Dispensing of fluticasone propionate/salmeterol combination in the summer and asthma-related outcomes in the fall

Article Outline

• Abstract
• Methods
  • Study design
  • Study population
  • Analysis
  • Sensitivity analysis
• Results
  • Sensitivity analysis
• Discussion
• Acknowledgment
• References
• Copyright

Background

Asthma exacerbations occur year-round; however, peak asthma-related events occur in the fall and are frequently associated with viral respiratory infections.

Objective

To compare the rates of asthma-related emergency department (ED) visits and hospitalizations in the fall (September, October, November) between users and nonusers of fluticasone propionate plus salmeterol in a single inhaler (FSC) in the preceding summer.

Methods

This was a retrospective, observational study using health care claims from a large managed care database. Patients age 4 to 55 years with both a medical claim for asthma and a pharmacy claim for FSC were categorized into 3 age groups: children (4-11 years), adolescents (12-18 years), and adults (19-55 years).

Results

There were 201,973 observations of FSC dispensings and 184,143 observations without FSC. Across all age groups, summertime dispensings of FSC were associated with a significantly lower (P < .001) risk of an asthma-related ED visit (4-11 years: adjusted odds ratio [OR], 0.54, 95% CI, 0.49-0.60; 12-18 years: OR, 0.59, 95% CI, 0.54-0.64; 19-55 years: OR, 0.53, 95% CI, 0.51-0.55) or hospitalization (4-11 years: OR, 0.43, 95% CI, 0.35-0.54; 12-18 years: OR, 0.49, 95% CI, 0.40-0.60; 19-55 years: OR, 0.61, 95% CI, 0.57-0.65) in the subsequent fall. This protective effect persisted even for patients with fall dispensings of FSC. The risk of oral corticosteroid dispensing in the fall was also significantly reduced in all age groups.

Conclusion

Summertime dispensings of FSC were associated with a decreased risk of serious asthma-related outcomes in the subsequent fall. Continuous use of FSC before seasonal viral exposure may decrease seasonally related exacerbations.

Key words: Fluticasone propionate, salmeterol, combination, asthma, utilization, emergency department, hospitalization, exacerbations, drug holiday

Abbreviations used: ED, Emergency department, FSC, Fluticasone propionate plus salmeterol in a single inhaler, GEE, Generalized estimating equation, NNT, Number needed to treat, OCS, Oral corticosteroid, OR, Odds ratio, SABA, Short-acting β-agonist

Seasonal trends in asthma exacerbations are widely recognized, with peaks of hospitalizations and emergency department (ED) visits occurring in the fall months.¹, ², ³, ⁴, ⁵, ⁶ This peak in asthma-related events appears to be associated with an increase in viral respiratory infections, especially rhinovirus, and the start of the school year.³, ⁷, ⁸ Although studies have been conducted worldwide, the Ontario study has been one of the most detailed examinations of both the prevalence and potential casual factors of the fall asthma epidemic.⁷ In addition to viral respiratory infections, the study suggests that inadequate use of asthma controller medication, specifically inhaled corticosteroids, may be related to higher rates of asthma exacerbations in the fall. Using a case-control analysis, the study of the fall asthma epidemic found that the reported use of inhaled corticosteroids was significantly less frequent among children with an asthma-related ED visit compared with matched controls. Asthma controller medications were prescribed only half as often for the children with ED visits than controls.

The current National Asthma Education and Prevention Program guidelines characterize asthma as a chronic inflammatory disorder with recurrent exacerbations.⁹ Consequently, the guidelines advocate the daily use of long-term control medication as the most effective therapy for persistent asthma. Despite these recommendations, as-needed use of asthma controller therapy has been suggested as a potential alternative to regular therapy, particularly in children who experience viral-induced asthma attacks.¹⁰, ¹¹, ¹² However, as-needed use of controller therapy has not been adequately evaluated as an effective strategy for treating asthma.

Because asthma exacerbations are less frequent during the summer months, a summer holiday from controller medications has been adopted in clinical practice. Although a summer adjustment of asthma controller medications may be appropriate for some patients, a search of the literature fails to identify studies to support this common clinical practice. Several important issues need to be investigated properly, including the identification of the asthma patient appropriate for this strategy, the required duration of well controlled asthma before initiating this strategy, and the appropriate timing of restarting controller medication before the fall epidemic of asthma.

This study was designed to assess clinical outcomes for patients who take a summer holiday from controller medications compared with consistent, regular treatment for patients with persistent asthma. Specifically, our seasonal outcomes analysis compared asthma-related ED visits and hospitalizations in the fall between patients dispensed fluticasone propionate plus salmeterol in a single inhaler (FSC) in the preceding summer and those who were not dispensed FSC in the summer with and without FSC dispensing in the fall.

Back to Article Outline

Methods 

Study design 

This was a retrospective observational analysis using health care claims from December 1, 2002, to May 31, 2006, from a large managed care database that contains more than 45 US health plans representing 58.5 million covered lives. This proprietary, Health Insurance Portability Accountability Act–compliant research database contains integrated enrollment, laboratory, pharmacy, and medical claims data.

Study population 

The study population consisted of patients age 4 to 55 years with both a medical diagnosis claim for asthma (International Classification of Diseases, Ninth Revision: 493.XX) and a pharmacy claim for FSC in the months before the summer season of interest. Subjects were required to have at least 1 pharmacy claim for FSC in the year of observation and were excluded if they were dispensed any other inhaled corticosteroid during that period. In addition, patients had to be continuously enrolled during the summer (June, July, August) and fall (September, October, November) in the year of observation. Patients were categorized as children (4-11 years), adolescents (12-18 years), and adults (19-55 years).

Analysis 

The summertime possession cohort was defined as patients with ≥1 pharmacy claim for FSC in the months of June, July, or August. The nonpossession cohort was defined as subjects without a FSC pharmacy claim history in these summer months. Numbers of observations in the summer were calculated; because 4 different years were observed, subjects could give from 1 to 4 observations over the entire study period. Asthma-related ED and hospitalization events were assessed during the subsequent fall (September, October, November). Asthma-related ED and hospitalization rates for both cohorts were then calculated as a mean percent of observations with the event. In addition, we also assessed the dispensing of short-acting β-agonists (SABAs) and oral corticosteroids (OCSs) in the fall months in a similar fashion.

Patient characteristics during the baseline period (summer) were compared across cohorts and included age, sex, and the percent of patients with ≥1 ED visit, ≥1 hospitalization, ≥1 SABA dispensing, or ≥1 OCS dispensing in the summer.

Statistical comparisons and unadjusted relative risks between the absolute mean ED and hospitalization percentages were determined by using χ² tests to compare categorical variables across cohorts. Adjusted relative risks controlling for age, sex, summertime asthma-related ED, hospital visits, OCS use, and albuterol use were generated by using generalized estimating equations (GEEs), which take into account multiple observations. To aid clinical interpretation, a number needed to treat (NNT) was calculated for OCS, ED, and hospitalization avoided in the fall for each age group. NNT is the number needed to treat to avoid 1 event in the fall.

Sensitivity analysis 

Two sensitivity analyses were performed. The first compared the same outcomes in patients dispensed FSC in the summer and those not dispensed FSC in the summer who were also dispensed FSC in the fall to see whether the use of FSC in the fall attenuated the risk of asthma-related fall events. The second was a matched sample using propensity score analysis with replacement. Propensity matching is a technique that matches cohorts on the propensity to receive a particular medication, in this case the propensity for FSC to be dispensed in the summer. The propensity score method can incorporate a variety of baseline variables present in the claims data that may alter the selection of one therapy over another and creates study cohorts closely balanced on observable factors.¹³, ¹⁴, ¹⁵, ¹⁶, ¹⁷ Observable baseline characteristics used in the propensity score are reported in Table I. SAS Proprietary Software, Release 9.1 (SAS Institute Inc, Cary, NC), was used for all statistical analyses.

Table I. Demographics and baseline characteristics in the summer

	Dispensing of FSC in the summer	No dispensing of FSC in the summer
Age 4-11 y
No. of observations	20,084	23,690
Mean age (y)	8.9∗	8.7
Male (%)	63.5†	62.1
ED visit in summer (%)	3.3∗	2.2
Hospitalization in summer (%)	0.6∗	0.3
OCS use in summer (%)	5.1∗	2.4
SABA use in summer (%)	37.7∗	17.0
Age 12-18 y
No. of observations	27,328	31,108
Mean age (y)	14.7‡	14.8
Male (%)	56.8∗	53.9
ED visit in summer (%)	3.5∗	2.2
Hospitalization in summer (%)	0.7∗	0.4
OCS use in summer (%)	9.0∗	4.5
SABA use in summer (%)	47.5∗	21.6
Age 19-55 y
No. of observations	154,561	129,345
Mean age (y)	40.7∗	39.5
Male (%)	38.4∗	34.0
ED visit in summer (%)	3.5∗	2.3
Hospitalization in summer (%)	1.5	1.1
OCS use in summer (%)	14.1	8.3
SABA use in summer (%)	42.1∗	20.6

Back to Article Outline

Results 

Overall, asthma-related events were greatest in the fall months throughout the entire study period (Fig 1). A comparison of baseline characteristics for patients with and without FSC dispensing during the summer is presented in Table I. A total of 386,116 observations were evaluated, including 201,973 observations of summertime FSC dispensing and 184,143 observations without FSC dispensed. Patients of all age groups who were dispensed FSC during the summer were more symptomatic during the summer months than those without FSC possession. The baseline characteristics of subjects dispensed FSC in the summer and those not dispensed FSC in the summer differed. Patients dispensed FSC in the summer had more baseline claims for SABA and OCS (except for the cohort 19-55 years old) and were more likely to have visited the ED or be hospitalized for asthma in the summer than those without FSC in the summer (P < .001). The baseline characteristics of the propensity-matched sample are shown in Table II.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  Seasonal asthma–related ED visits and hospitalizations throughout the observation period.

Table II. Demographics and baseline characteristics in the summer after propensity matching

	Dispensing of FSC in the summer	No dispensing of FSC in the summer
Age 4-11 y
No. of observations	19,280	22,140
Mean age (y)	8.91	8.91
Male (%)	63.6	63.6
ED visit in summer (%)	3.2	3.1
Hospitalization in summer (%)	0.6∗	0.4
OCS use in summer (%)	5.1	5.0
SABA use in summer (%)	37.8	37.7
Age 12-18 y
No. of observations	26,414	29,476
Mean age (y)	14.7	14.7
Male (%)	56.8	56.7
ED visit in summer (%)	3.5	3.5
Hospitalization in summer (%)	0.7†	0.5
OCS use in summer (%)	9.0	9.0
SABA use in summer (%)	47.6	47.6
Age 19-55 y
No. of observations	147,757	119,723
Mean age (y)	40.8	40.8
Male (%)	38.5	38.4
ED visit in summer (%)	3.4	3.4
Hospitalization in summer (%)	1.5∗	1.4
OCS use in summer (%)	14.1	14.1
SABA use in summer (%)	42.1	42.1

Table III presents the results of the complete population. The dispensing of FSC in the summer was associated with a statistically significantly (P < .0001) lower risk of asthma-related ED visit or hospitalization in the subsequent fall for all age groups. For children age 4 to 11 years, the risks of an asthma-related ED visit (adjusted odds ratio [OR], 0.60; 95% CI, 0.54-0.67) and hospitalization (OR, 0.49; 95% CI, 0.39-0.61) were significantly reduced (P < .001) for patients dispensed FSC in the summer compared with those without a FSC claim. Similar results were observed for adolescents; the risk of an asthma-related ED visit (OR, 0.624; 95% CI, 0.59-0.70) and hospitalization (OR, 0.55; 95% CI, 0.45-0.67) was also significantly lower (P < .001) for summertime dispensing of FSC compared with those who did not receive FSC in the summer. The same pattern of results was noted for adults (ED visit: OR, 0.60, 95% CI, 0.57-0.63; hospitalization: OR, 0.68, 95% CI, 0.64-0.73). For each of the age groups, dispensing FSC in the summer was associated with a significantly lower (P < .001) risk of both OCS and SABA use in the fall compared with no FSC summertime claim. The NNT to prevent 1 severe exacerbation as defined by an inpatient hospitalization, ED visit, or OCS claim during the fall season from the unmatched sample is presented in Table IV.

Table III. Comparison of asthma-related outcomes in the fall

Table IV. NNT for asthma-related exacerbations by age group

Sensitivity analysis 

Because patients dispensed FSC in the summer were more likely to have FSC dispensed in the fall compared with summertime nondispensing, a sensitivity analysis was performed on only patients who received FSC in the fall. Fall asthma events were then compared on the basis of summertime dispensing of FSC in this group of patients who were dispensed FSC during the fall months. The results remained consistent with the overall findings demonstrating a statistically significant lower risk of asthma-related events observed in the fall that was associated with a history of FSC in the summer and not related to the dispensing of FSC in the fall. In fact, both the crude relative risk and adjusted OR of a fall asthma-related exacerbation or SABA use was higher in the cohort of subjects possessing FSC in the fall without the dispensing of FSC in the summer (Table V). An increase in asthma ED visits and hospitalizations in the fall compared with the summer appeared in all 3 age groups without FSC claims in the summer but not in the FSC summertime possession cohort (Fig 2, Fig 3).

Table V. Comparison of asthma-related outcomes in the fall (among fall users of FSC with or without summertime use of FSC)

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  Percent of patients with an asthma-related ED visit in the fall for patients with FSC dispensed in the summer and those not dispensed FSC. Yellow lines represent baseline (summer) rate.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 3. 

  Percent of patients with an asthma-related hospitalization in the fall for patients dispensed and not dispensed FSC in the summer. Yellow lines represent baseline (summer) rate.

Results from propensity matching provided an additional sensitivity analysis and support the findings from the base regression analysis (Table VI).

Table VI. Comparison of asthma-related outcomes in the fall in propensity-matched sample

∗P < .001 for χ² test.

Back to Article Outline

Discussion 

The results of our longitudinal database claims analysis demonstrate that the dispensing of an anti-inflammatory such as FSC in the summer was associated with a reducing the risk of serious asthma-related exacerbations in the subsequent fall. Summertime dispensing of FSC had lower rates of asthma-related ED visits or hospitalization in the fall compared with the cohort without summertime possession of FSC. Pharmacy claims for OCS and SABA in the fall were also lower among patients dispensed FSC in the summer. The results were observed for all age groups and were not altered by pharmacy claims for FSC in the fall months. Further, fall dispensing of FSC in the absence of summer dispensing did not afford this protection.

Not unexpectedly, the NNT in the summer with FSC to avoid 1 event in the fall was lowest for the avoidance of OCS and highest for avoidance of an asthma-related hospitalization. NNT is a function of the background rate of events, with rare occurrences usually having high NNTs. For example, 167 patients age 4 to 11 years would need to be treated with FSC to avoid 1 hospitalization, whereas 38 would need to be treated to avoid 1 course of OCS in the fall.

Patients without summertime dispensing of FSC had less baseline disease morbidity, as evidenced by lower rates of asthma-related ED visits or hospitalizations during the summer, than patients dispensed FSC in the summer. Paradoxically, patients who did not possess FSC during the summer had greater asthma disease morbidity in the fall, whether they started or did not start FSC in the fall. This observation suggests that patients who are weaned from controllers during the summer months may be at increased risk for asthma-related ED visits or hospitalization even if the controller medication is restarted in the fall. One hypothesis to support this observation is that epithelial integrity of the airway improves with controller treatment and that this may provide better protection for the patient with asthma when subsequently exposed to ubiquitous viral infections.¹⁸, ¹⁹ In addition, the benefits of treatment with a combination inhaled corticosteroid and long-acting bronchodilator may provide additional benefits to ciliary function, decreased vascularity, and reduction in inflammation.²⁰, ²¹, ²², ²³ Although further research is needed to demonstrate the need for year-round treatment in patients, the current recommendation for a summertime holiday from controller medication is based on clinical impression with little support from the literature. This too needs to be validated if it is to be included in any treatment algorithm. Our findings are of particular importance given the seasonal patterns of asthma. Increases in ED visits and hospital admissions for asthma occur with predictable regularity every fall,¹, ², ³, ⁴, ⁵ after a summertime decline in asthma medication use (data not shown). The results of our analysis suggest that the continued use of FSC throughout the summer may help mitigate asthma worsening in the fall.

Autumn peaks of asthma-related exacerbations have been well described.¹, ², ³, ⁴, ⁵ Several investigators have shown that the increase in ED visits and hospitalizations in early fall is closely associated with return to school after summer vacation.⁴, ⁷, ⁸ Using Canadian asthma hospital admission data from 1990 to 2002, Johnston et al⁸ applied mathematical modeling to estimate the exact timing and magnitude of September asthma hospitalization epidemics in children and adults. Peaks in asthma hospitalizations were estimated to occur 17.7 days after Labor Day in Canadian school-age children. Epidemics of lesser magnitude occurred later in preschool children and adults, which suggests that school-age children may be responsible for transmitting the agents causing exacerbations to younger and older family members.

Peaks in asthma exacerbations after school return have been linked to the coincident increases in viral infections, particularly rhinovirus, as well as inadequate asthma controller medicine. In a case-control study, respiratory viral infections were identified in 62% children with ED visits for asthma during September versus 41% of control children with comparable asthma severity (P = .011).⁷ Picornaviruses, of which more than 80% were rhinovirus, were detected in 52% of cases versus 29% of controls (P = .002). Despite comparable asthma severity, the ED cases experienced more frequent exacerbations than controls in the previous year. Importantly, asthma controller medications, particularly inhaled corticosteroids (ICS), were prescribed only half as often for the ED cases as for controls (ICS, 49% vs 85%; P < .0001). Furthermore, the number of inhaled corticosteroid prescriptions was lowest in August compared with other months. These findings suggest that the peak in asthma exacerbations after school return may by reduced by the use of controller medication in the months before children are at greatest risk of a viral-induced exacerbation.

Treatment guidelines advocate the continual, daily use of long-term control medication⁹; however, the intermittent use of asthma controllers occurs commonly in clinical practice. Intermittent use or summer drug holiday is a dosing strategy for asthma that is adopted by many patients. The summer drug holiday strategy suggests that restarting controller medication in the weeks before the fall epidemic may provide adequate protection against the risk of an asthma-related exacerbation. However, there are no studies in the literature to validate this strategy, nor are there any that provide evidence for how many weeks before viral exposure are needed to protect against the risk of an exacerbation.

Observational studies using commercial claims data have some limitations that are common to all analyses conducted with this type of data source. For example, there are limits to the degree to which claims data can accurately capture an individual's medical history, including the inability to elucidate asthma phenotypes. Although claims data are useful for understanding real-world patterns of use and cost, they are subject to possible coding errors. In addition, exposure estimates derived from claims data can be underestimated because these data do not account for physician samples or medication purchased with cash.

Because the data for our analysis were obtained from a managed care population, the results are primarily applicable to the treatment of asthma in the managed care setting. However, the plans used for the analysis are discounted fee-for-service, independent practice association network plans and not capitation or gate-keeper models. In addition, the plans include a wide geographic distribution across the United States and therefore allow for generalization to managed care populations on a national level. Finally, pharmacy claims can provide evidence of prescriptions filled but cannot supply information regarding whether a patient actually took medications, used all medication supplied, or hoarded medication.

Because of the observational nature of this study, without the benefit of clinical information and more historical information, asthma severity can be approximated only with the use of surrogates within this database. Confounding by indication or severity where only those subjects that need a medication receive it is a potential bias in most pharmacoepidemiology research such as this. This can lead to spurious findings and bias conclusions. This bias was mitigated by comparing subjects that were dispensed only 1 particular type of asthma treatment, FSC, and subjects were required to have a medical claim history for asthma, which mitigated some of the confounding that is related to treatment choice. Bias was also reduced by using statistical regression models that adjusted for differences in patient demographics and summertime asthma events, sensitivity analyses with stratification by dispensing of FSC in the fall, and propensity matching. This study does not address whether a similar reduction in fall asthma exacerbations could be achieved with other controller treatments such as inhaled corticosteroid alone.²⁴, ²⁵

The large sample of this study may make any difference between the groups statistically significant. However, the adjusted ORs for asthma-related events (ED or hospitalizations) were all ≤0.70, and 95% CIs were very tight around that estimate. This lower risk should be considered clinically meaningful, especially when patients dispensed FSC in the summer had no change in asthma-related events in the fall, whereas the cohort without dispensing had as much as a 3-fold increase in asthma-related hospitalizations. In addition, although no adjustments were made for multiple testing in this analysis (for the multiple relevant endpoints), we assert that the results are statistically valid given the very low P values found. The Bonferroni correction is a simple multiplicity adjustment method with no assumptions about data distributions or about the correlation structure among the endpoints. Applying the Bonferroni correction to our data would have resulted in an adjusted α of 0.05/(number of age groups × number of tests) = 0.05/12, or 0.004. All P values in this analysis were found to be less than .001. However, replication of this analysis in another population or dataset would be helpful in interpretation and could support these findings.

In conclusion, our longitudinal analysis demonstrates that the dispensing of FSC in the summer was associated with a decreased risk of asthma-related ED visits and hospitalizations in the subsequent fall. This reduction in risk was not observed in patients who were dispensed their medication in the fall but not during the summer months. Continued use of FSC or other anti-inflammatory medications in the summer may help to prevent the predictable worsening of asthma in the fall.

Back to Article Outline

We thank Cynthia Toso, PharmD, for assistance in preparing the manuscript.

Back to Article Outline

References 

1. Lin RY, Pitt TJ, Lou W, Yi Q. Asthma hospitalization patterns in young children relating to admission age, infection presence, sex, and race. Ann Allergy Asthma Immunol. 2007;98:139–145
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (964 KB)
   • CrossRef
2. Gergen PJ, Mitchell H, Lynn H. Understanding the seasonal pattern of childhood asthma: results from the National Cooperative Inner-City Asthma Study (NCICAS). J Pediatr. 2002;141:631–636
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (96 KB)
   • CrossRef
3. Sears MR, Johnston NW. Understanding the September asthma epidemic. J Allergy Clin Immunol. 2007;120:526–529
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (198 KB)
   • CrossRef
4. Silverman RA, Ito K, Stevenson L, Hastings HM. The relationship of fall school opening and emergency department asthma visits in a large metropolitan area. Arch Pediatr Adolesc Med. 2005;159:818–823
   • View In Article
   • MEDLINE
   • CrossRef
5. Fleming DM, Cross KW, Sunderland R, Ross AM. Comparison of the seasonal patterns of asthma identified in general practitioner episodes, hospital admissions, and deaths. Thorax. 2000;55:662–665
   • View In Article
   • MEDLINE
   • CrossRef
6. Pendergraft TB, Stanford RH, Beasley R, Stempel DA, McLaughlin T. Seasonal variation in asthma-related hospital and intensive care unit admissions. J Asthma. 2005;42:265–271
   • View In Article
   • MEDLINE
   • CrossRef
7. Johnston NW, Johnston SL, Duncan JM, Greene JM, Kebadze T, Roy M, et al. The September epidemic of asthma exacerbations in children: a search for etiology. J Allergy Clin Immunol. 2005;115:132–138
   • View In Article
8. Johnston NW, Johnston SL, Norman GR, Dai J, Sears MR. The September epidemic of asthma hospitalization: school children as disease vectors. J Allergy Clin Immunol. 2006;117:557–562
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (294 KB)
   • CrossRef
9. Expert Panel report 3: guidelines for the diagnosis and management of asthma. Full report 2007. National Heart, Lung, and Blood Institute, National Institutes of Health, National Heart, Lung, and Blood Institute. Available at: http://www.nhlbi.nih.gov/guidelines/asthma/asthgdln.htm. Accessed August 21, 2008.
   • View In Article
10. Connett G, Lenney W. Prevention of viral induced asthma attacks using inhaled budesonide. Arch Dis Child. 1993;68:85–87
    • View In Article
    • CrossRef
11. Svedmyr J, Nyberg E, Thunqvist P, Asbrink-Nilsson E, Hedlin G. Prophylactic intermittent treatment with inhaled corticosteroids of asthma exacerbations due to airway infections in toddlers. Acta Paediatr. 1999;88:42–47
    • View In Article
    • MEDLINE
    • CrossRef
12. Doull IJM. Limitations of maintenance therapy for viral respiratory infection-induced asthma. J Pediatr. 2003;142:S21–S25
    • View In Article
    • Abstract
    • Full-Text PDF (77 KB)
    • CrossRef
13. Braitman LE, Rosenbaum PR. Rare outcomes, common treatments: analytic strategies using propensity scores. Ann Intern Med. 2002;137:693–695
    • View In Article
14. Cepeda MS, Boston R, Farrar JT, Strom BL. Comparison of logistic regression versus propensity score when the number of events is low and there are multiple confounders. Am J Epidemiol. 2003;158:280–287
    • View In Article
    • MEDLINE
    • CrossRef
15. D'Agostino RB. Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group. Stat Med. 1998;17:2265–2281
    • View In Article
    • MEDLINE
    • CrossRef
16. Joffe MM, Rosenbaum PR. Invited commentary: propensity scores. Am J Epidemiol. 1999;150:327–333
    • View In Article
    • MEDLINE
17. Stone RA, Obrosky DS, Singer DE, Kapoor WN, Fine MJ. Propensity score adjustment for pretreatment differences between hospitalized and ambulatory patients with community-acquired pneumonia. Pneumonia Patient Outcomes Research Team (PORT) Investigators. Med Care. 1995;33:AS56–AS66
    • View In Article
    • MEDLINE
18. Laitinen LA, Laitinen A, Haahtela T. A comparative study of the effects of an inhaled corticosteroid, budesonide, and a beta 2-agonist, terbutaline, on airway inflammation in newly diagnosed asthma: a randomized, double-blind, parallel-group controlled trial. J Allergy Clin Immunol. 1992;90:32–42
    • View In Article
    • MEDLINE
    • CrossRef
19. Chanez P, Bourdin A, Vachier I, Godard P, Bousquet J, Vignola AM. Effects of inhaled corticosteroids on pathology in asthma and chronic obstructive pulmonary disease. Proc Am Thorac Soc. 2004;1:184–190
    • View In Article
    • MEDLINE
    • CrossRef
20. Orsida BE, Ward C, Li X, Bish R, Wilson JW, Thien F, et al. Effect of a long-acting beta2-agonist over three months on airway wall vascular remodeling in asthma. Am J Respir Crit Care Med. 2001;164:117–121
    • View In Article
    • MEDLINE
21. Wallin A, Sue-Chu M, Bjermer L, Ward J, Sandström T, Lindberg A, et al. Effect of inhaled fluticasone with and without salmeterol on airway inflammation in asthma. J Allergy Clin Immunol. 2003;112:72–78
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (164 KB)
    • CrossRef
22. Dowling RB, Johnson M, Cole PJ, Wilson R. Effect of fluticasone propionate and salmeterol on Pseudomonas aeruginosa infection of the respiratory mucosa in vitro. Eur Respir J. 1999;14:363–369
    • View In Article
    • MEDLINE
    • CrossRef
23. Nelson HS, Chapman KR, Pyke SD, Johnson M, Pritchard JN. Enhanced synergy between fluticasone propionate and salmeterol inhaled from a single inhaler versus separate inhalers. J Allergy Clin Immunol. 2003;112:29–36
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (156 KB)
    • CrossRef
24. Bateman ED, Boushey HA, Bousquet J, Busse WW, Clark TJH, Pauwels RA, et al. Can guideline-defined asthma control be achieved? the gaining optimal asthma control study. Am J Respir Crit Care Med. 2004;170:836–855
    • View In Article
    • MEDLINE
    • CrossRef
25. Pauwels RA, Pedersen S, Busse WW, Tan WC, Chen YZ, Ohlsson SV, et al. Early intervention with budesonide in mild persistent asthma: a randomized, double-blind trial. Lancet. 2003;361:1071–1076
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (101 KB)
    • CrossRef