Determinants of response to fluticasone propionate and salmeterol/fluticasone propionate combination in the Gaining Optimal Asthma controL study

Article Outline

• Abstract
• Methods
  • Definitions of asthma control
  • Criteria used to assess clinical benefits in patients with NWC asthma
  • Statistical analysis
• Results
  • Baseline factors influencing control
  • Individual clinical benefits of treatment in patients with NWC asthma
• Discussion
• Acknowledgment
• References
• Copyright

Background

During the Gaining Optimal Asthma controL study, 3416 patients with uncontrolled asthma were randomized to receive salmeterol/fluticasone propionate combination (SFC) or fluticasone propionate (FP) for 1 year. Approximately two thirds of patients achieved well-controlled (WC) asthma, and one third continued to have asthma that was not well controlled (NWC).

Objective

This analysis aimed to (1) identify factors influencing treatment response and (2) assess the clinical benefits of SFC and FP in patients with NWC asthma.

Methods

Logistic regression analysis was used to investigate whether covariates influenced the achievement of at least WC asthma in the study population. In patients with NWC asthma, predefined criteria were used to assess improvements in 6 clinical outcomes.

Results

Factors affecting the probability of having NWC asthma included smoking status (current vs never: odds ratio [OR], 2.757; 95% CI, 2.061−3.689; P < .0001; former vs never: OR, 1.274; 95% CI, 1.031−1.574; P = 0.0273), sex (women vs men: OR, 0.652; 95% CI, 0.527–0.806; P < .0001), history of inhaled corticosteroid use (no history vs history: OR, 0.546; 95% CI, 0.437−0.683; P < .0001), and treatment (FP vs SFC: OR, 1.972; 95% CI, 1.686–2.308; P < .0001). Of patients with NWC asthma, 86% to 96% showed improvements in 1 or more clinical outcomes.

Conclusion

It is imperative for good asthma control that patients stop smoking. Patients who did not have at least WC asthma demonstrated clinical improvements in individual asthma outcomes.

Clinical implications

Although not all patients can achieve guideline-defined control, long-term treatment with SFC or FP is associated with clinical improvements in nearly all patients, regardless of smoking history or inhaled corticosteroid use.

Key words: Asthma, salmeterol/fluticasone propionate combination, asthma control, treatment response, smoking

Abbreviations used: AQLQ, Asthma quality of life questionnaire, BDP, Beclomethasone dipropionate, FP, Fluticasone propionate, GOAL, Gaining Optimal Asthma controL, GR, Glucocorticoid receptor, ICS, Inhaled corticosteroid, NWC, Not well controlled, OR, Odds ratio, PEF, Peak expiratory flow, PEFR, Peak expiratory flow rate, SFC, Salmeterol/fluticasone propionate combination, TC, Total control, WC, Well controlled

The Gaining Optimal Asthma controL (GOAL) study¹ provided the first evidence of the clinical benefits of aiming for complete and sustained guideline-defined control of asthma. In the GOAL study, definitions of totally controlled (TC) and well-controlled (WC) asthma were derived from the Global Initiative for Asthma/National Institutes of Health guidelines.², ³ Both control definitions were composite measures, incorporating a range of end points in a single definition. These composite measures represent a more comprehensive and sustained assessment of control and are therefore more difficult to achieve than a single end point outcome. Depending on the treatment group, between 59% and 71% of patients achieved the composite criteria for WC asthma for at least 7 of 8 weeks at the end of the 12-month study. Patients who did not meet these stringent control levels were categorized as having asthma that was not well controlled (NWC), even if they had improvements in some or all of the individual end points comprising the TC and WC definitions.

A number of factors have been suggested to influence the response to corticosteroid therapy, including smoking,⁴, ⁵, ⁶ age,⁷ asthma duration,⁷ level of blood and sputum eosinophils,⁷ pretreatment FEV₁,⁷ and higher response rates to short-acting bronchodilators before treatment.⁷ However, these studies used single outcomes rather than composite end points for assessing responses and did not include an assessment of responses to combination treatment with an inhaled corticosteroid (ICS) and a long-acting β₂-agonist.

The aim of this analysis was first to identify factors that influenced treatment response in the treatment arms of the GOAL study (treatment with the long-acting β₂-agonist salmeterol combined with the corticosteroid fluticasone propionate [SFC] or treatment with fluticasone propionate [FP] alone). The second aim was to assess individual clinical improvements in patients who did not achieve the strict criteria of at least WC asthma with these treatment options.

Back to Article Outline

Methods 

A detailed methodology of the GOAL study has been published previously.¹ The GOAL study was a 12-month, stratified, double-blind, parallel-group study in 3416 patients with mainly uncontrolled asthma randomized to receive either SFC (Seretide/Advair/Viani; GlaxoSmithKline, Middlesex, United Kingdom) or FP alone (Flixotide/Flovent, GlaxoSmithKline). Study treatment was stratified according to ICS use in the previous 6 months: stratum 1, no ICS; stratum 2, 500 μg or less of beclomethasone dipropionate (BDP) or equivalent; and stratum 3, 500 to 1000 μg of BDP or equivalent. Patients who had a smoking history of 10 pack-years or more (eg, 20 cigarettes per day for 10 years or 40 cigarettes per day for 5 years) or those who were likely to change their smoking habits during the course of the study or who had given up smoking within 4 weeks of visit 1 were excluded from the study.

Patients who did not have WC asthma during the run-in period were randomized and entered into phase I of the study (the step-up phase), during which their treatment was increased every 12 weeks until TC or a maximum daily dose of 1000 μg of FP/SFC was reached. Control was assessed over an 8-week period before each clinic visit at weeks 12, 24, 36, and 52.¹ Patients who achieved TC during phase I entered phase II, continuing on the dose at which TC had been achieved for the next 16 to 40 weeks of the study. Patients who did not achieve TC during phase I entered phase II after 12 weeks of the maximum study dose and stayed on that dose for the remainder of the 52-week study.

Definitions of asthma control 

In the GOAL study, asthma control was assessed over an 8-week period before each clinic visit by using a composite measure of symptoms: predicted morning peak expiratory flow (PEF; as a percentage), FEV₁, nighttime awakenings, daytime symptoms, use of rescue β₂-agonist, rate of exacerbations, and adverse events (Table I). Patients who failed to meet the stringent control level of TC or at least WC at the end of phase II were categorized as having NWC asthma, even if they had improvements in some or all of the individual clinical end points comprising the WC definition.

Table I. Definitions of WC and TC asthma based on GINA/NIH guideline aims of treatment

TC was achieved if the patient had at least 7 of 8 TC weeks, and WC was achieved if the patient had at least 7 of 8 WC weeks. Patients who had at least WC asthma met all the criteria for WC asthma; a proportion of these patients also met the criteria for TC asthma.

GINA, Global Initiative for Asthma; NIH, National Institutes of Health.

Criteria used to assess clinical benefits in patients with NWC asthma 

To characterize improvements in individual clinical outcomes in patients with NWC disease, the following criteria were identified a priori to define what were considered to be important improvements: (1) PEF, a change from baseline of 15 L/min or greater at week 52; (2) predicted FEV₁, greater than 90% (and greater than baseline) or change of 6% or greater at week 52; (3) nighttime awakenings, greater than 50% reduction in the number of awakenings experienced in the last 4 weeks of treatment compared with the 4-week run-in period; (4) symptoms, greater than 50% reduction in the number of days with symptoms in the last 4 weeks of treatment compared with the 4-week run-in period; (5) rescue medication use, greater than 50% reduction in the number of days on which rescue medication was used in the last 4 weeks of treatment compared with the 4-week run-in period; and (6) total Asthma Quality of Life Questionnaire overall score, change of greater than 0.5 from baseline score at the end of treatment. If a patient achieved any single criterion, they were classified as “improving.” Diary card data from the last 4 weeks of the study (if these data were missing, the last 4 weeks of available data were used) were analyzed and compared directly with those during the 4-week run-in period.

Statistical analysis 

A logistic regression analysis⁸ was performed to investigate whether various covariates affected whether patients' symptoms were at least WC at the end of the study. This logistic regression (using the binary response of NWC vs at least WC) used the main effects from the primary analysis of the GOAL study (ie, country amalgamation, age, sex, and baseline FEV₁). The stepwise regression procedure⁹ was used for each stratum to assess whether covariates had a significant influence in the model (at the 5% significance level). The stepwise regression used was a combination of forward regression to choose the model, followed by backward regression to check any previous covariates that were no longer significant and could be removed.⁸ The covariates tested (in addition to the main effects listed previously) were height (patients were grouped by increments of 10 cm), tobacco use (current smoker [≤10 cigarette pack-years], former smoker, or never smoked), history of atopy (diagnosed allergic reaction) and history of an allergy (specific allergic reaction), history of moderate (requiring antibiotics, oral corticosteroids, or both) and severe (requiring hospitalization) exacerbations, baseline PEF (in liters per minute), duration of asthma (<6 months, ≥6 months to <1 year, ≥1 year to <5 years; durations increased in 5-yearly intervals from ≥5 years until <25 years, and the final group was ≥25 years), previous ICS use, and baseline nighttime awakenings. Interactions were not tested in the model because of the high number of covariates. Where no unit or category is specified, a yes or no variable was used. Baseline diary card data were those collected over the last week of the run-in period. No corrections for multiple comparisons were made because of the selection procedure used and also because one model was selected for each stratum. Odds ratios (ORs) were calculated to provide an estimate of the odds of having WC asthma for each of the various covariates affecting control of asthma. Exacerbation rates were calculated by using the Poisson model, adjusting for baseline, stratum, sex, age, smoking status, and smoking by treatment interaction. All statistical analyses were conducted with SAS version 8.2 software (SAS Institute, Cary, NC). Descriptive statistics were used to describe the proportion of patients with NWC asthma achieving each of the predefined criteria for improvement in the individual outcomes during the treatment period.

Back to Article Outline

Results 

A total of 1206 (35%) patients (710 treated with FP and 496 treated with SFC) in the GOAL population failed to achieve at least WC during phase I or at the end of phase II. Compared with those who had at least WC asthma, the NWC group had a larger proportion of current and former smokers, higher baseline rescue medication use, a greater mean number of nighttime awakenings, higher mean symptom scores in the previous 24 hours, and lower mean morning PEF rate (PEFR); the 2 groups were similar in terms of male to female ratio and baseline age (Table II). These were baseline data in a randomized clinical trial, and therefore the differences between the WC and NWC groups were not assessed for statistical significance at baseline.⁹ Results are presented for all strata combined because there were few differences between strata.

Table II. Baseline characteristics of the total population, those with at least WC asthma, and those with NWC asthma

Baseline factors influencing control 

The factors that are most predictive of NWC asthma are presented in Table III for all patients. Smoking status and treatment had the greatest influence on achieving at least WC asthma (Fig 1). The observed ORs of having NWC asthma were increased by 176% in current smokers compared with those who had never smoked, by 97% in patients who received FP compared with those treated with SFC, by 90% for each nocturnal asthma awakening during baseline (all P < .0001), and by 27% in former smokers compared with those who had never smoked (all P = .0273). Female patients and patients with no history of prior ICS use were 65% and 55% more likely to have at least WC asthma than male patients or patients with a history of ICS use, respectively (both P < .0001). For every 10-cm increase in height, the patients increased their odds of having NWC asthma by 21% (P = .0023). The ORs were similar in the various strata for age, FEV₁, baseline nighttime awakenings, and smoking status. The effects of sex and treatment were greater in both stratum 2 and stratum 3 than in stratum 1. Sex was only statistically significant in stratum 2 and overall (both P < .0001). The effects of height and baseline PEFR were included in stratum 3 and the overall model but not selected for stratum 1 and stratum 2 models because they did not significantly influence the response.

Table III. ORs of factors affecting the probability of the asthma being NWC in the total patient population

Patients with missing covariates were excluded. Numbers of patients included were as follows: FP, 1669; SFC, 1676. Baseline is calculated as the week before randomization for PEFR, nighttime awakenings, and the randomization visit for FEV₁. ORs are the result of the logistic regression analysis and are adjusted for all other significant covariates.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  Influence of smoking status on the proportion of patients achieving WC asthma at the end of phase II (A) and exacerbation rates (B; all strata). Comparison for exacerbation rates was made by using the Poisson model, adjusting for baseline, stratum, sex, age, smoking status, and smoking by treatment interaction.

A total of 1259 (74%) patients in the FP group and 1267 (74%) patients in the SFC group never smoked; 306 (FP; 18%) and 312 (SFC; 18%) were former smokers, and 142 (FP; 8%) and 130 (SFC; 8%) were current smokers. A higher proportion of patients achieved WC asthma in the SFC group than in the FP group, irrespective of smoking status (Fig 1, A).

Asthma severity did not appear to be related to smoking status. Patients who had never smoked, current smokers, and former smokers had baseline FEV₁ predicted values of 77%, 77%, and 76%, respectively. Reversibility of airflow obstruction based on FEV₁ after inhalation of a short-acting β₂-agonist was 27%, regardless of smoking status.

The influence of smoking on achievement of at least WC asthma and risk of exacerbation is shown in Fig 1. For FP-treated patients, the exacerbation rate was twice as high in current smokers as in patients who had never smoked (0.35 vs 0.17 per patient per year). The difference between these 2 groups was less marked in SFC-treated patients (0.20 vs 0.13 per patient per year). The proportion of patients achieving WC asthma paralleled the results for exacerbations; patients who had never smoked were more likely to achieve at least WC asthma than former or current smokers.

Smoking status also influenced the improvement in lung function with study treatment. In the nonsmoking group the improvement in FEV₁ was 0.25 L and 0.43 L in FP-treated and SFC-treated patients, respectively. Improvement in lung function for former smokers was 0.18 L and 0.32 L in FP-treated and SFC-treated patients, respectively, and 0.19 L and 0.28 L, respectively, for current smokers.

Individual clinical benefits of treatment in patients with NWC asthma 

Between 86% and 96% of the patients with NWC asthma (depending on stratum) showed overall improvements in at least 1 of the criteria used to assess clinical benefit (Table IV). More patients showed improvements in lung function (PEF, 45% to 73%; FEV₁, 42% to 67%; depending on stratum) and nighttime awakenings (63% to 69% depending on stratum) than in other criteria, such as daytime symptom scores (15% to 38% depending on stratum) and rescue use of β₂-agonist (28% to 50% depending on stratum). Approximately half of the patients showed improvements in 3 to 6 criteria (Fig 2).

Table IV. Number of patients with NWC asthma passing individual criteria used to assess benefit

Stratum 1, No ICS at baseline; stratum 2, 500 μg or less of BDP or equivalent at baseline; stratum 3, 500 to 1000 μg of BDP or equivalent at baseline; symptom score, greater than 50% reduction in the number of days with symptoms in the last 4 weeks of treatment compared with the 4-week run-in period; rescue medication use, greater than 50% reduction in the number of days on which rescue medication was used in the last 4 weeks of treatment compared with the 4-week run-in period; nighttime awakenings, greater than 50% reduction in the number of awakenings experienced in the last 4 weeks of treatment compared with the 4-week run-in period; PEF, a change from baseline of 15 L/min or greater at week 52; predicted FEV₁, greater than 90% (and greater than baseline) or change of 6% or greater at week 52; AQLQ, a change of greater than 0.5 from baseline at the end of treatment.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  Number of individual criteria passed by patients with NWC asthma (all strata).

When analyzed according to smoking status, between 89% and 95% of patients with NWC asthma showed overall improvements in at least 1 criterion used to assess clinical benefit. Smoking status appeared to mainly influence the improvements in rescue remedy use; other factors varied according to smoking status.

In general, SFC-treated patients showed improvements in more criteria than those in the FP arm (Fig 2), regardless of smoking status. For all strata combined, the proportions of patients in the SFC and FP arms who did not achieve any of the criteria for improvement were, respectively, 5% and 11% in nonsmokers, 10% and 10% in former smokers, and 10% and 10% in current smokers.

Back to Article Outline

Discussion 

The GOAL study found that approximately 65% of patients with persistent, mainly severe uncontrolled asthma achieved at least WC asthma after 1 year of treatment. The current analyses established that smoking status, treatment received (either FP or SFC), and presence or absence of baseline nighttime awakenings were the most important factors influencing response to treatment. Even if the patients were not or had not been heavy smokers, smoking history had the greatest negative influence on the probability of a patient achieving WC asthma. These results are in agreement with the findings of previous small, short-term (2–12 weeks) crossover studies, which suggested that, compared with patients who have never smoked, current smokers have diminished responses to corticosteroid therapy, as assessed on the basis of single outcomes, such as FEV₁, PEFR, PC₂₀, and sputum eosinophil count.⁴, ⁵, ⁶, ¹⁰

The findings of the current analysis add to our knowledge about the effects of smoking. The effect was evident on single selected outcomes but also on overall control status. Furthermore, previous studies have been of a shorter duration and therefore have not been able to address whether the reduced response to treatment represented a delayed effect that might in due course improve with prolonged treatment. Our data showed that a mere delay in effect is unlikely because the impairment in response caused by smoking was still present, even after 1 year of treatment. The effect of smoking status was similar across all strata in the study. This, together with the findings of short-term studies in patients with mild asthma,⁴, ⁵, ⁶, ¹⁰ suggests that the effects of smoking are applicable to all asthmatic patients. The current data also provide indirect evidence that smoking cessation could at least partially restore response to treatment because the deleterious effects in former smokers were less pronounced than those in current smokers.

Previous short-term studies assessing the effects of ICS in current smokers indicated that there would not be any statistically significant treatment effect during a 2- or 3-week treatment period.⁴, ⁵, ⁶ However, the current study showed improvements in both treatment groups over the course of the year, and asthma control in current smokers was markedly better at the end of the study than at study start. As with the primary analysis previously reported for the whole population, the clinical effect of SFC was greater than the effect of FP in both former and current smokers. Therefore it can be concluded that SFC treatment of asthma in patients who smoke is associated with clinically important benefits. However, this study was not placebo controlled, and therefore the conclusions for FP are less strong. Nevertheless, the marked clinical improvement from baseline seen in FP-treated patients suggests that long-term treatment with FP is also associated with important clinical benefits in smokers. Therefore it appears that active smoking could decrease the effect of treatment rather than abolish it, as previously suggested by some.⁶ Heavy smokers were not included in the study, therefore our data do not allow any conclusions for such patients.

Compared with nonsmokers with asthma, active smokers have been shown to have more severe asthma symptoms, more exacerbations, poorer asthma control, and an accelerated decrease in lung function.¹⁰, ¹¹, ¹² However, the mechanism underlying this and the reduced therapeutic response in smokers is not clear.¹⁰ Studies assessing the cellular effects have provided several suggestions of the effects of smoking on asthma pathophysiology, including alterations in airway inflammatory cell phenotypes, changes in glucocorticoid receptor (GR) α/β ratio, reduced histone deacetylase activity, reduced number of GRs, altered affinity of the ligand for the GR, reduced ability of the GR to bind with DNA or increased expression of inflammatory transcription factors that interfere with GR–DNA binding, and excessively remodeled airways.¹³, ¹⁴ However, a recent study found similar statistically significant effects of 160 μg of BDP on sputum eosinophils and eosinophilic cationic protein in smoking and nonsmoking asthmatic patients, even though the effect on other outcomes was reduced in smokers.¹⁰

Smoking induces an increase in urinary leukotriene E₄ levels in asthmatic patients, and therefore the study also assessed the effects of montelukast in smoking and nonsmoking asthmatic patients. The interpretation of data was difficult because montelukast had no statistically significant effects on any of these outcomes in nonsmokers, whereas it increased morning PEF in smokers.¹⁰ Therefore further studies are needed to elucidate the relative importance of the proposed mechanisms by which smoking affects response to treatment.

The finding that nocturnal awakenings in asthmatic patients before treatment increased the probability of not achieving WC asthma was surprising and, to our knowledge, not previously reported. The reason for this is not known. Nocturnal awakenings could merely be a reflection of more severe asthma, which is likely to respond less well to treatment.¹¹ Height was also predictive of not achieving WC asthma and might be related to the link between height and FEV₁.¹⁵ In the current analysis female patients were more likely to have WC asthma than male patients. This does not align with previous findings suggesting that female patients have more severe asthma and greater bronchial hyperresponsiveness than male patients.¹⁶, ¹⁷, ¹⁸, ¹⁹ Perhaps these sex differences reflect a higher susceptibility of female patients to the effects of smoking.¹⁶ Polymorphisms in the estrogen receptor α-gene might be associated with increased hyperresponsiveness and rapid loss of lung function in female patients with asthma.²⁰ The effect of ICS on lung function has also been found to be smaller in women than in men.²⁰ Nevertheless, it is still not clear whether women respond differently to therapy than men.¹⁶, ¹⁷, ¹⁸ The sex effect in the current analysis was only statistically significant in stratum 2 and overall, making the interpretation even more difficult. The reason for the sex effect in the GOAL study is not known, and it was not explained by differences in baseline severity, compliance, or smoking status.

A composite end point of several outcomes was used in the GOAL study, all of which required an improvement in the specified threshold control level over long periods (every day of the week and up to 7 of 8 weeks). Based on findings from earlier studies, it was obvious that such a degree of asthma control was more difficult to achieve than achieving improvements in a single outcome over a shorter period of time. Therefore for the purposes of assessing the response in the individual outcomes measures used in the study, criteria were developed for what we considered to be clinically significant improvements. The magnitude of these improvements was chosen based on the magnitude of the effects reported in randomized controlled studies that were generally accepted to have demonstrated significant effects of different asthma treatments. Of the criteria used, only the change in Asthma Quality of Life Questionnaire has been formally validated.²¹

By using these criteria, it was possible to determine that the majority of patients with NWC asthma benefited to some extent from the treatment, regardless of smoking status or stratum. The threshold limits for each parameter are arbitrary, and the difference in response rates should be interpreted with caution, as should the significance of apparently negative responses to all items. It should also be remembered that all patients with unavailable data in the GOAL study were classified as not improving (adding to the number of patients without improvement). The lack of a placebo arm complicates an exact assessment of the treatment effect; however, among patients with NWC asthma, SFC-treated patients tended to show more improvement than FP-treated patients, although the differences between the 2 treatment groups were less marked than in the whole population.

In conclusion, it is imperative for good asthma control that asthmatic patients stop smoking. In the GOAL study most patients who failed to achieve at least WC asthma status still demonstrated clinical improvements in several individual asthma outcomes. The clinical effects of SFC treatment are significantly greater than the effects of FP treatment in patients with asthma who smoke.

Back to Article Outline

We acknowledge the contribution of Homer Boushey as a former member of the GOAL Steering Committee. We also would like to pay tribute to Romain Pauwels and Anne Woolcock (both deceased) for their individual outstanding achievements in asthma research and for their respective contributions to the GOAL study as former members of the Steering Committee. We thank the investigators and staff at all study sites and at GlaxoSmithKline. We also thank Arthur Smyth Medina for writing assistance and manuscript management.

Back to Article Outline

References 

1. 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;170:836–844
   • View In Article
   • MEDLINE
   • CrossRef
2. Global Initiative for Asthma . Global Initiative for Asthma (GINA) Pocket guide for asthma management and prevention. Bethesda (MD): National Institutes of Health, National Heart, Lung, and Blood Institute; 1998;Publication no. 95-3659B
   • View In Article
3. National Institutes of Health . National Asthma Education and Prevention Program. Guidelines for the diagnosis and management of asthma: expert panel report 2. Bethesda (MD): National Institutes of Health, National Heart, Lung and Blood Institute; 1997;Publication no. 97-4051
   • View In Article
4. Chaudhuri R, Livingston E, McMahon AD, Thomson L, Borland W, Thomson NC. Cigarette smoking impairs the therapeutic response to oral corticosteroids in chronic asthma. Am J Respir Crit Care Med. 2003;168:1308–1311
   • View In Article
   • MEDLINE
   • CrossRef
5. Chalmers GW, Macleod KJ, Little SA, Thomson LJ, McSharry CP, Thomson NC. Influence of cigarette smoking on inhaled corticosteroid treatment in mild asthma. Thorax. 2002;57:226–230
   • View In Article
   • MEDLINE
   • CrossRef
6. Tomlinson JE, McMahon AD, Chaudhuri R, Thompson JM, Wood SF, Thomson NC. Efficacy of low and high dose inhaled corticosteroid in smokers versus non-smokers with mild asthma. Thorax. 2005;60:282–287
   • View In Article
   • MEDLINE
   • CrossRef
7. Jang AS, Lee JH, Park SW, Lee YM, Uh ST, Kim YH, et al. Factors influencing the responsiveness to inhaled glucocorticoids of patients with moderate-to-severe asthma. Chest. 2005;128:1140–1145
   • View In Article
   • MEDLINE
   • CrossRef
8. Hosmer DW, Lemeshow S. Applied logistic regression. New York: Wiley; 2000;
   • View In Article
9. CPMP/EWP/2863/99 Points to consider on adjustment for baseline covariates. European Agency for the Evaluation of Medicinal Products. Available at: http://www.emea.europa.eu/pdfs/human/ewp/286399en.pdf.Accessed June 1, 2007.
   • View In Article
10. Lazarus SC, Chinchilli VM, Rollings NJ, Boushey HA, Cherniack R, Craig TJ, et al. Smoking affects response to inhaled corticosteroids or leukotriene receptor antagonists in asthma. Am J Respir Crit Care Med. 2007;175:783–790
    • View In Article
    • MEDLINE
    • CrossRef
11. Schatz M, Zeiger RS, Vollmer WM, Mosen D, Cook EF. Determinants of future long-term asthma control. J Allergy Clin Immunol. 2006;118:1048–1053
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (126 KB)
    • CrossRef
12. Lange P, Parner J, Vestbo J, Schnohr P, Jensen G. A 15-year follow-up study of ventilatory function in adults with asthma. N Engl J Med. 1998;339:1194–1200
    • View In Article
    • MEDLINE
    • CrossRef
13. Ito K, Chung KF, Adcock IM. Update on glucocorticoid action and resistance. J Allergy Clin Immunol. 2006;117:522–543
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (584 KB)
    • CrossRef
14. Thomson NC, Spears M. The influence of smoking on the treatment response in patients with asthma. Curr Opin Allergy Clin Immunol. 2005;5:57–63
    • View In Article
    • MEDLINE
    • CrossRef
15. Global Strategy for Asthma Management and Prevention. Global Initiative for Asthma (GINA). Available at: http://www.ginasthma.org/Guidelineitem.asp??l1=2&l2=1&intId=60. Accessed January 9, 2007.
    • View In Article
16. Leynaert B, Bousquet J, Henry C, Liard R, Neukirch F. Is bronchial hyperresponsiveness more frequent in women than in men? A population-based study. Am J Respir Crit Care Med. 1997;156:1413–1420
    • View In Article
    • MEDLINE
17. Trigg CJ, Bennett JB, Tooley M, Sibbald B, D'Souza MF, Davies RJ. A general practice based survey of bronchial hyperresponsiveness and its relation to symptoms, sex, age, atopy, and smoking. Thorax. 1990;45:866–872
    • View In Article
    • MEDLINE
    • CrossRef
18. Britton J, Pavord I, Richards K, Knox A, Wisniewski A, Wahedna I, et al. Factors influencing the occurrence of airway hyperreactivity in the general population: the importance of atopy and airway calibre. Eur Respir J. 1994;7:881–887
    • View In Article
    • MEDLINE
19. Bottema RW, Reijmerink NE, Koppelman GH, Kerkhof M, Postma DS. Phenotype definition, age, and gender in the genetics of asthma and atopy. Immunol Allergy Clin North Am. 2005;25:621–639
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (313 KB)
    • CrossRef
20. Dijkstra A, Howard TD, Vonk JM, Ampleford EJ, Lange LA, Bleecker ER, et al. Estrogen receptor 1 polymorphisms are associated with airway hyperresponsiveness and lung function decline, particularly in female subjects with asthma. J Allergy Clin Immunol. 2006;117:604–611
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (191 KB)
    • CrossRef
21. Juniper EF, Guyatt GH, Epstein RS, Ferrie PJ, Jaeschke R, Hiller TK. Evaluation of impairment of health related quality of life in asthma: development of a questionnaire for use in clinical trials. Thorax. 1992;47:76–83
    • View In Article
    • MEDLINE
    • CrossRef