, poor drug adherence, or both? Pharmacotherapy 2010;30(11):1109e16. 32. Hess LM, Raebel MA, Conner DA, Malone DC. Measurement of adherence in pharmacy administrative databases: a proposal
Inhaled corticosteroids vs. leukotriene-receptor antagonists
33. 34.
35.
36.
for standard definitions and preferred measures. Ann Pharmacother 2006 Jul;40(7e8):1280e8. Greenland S. Modeling and variable selection in epidemiologic analysis. Am J Pub Health 1989;79(3):340e9. Cabana MD, Rand CS, Becher OJ, Rubin HR. Reasons for pediatrician nonadherence to asthma guidelines. Arch Pediatr Adolesc Med 2001 Sep;155(9):1057e62. Cabana MD, Abu-Isa H, Thyne SM, Yawn B. Specialty differences in prescribing inhaled corticosteroids for children. Clin Pediatr (Phila) 2007 Oct;46(8):698e705. Bateman ED, Boushey HA, Bousquet J, Busse WW, Clark TJ, Pauwels RA, et al. Can guideline-defined asthma control be achieved? The gaining optimal asthma control study. Am J Respir Crit Care Med 2004 Oct 15;170(8):836e44.
855 37. Rabe KF, Atienza T, Magyar P, Larsson P, Jorup C, Lalloo UG. Effect of budesonide in combination with formoterol for reliever therapy in asthma exacerbations: a randomised controlled, double-blind study. Lancet 2006 Aug;368(9537):744e53. 38. Dombrowski MP, Schatz M, Wise R, Thom EA, Landon M, Mabie W, et al. Randomized trial of inhaled beclomethasone dipropionate versus theophylline for moderate asthma during pregnancy. Am J Obstetrics Gynaecol 2004 Mar;190(3):737e44. 39. Akinbami LJ, Schoendorf KC. Trends in childhood asthma: prevalence, health care utilization, and mortality. Pediatrics 2002 Aug;110(2 Pt 1):315e22. 40. Dales RE, Cakmak S, Judek S, Dann T, Coates F, Brook JR, et al. Influence of outdoor aeroallergens on hospitalization for asthma in Canada. J Allergy Clin Immunol 2004 Feb;113(2):303e6.