If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The objective of this analysis was to investigate whether patients with severe or difficult-to-treat asthma who experienced recent severe asthma exacerbations are at increased risk of future asthma exacerbations.
Methods
We conducted a 1.5-year prospective analysis of 2780 patients ⩾12 years of age from The Epidemiology and Natural History of Asthma: Outcomes and Treatment Regimens study. Severe exacerbations were defined as either an asthma-related emergency department visit or night of hospitalization in the 3 months prior to study visit; a secondary analysis assessed prior steroid bursts as an independent predictor and outcome. Potential confounding was assessed by statistical adjustment for demographic and clinical factors, as well as asthma severity and asthma control.
Results
Compared with patients without a recent severe exacerbation, patients with a recent exacerbation were at increased risk of future exacerbation (odds ratio=6.33; 95% CI 4.57, 8.76), even after adjustment for demographics and clinical factors (odds ratio=3.77; 95% CI 2.62, 5.43), asthma severity (physician-assessed: odds ratio=5.62; 95% CI 4.03, 7.83), National Asthma Education and Prevention Program (odds ratio=5.07; 95% CI 3.62, 7.11), Global Initiative for Asthma (odds ratio=5.32; 95% CI 3.80, 7.47), and asthma control (odds ratio=3.90; 95% CI 2.77, 5.50).
Conclusion
This analysis suggests that recent severe asthma exacerbations are a strong independent factor predicting future exacerbations and, as such, should be considered as part of the clinical assessment of patients with severe or difficult-to-treat asthma.
An important factor in asthma management is the reduction of exacerbations that require intensive healthcare use. Identification of patients at greatest risk for future exacerbations would help focus available healthcare resources, improve prevention strategies, and reduce morbidity of asthma.
Studies of asthma populations have shown that patients with a history of asthma exacerbations are at significant risk for future exacerbations.
Prospective multicenter study of relapse following treatment for acute asthma among adults presenting to the emergency department. MARC investigators. Multicenter Asthma Research Collaboration.
Prospective multicenter study of relapse following treatment for acute asthma among adults presenting to the emergency department. MARC investigators. Multicenter Asthma Research Collaboration.
have been identified in several asthma studies as contributing factors to this relationship. It has not been established, however, whether the association between recent exacerbations and future exacerbations persists after controlling for guideline-assessed asthma severity, asthma control, and clinical and demographic factors in patients with severe or difficult-to-treat asthma.
The Epidemiology and Natural History of Asthma: Outcomes and Treatment Regimens (TENOR) is a 3-year, multicenter, observational study of the natural history, treatment regimens, and outcomes of subjects with severe or difficult-to-treat asthma. The objective of this analysis was to investigate whether patients with severe or difficult-to-treat asthma who experienced recent asthma exacerbations are at increased risk of future asthma exacerbations, independent of known risk factors, as well as measures of severity and control. This analysis adjusted for demographic and clinical risk factors, asthma severity (National Asthma Education and Prevention Program (NAEPP),
National Asthma Education and Prevention Program Expert Panel Report: guidelines for the diagnosis and management of asthma update on selected topics-2002.
Our hypothesis was that recent exacerbations would persist in being an important indicator of future exacerbations, despite adjustment for severity and other potential risk factors.
Methods
Study design and population
TENOR is a prospective, observational, multicenter, 3-year study of patients in the United States with severe or difficult-to-treat physician-diagnosed asthma. Complete details of TENOR study methods have been previously described.
Design and baseline characteristics of the epidemiology and natural history of asthma: outcomes and treatment regimens (TENOR) study: a large cohort of patients with severe or difficult-to-treat asthma.
Briefly, patients were eligible for TENOR if they were categorized by physicians as having severe or difficult-to-treat asthma; patients with mild or moderate asthma were eligible for enrollment if their physicians considered their asthma difficult-to-treat and they met the additional inclusion and exclusion criteria. Patients were excluded from TENOR if they were heavy smokers (⩾30 pack-years) and if they had cystic fibrosis. Patients were selected for the TENOR study if they had received care for at least 1 year from their current provider and if they demonstrated high healthcare or medication usage in the 12 months prior to study entry.
Design and baseline characteristics of the epidemiology and natural history of asthma: outcomes and treatment regimens (TENOR) study: a large cohort of patients with severe or difficult-to-treat asthma.
High healthcare utilization (HCU) was defined as 2 or more unscheduled care visits for asthma or 2 or more oral steroid bursts. High medication use was defined at the time of enrollment as requiring 3 or more medications to control asthma or requiring long-term, daily high doses of inhaled steroids or use of 5mg/d or more of oral prednisone.
Data collection
Data utilized for this analysis was collected at the 12-month visit and subsequent 6-month visits. Demographic data collected in TENOR included age, sex, race/ethnicity. Clinical data included body mass index (BMI), spirometry measurements, as well as histories of comorbid conditions, smoking, and allergic/non-allergic triggers collected by the study site physician and/or study coordinator. Current actual asthma control medication use, including asthma control and quick-relief medications were collected by study coordinator interview. To collect medication data as accurately and consistently as possible, patients were instructed to bring their medications to the interview. However, medication use was not verified against prescriptions. Controller medications included inhaled corticosteroids, long-acting beta-agonists, leukotriene modifiers, methylxanthines, and cromolyn sodium or nedocromil. Quick-relief medications included short-acting beta-agonists and anticholinergics. Systemic corticosteroid use was also collected. Patient-reported information on asthma-related HCU included emergency department (ED) visits, overnight hospitalizations, corticosteroid bursts or ‘steroid bursts’ (defined as a short-term increase in corticosteroid therapy to treat an exacerbation of symptoms), and unscheduled office visits/contacts to physician during the previous 3 months; confirmation with patient medical records was not done.
Asthma severity assessment
Asthma severity was assessed using three different methods. Physicians were instructed to use their subjective clinical opinion to categorize each patient as having either mild, moderate, or severe asthma. In addition to assessing each subject's asthma severity, physicians reported whether their patient's asthma was considered difficult-to-treat based on specified parameters (i.e., complex treatment regimen, multiple drugs required, unable to avoid triggers, frequent exacerbations, severe exacerbations and/or unresponsive to therapy). Asthma severity was also derived according to NAEPP and GINA guidelines using patient responses to questions relating to asthma symptoms from the patient-reported Mini Asthma Quality of Life Questionnaire (Mini AQLQ)
was used to evaluate patient-reported level of asthma control.
Asthma exacerbations: predictor and outcomes
Recent severe exacerbations as a predictor were defined as the composite outcome of an ‘asthma-related ED visit or night of hospitalization in the 3 months prior to the study visit.’ Future severe exacerbations (FSE) as outcome at 18-month follow-up were defined as an ‘asthma-related ED visit or night of hospitalization in the 3 months prior to the study visit, or an asthma-related death that occurred prior to the scheduled study visit.’ These definitions were chosen because these serious events would likely result in medical encounters, therefore minimizing incomplete or misclassified data. There were few deaths in TENOR; therefore FSE is mostly a composite of hospitalization and ED visits.
Secondly, ‘any steroid burst in the 3 months prior to the study visit’ assessed at baseline and 18-month follow-up was used as a more sensitive measure of exacerbation as predictor and outcome, respectively. Steroid bursts were defined as a short-term increase in steroid therapy to treat an exacerbation of asthma symptoms. This definition was used only for the secondary analysis.
Statistical analysis
The present analysis includes patients ⩾12 years of age who had both a 12-month follow-up assessment and at least 1 subsequent follow-up visit. In order to avoid potential over-reporting of HCU due to the TENOR inclusion criteria, the 12-month assessment was used at baseline for this analysis and the TENOR 30-month assessment is used for this study's 1.5-year (18-month) follow-up outcome. Use of a single follow-up visit allowed outcomes to be defined as binary variables that could be modeled using logistic regression. For patients without an 18-month visit, data was imputed using the nearest available post-baseline follow-up visit.
Recent exacerbations, the three asthma severity measures and asthma control were used in univariable and bivariable models of the outcomes. In addition, a set of candidate predictor variables were pre-specified from demographic and clinical variables collected at baseline. Candidate variables included age; sex; race; BMI (kg/m2); duration of asthma; smoking status; history of chronic obstructive pulmonary disease (COPD), emphysema, and chronic bronchitis; post-bronchodilator forced expiratory volume in 1s (FEV1), forced expiratory vital capacity (FVC) ratio; post-bronchodilator % predicted FVC; allergic rhinitis; count of allergic asthma triggers (i.e., pollen, dust, mold, pets); and the count of non-allergic asthma triggers (i.e., cold and sinus infections). This set of candidate predictors was evaluated using stepwise selection and the reduced subset was then used in a multivariable model. Interactions between recent exacerbations and seasonality were also assessed, but no interaction was detected; therefore, these results are not presented.
Confirmatory analyses were conducted using generalized estimating equations (GEE) repeated measures logistic regression model. The GEE method employs all available follow-up data and accounts for multiple correlations.
With GEE, the analysis utilized all 4 of the patients’ possible follow-up assessments so that the outcome covered 6–24 months of follow-up rather than only the 18-month outcome. The GEE results were very similar to the logistic regression results and are therefore not presented.
Odds ratios (ORs) were computed first using recent exacerbations as the only predictor. Adjustment for asthma severity was accomplished using a set of multivariable models, each of which included RSE and a single severity measure. Adjustment for demographics and clinical data included the previously identified covariates in the logistic regression. Asthma-related medications were not considered as a separate model adjustment because they are a component of the GINA classification. Primary results are based on adjustments for the collection of variables in the stepwise model or one of the individual measures of asthma control or severity. Adjustment for all of measures in a single statistical model may create issues of colinearity; however, for completeness, results from the single statistical model have been reported. Where measures of precision are reported,
95% confidence intervals were used and percentages were computed as a function of non-missing data. All analyses were conducted using SAS version 9.1.
Results
Demographics by RSE
Of the entire TENOR population, 2780 patients met the criteria for this analysis, including 244 (8.8%) who had an RSE. Among these 244 patients, 155 had a prior ED visit, 17 had a hospitalization, and 72 had both an ED visit and hospitalization. There were 2536 patients with no recent severe exacerbations (NRSE). Nearly 700 patients () had a recent steroid burst.
Patients in the RSE group were more likely to be young adults (18% vs. 11% in the NRSE group) (Table 1). The gender distribution in this analysis reflected the entire TENOR population, where 71% of patients ⩾18 years of age were female.
Patients in the RSE cohort were also more likely to be black (30%; ), overweight (49% with a BMI ⩾30; ), and have a history of bronchitis (42%; ) compared with only 11%, 39%, and 32%, respectively, in the NRSE group. Patients with RSE were more likely to have multiple allergic triggers (57% with ⩾2; ) and multiple non-allergic triggers (49% with ⩾2; ) compared with patients in the NRSE group (44% and 26%), respectively.
Table 1Baseline TENOR patient demographics and characteristics.
More patients in the RSE group had severe asthma by all measures compared with patients in the NRSE group. Patients in the RSE cohort had more asthma control problems than the NRSE patients (Table 2).
Table 2Baseline TENOR measures of asthma severity and asthma control.
Assessment Method
NRSE, n (%)
RSE, n (%)
Total patients ()
2536
244
Physician-assessed severity
Mild
182 (7.2)
8 (3.3)
Moderate
1346 (53.2)
83 (34.2)
Severe
1002 (39.6)
152 (62.6)
NAEPP severity
Mild
1025 (41.5)
43 (18.0)
Moderate
689 (27.9)
61 (25.5)
Severe
756 (30.6)
135 (56.5)
GINA severity
Mild
145 (5.9)
2 (0.8)
Moderate
1128 (46.2)
55 (23.1)
Severe
1170 (47.9)
181 (76.1)
ATAQ control index
0
835 (34.2)
7 (2.9)
1
630 (25.8)
30 (12.6)
⩾2
979 (40.1)
201 (84.5)
Recent steroid burst
679 (26.8)
191 (78.6)
NRSE=no recent severe exacerbation; RSE=recent severe exacerbation; NAEPP=National Asthma Education and Prevention Program; GINA=Global Initiative for Asthma; ATAQ=Asthma Therapy Assessment Questionnaire.
P<0.0001 for patients experiencing an exacerbation vs. not experiencing an exacerbation for all assessment methods used.
The odds of RSE predicting FSE were over 6-fold (OR=6.33; 95% CI 4.57, 8.76) (Fig. 1a). After adjustment for asthma severity by physician-assessment, there was over a 5-fold risk of FSE (OR=5.62; 95% CI 4.03, 7.83). When adjusted by guideline assessment (OR=5.07; 95% CI 3.62, 7.11) for NAEPP and (OR=5.32; 95% CI 3.80, 7.47) for GINA. Adjustment for asthma control resulted in an OR of 3.90 (95% CI 2.77, 5.50) and adjustment for demographic and clinical characteristics produced an OR of 3.77 (95% CI 2.62, 5.43). Although many of the components of severity are captured in the adjustment for demographic and clinical characteristics, a model adjusting for asthma control, all three measures of severity, and demographic and clinical characteristics produced an OR of 3.09 (95% CI 2.35, 4.06).
Figure 1Odds of future exacerbations associated with recent exacerbations, adjusted for demographics, asthma severity, and asthma control. (a) Odds of future severe exacerbation associated with recent severe exacerbations, (b) Odds of future steroid burst associated with recent steroid bursts. x-Axis is on logarithmic scale.
When steroid bursts were defined as predictor and outcome, patients with a recent exacerbation were also more likely to report a future exacerbation (unadjusted OR=3.91; 95% CI 3.28, 4.67) (Fig. 1b). The odds of having a future steroid burst were also statistically significant () after adjusting for severity classification, ATAQ control index and/or demographic and clinical characteristics. Adjustment for all variables in a single model led to an OR of 2.99 (95% CI 2.57, 3.47)
Table 3a, Table 3b show the relationship between recent exacerbations and future exacerbations after adjustment for level of asthma severity. Asthma severity classification of severe vs. moderate nearly doubled the odds of a future severe exacerbation, regardless of the classification mechanism. Patients with ⩾2 asthma control problems were also at increased risk (OR=2.00, 95% CI 1.37, 2.92; ). A similar pattern was seen when recent steroid bursts were assessed as predicting future steroid bursts (Table 3b).
Table 3aOdds of future exacerbations associated with recent exacerbations adjusted for asthma severity assessment and asthma control. Predictors of future severe exacerbations (FSE).
Future severe exacerbation=the composite outcome of an ‘asthma-related ED visit’ or ‘night of hospitalization’ in the 3 months prior to the 18-month study visit, or an asthma-related death that occurred prior to the scheduled 18-month study visit; NAEPP=National Asthma Education and Prevention Program; GINA=Global Initiative for Asthma; ATAQ=Asthma Therapy Assessment Questionnaire.
Predictors
OR
95% CI
Unadjusted
Recent exacerbations
6.33
(4.57, 8.76)
Adjusted
Physician-assessed severity
Recent exacerbations
5.62
(4.03, 7.83)
Severe vs. moderate asthma
1.70
(1.26, 2.29)
NAEPP guidelines severity
Recent exacerbations
5.07
(3.62, 7.11)
Severe vs. moderate asthma
1.81
(1.26, 2.59)
GINA guidelines severity
Recent exacerbations
5.32
(3.80, 7.47)
Severe vs. moderate asthma
1.87
(1.35, 2.59)
ATAQ control index
Recent exacerbations
3.90
(2.77, 5.50)
⩾2 vs. 1 ATAQ control problems
2.00
(1.37, 2.92)
Future severe exacerbation=the composite outcome of an ‘asthma-related ED visit’ or ‘night of hospitalization’ in the 3 months prior to the 18-month study visit, or an asthma-related death that occurred prior to the scheduled 18-month study visit; NAEPP=National Asthma Education and Prevention Program; GINA=Global Initiative for Asthma; ATAQ=Asthma Therapy Assessment Questionnaire.
Future steroid burst=‘Any steroid burst in the 3 months prior to the 18 month study visit.’; NAEPP=National Asthma Education and Prevention Program; GINA=Global Initiative for Asthma; ATAQ=Asthma Therapy Assessment Questionnaire.
Predictors
OR
95% CI
Unadjusted
Recent exacerbations
3.91
(3.28, 4.67)
Adjusted
Physician-assessed severity
Recent exacerbations
3.47
(2.90, 4.16)
Severe vs. moderate asthma
1.87
(1.56, 2.25)
NAEPP guidelines severity
Recent exacerbations
3.58
(2.98, 4.30)
Severe vs. moderate asthma
1.43
(1.15, 1.79)
GINA guidelines severity
Recent exacerbations
3.62
(3.01, 4.37)
Severe vs. moderate asthma
1.49
(1.23, 1.81)
ATAQ control index
Recent exacerbations
3.23
(2.67, 3.90)
⩾2 vs. 1 ATAQ control problems
1.48
(1.18, 1.86)
† Future steroid burst=‘Any steroid burst in the 3 months prior to the 18 month study visit.’; NAEPP=National Asthma Education and Prevention Program; GINA=Global Initiative for Asthma; ATAQ=Asthma Therapy Assessment Questionnaire.
Multivariate analysis found that RSE remained significantly associated with FSE after additional covariate adjustment. Patients with BMI >30, non-allergic or allergic triggers were at increased risk of FSE while lower FVC, white race/ethnicity or age >55 years were protective factors of risk. Similarly, recent steroid bursts were significantly associated with future steroid bursts after multivariate adjustment. Having a history of COPD, non-allergic triggers or being a young adult increased risk of future exacerbations while lower FEV1/FVC ratio or white race/ethnicity were protective (Table 4).
Table 4Odds of future exacerbations using multivariate analysis adjusted for demographic and clinical factors.
Risk factors
OR
95% CI
Future severe exacerbations
Recent severe exacerbation
3.77
(2.62, 5.43)
BMI >30
1.67
(1.21, 2.31)
Non-allergic triggers
1.44
(1.15, 1.80)
Allergic triggers
1.14
(1.01, 1.29)
Per 10% predicted FVC
0.87
(0.80, 0.95)
Race: white vs. non-white
0.63
(0.45, 0.89)
Age: >55 vs. 12–18
0.57
(0.34, 0.94)
Future steroid burst
Recent steroid burst
3.28
(2.72, 3.96)
COPD
1.67
(1.19, 2.34)
Non-allergic triggers
1.57
(1.38, 1.79)
Age: 19–35 vs. 12–18
1.42
(1.00, 2.01)
Per 10% predicted FEV1/FVC ratio
0.94
(0.89, 0.99)
Race: white vs. non-white
0.77
(0.61, 0.97)
BMI=body mass index; FVC=forced vital capacity; COPD=chronic obstructive pulmonary disease; FEV1=forced expiratory volume in 1s.
This analysis has found that recent severe asthma exacerbations are strongly associated with future exacerbations, most notably demonstrated by an OR of 6.33 (95% CI 4.57, 8.76) for RSE predicting FSE. When steroid bursts were used to define exacerbation and after risk factor adjustment, the association persisted. This implies that recent severe exacerbations are a strong independent factor predicting future exacerbations and should be considered when clinically evaluating patients with severe or difficult-to-treat asthma. For appropriate disease management, it is imperative to understand the factors contributing to recurrent and intense HCU in “frequent flier”
Prospective multicenter study of relapse following treatment for acute asthma among adults presenting to the emergency department. MARC investigators. Multicenter Asthma Research Collaboration.
these results expand upon previous research by using prospectively evaluated data and including more than a year of patient follow-up in a unique cohort of patients with severe and difficult-to-treat disease. In addition, our study used a large asthma patient sample and sizeable numbers of clinical predictor and outcome variables, including measures of asthma severity and asthma control. These aspects of our study allowed for a thorough and robust examination of risk factors for future exacerbations in severe or difficult-to-treat asthma. The current analysis also controlled for important potential confounding variables, such as asthma severity and asthma control, which were not previously addressed in other studies.
While a lack of agreement between asthma severity classification by physicians’ assessment, NAEPP, and GINA criteria exists, it has been reported that a relationship exists between recent HCU and asthma severity, by several measures.
We found recent exacerbations were predictive of future exacerbations after adjustment for asthma severity by three different asthma severity classification methods and that patients with severe asthma were at greatest risk. This finding suggests that recent exacerbations are not adequately represented in asthma severity classifications, even when those classifications incorporate treatment as is the case with the GINA guidelines. However, when asthma control was included in the model, the odds ratio (OR=3.90; 95% CI 2.77, 5.50) was lower than in the asthma severity models. This suggests that asthma control better reflects risk of future asthma exacerbations than measures of asthma severity. Therefore, patient-reported asthma control may be an intermediate concept between asthma severity assessment and recent exacerbations.
Our findings are consistent with results from other studies which identified certain demographic and clinical risk factors associated with future exacerbations. Specifically, patients with BMI greater than 30, non-allergic or allergic triggers, non-white race/ethnicity, lower % predicted FVC or history of COPD were at greater risk of future exacerbations compared with patients without these characteristics. The FVC finding may be related to air trapping in the context of an obstructive ventilatory deficit or a proxy for restrictive lung physiology associated with obesity, which was relatively common in our study group. The role of obesity in asthma may have mechanical, inflammatory, and hormonal influences
; however, our analysis findings could be related to steroid use, which also was prominent in the TENOR population. Older patients (aged >55) were identified as being at lowest risk of FSE while young adults (ages 19–35) were found to be at higher risk of future steroid bursts compared with adolescents (ages 12–18). These mixed age-related findings may be explained by lifestyle differences, including use of medications and healthcare resources, or could be due to biological differences between these age groups. Additionally, recent research
Study of modifiable risk factors for asthma exacerbations: virus infection and allergen exposure increase the risk of asthma hospital admissions in children.
has found that, independently, viral illness and allergen exposure did not increase the likelihood of asthma-related hospitalizations in children; however, in combination, these factors did increase hospitalization risk. While our analysis found no association between seasonality and RSE, future TENOR analyses may help clarify the role of allergic triggers in the relationship between RSE and FSE in patients with severe or difficult-to-treat asthma.
This study has several limitations. Steroid bursts as predictor identified more recent exacerbations () compared with severe exacerbations (); however, steroid bursts may be a less-specific measure of exacerbations compared with RSE and may be subject to recall bias. Also, the primary and secondary analysis predictors and outcomes are not mutually exclusive as severe exacerbations occurring under supervision of appropriate asthma medical care likely resulted in steroid bursts.
In TENOR, there may be over-reporting of exacerbations by patients at earlier assessment visits than at later visits; therefore, we used the 12-month visit as baseline. Additionally, use of a 3-month patient recall period for HCU has been shown to be a more precise patient recall period than longer periods.
Also, it is not clear which clinical parameters physicians used in classifying patient severity. Furthermore, NAEPP and/or GINA criteria-based severity assessments were derived retrospectively, from patient-reported outcome instruments that were not designed to evaluate asthma severity and may not fully represent asthma severity.
Our results suggest that recent severe exacerbations should be given greater consideration in the clinical evaluation of patients with severe or difficult-to-treat asthma. Furthermore, after consideration of other factors, including asthma severity classification and control, recent asthma exacerbations were found to contribute strongly and independently to the prediction of a future exacerbation. This suggests that the pathobiologic events driving these exacerbations may differ from those contributing to other characteristics of asthma. Physicians should be aware of their patients’ recent exacerbation status and respond to the occurrence to reduce the risk of repeat episodes. Asthma treatment guidelines should consider recent exacerbations in determination of asthma severity classification and recommendations for treatment intervention to prevent future events.
Prospective multicenter study of relapse following treatment for acute asthma among adults presenting to the emergency department. MARC investigators. Multicenter Asthma Research Collaboration.
Design and baseline characteristics of the epidemiology and natural history of asthma: outcomes and treatment regimens (TENOR) study: a large cohort of patients with severe or difficult-to-treat asthma.
Study of modifiable risk factors for asthma exacerbations: virus infection and allergen exposure increase the risk of asthma hospital admissions in children.
For entire category.
00356-8&id=gr1.jpg){kind=link}