BRIEF COMMUNICATION Body Mass Index and Risk of Hospitalization among Adults Presenting with Asthma Exacerbation to the Emergency Department Kohei Hasegawa1, Yusuke Tsugawa2, Bernard L. Lopez3, Howard A. Smithline4, Ashley F. Sullivan1, and Carlos A. Camargo, Jr.1 1
Department of Emergency Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts; 2Harvard Interfaculty Initiative in Health Policy, Cambridge, Massachusetts; 3Department of Emergency Medicine, Thomas Jefferson University Hospital, Philadelphia, Pennsylvania; and 4Department of Emergency Medicine, Baystate Medical Center, Springfield, Massachusetts
Abstract Rationale: Studies have linked obesity to incident asthma and worse chronic severity/control. However, the relationship between obesity and acute asthma morbidity remains unclear. Objectives: To determine whether obese adults presenting to the emergency department (ED) with asthma exacerbation are at higher risk of hospitalization compared with normal-weight adults. Methods: Multicenter chart review study of 48 EDs across 23 U.S. states. We identified ED patients aged 18 to 54 years with asthma exacerbation during 2011 to 2012. Primary outcome was hospitalization. Measurements and Main Results: The analytic cohort comprised 1,227 patients. Of these, 323 patients (27%) were overweight (body mass index [BMI], 25–29.9 kg/m2), and 607 (50%) were obese (BMI > 30 kg/m2). Among the 607 obese patients, 364 patients (60%) were severely obese (BMI > 35 kg/m2).
Several markers of chronic severity/control of asthma and acute severity did not differ across BMI groups. By contrast, compared with normal-weight patients, the risk of hospitalization was higher in patients who were overweight (11 vs. 18%; odds ratio [OR], 1.68; 95% confidence interval [CI], 1.05–2.68; P = 0.03) or obese (11 vs. 23%; OR, 2.30; 95% CI, 1.53–3.49; P , 0.001). In the adjusted analysis with multiple imputation, the association lost statistical significance in overweight patients (OR, 1.56; 95% CI, 0.90–2.71; P = 0.11) but persisted in obese patients (OR, 1.69; 95% CI, 1.02– 2.81; P = 0.04). The latter finding was driven by an even higher risk of hospitalization in severely obese patients (OR, 1.95; 95% CI, 1.13–3.34; P = 0.02). Conclusions: In this multicenter study of ED patients with asthma exacerbation, we found that obese adults were at a higher risk of hospitalization compared with normal-weight adults. Keywords: acute asthma; obesity; body mass index; hospitalizations; emergency department
(Received in original form June 19, 2014; accepted in final form August 29, 2014 ) The sponsor had no role in the conduct of the study; collection, management, analysis, or interpretation of the data; or preparation, review, or approval of the manuscript. Funded by a grant from Novartis Pharmaceuticals Corporation to Massachusetts General Hospital (primary investigator: C.A.C.). Author Contributions: K.H. was principally responsible for acquisition, collation, statistical analysis, and interpretation of the data, writing of the submitted article, and coordinating the submission process. Y.T. was involved in statistical analysis and interpretation of the data and substantial revision of the submitted article. B.L.L. and H.A.S. were involved in acquisition and interpretation of data and substantial revision of the article. A.F.S. was involved in the conception, hypothesis delineation, and design of the study, acquisition and interpretation of data, revision of the article, coordination of this study, and maintenance of the database from which these data were extracted. C.A.C. was involved in the conception, hypothesis delineation, and design of the study, supervision of statistical analysis, and substantial revision of the submitted article. Correspondence and requests for reprints should be addressed to Kohei Hasegawa, M.D., M.P.H., Department of Emergency Medicine, Massachusetts General Hospital, 326 Cambridge Street, Suite 410, Boston, MA 02114. E-mail: [email protected] This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org Ann Am Thorac Soc Vol 11, No 9, pp 1439–1444, Nov 2014 Copyright © 2014 by the American Thoracic Society DOI: 10.1513/AnnalsATS.201406-270BC Internet address: www.atsjournals.org
Asthma is an important public health problem in the United States. The prevalence remains at historically high levels, affecting 26 million Americans (1).
Additionally, the public health burden of asthma hospitalizations is significant: 385,000 hospitalizations in 2011, with an estimated direct cost of $2.3 billion
Hasegawa, Tsugawa, Lopez, et al.: Obesity and Hospitalization in Acute Asthma
annually (2). Parallel to the high prevalence of asthma, the United States is in the midst of an obesity epidemic, with 35% of U.S. adults obese in 2011 to 2012 (3). 1439
BRIEF COMMUNICATION Many studies have linked obesity to the etiology of asthma (4, 5) and to worse chronic severity (6–8) and control (4, 6, 8, 9). In contrast, fewer studies have examined the severity or clinical course of asthma exacerbations in obese individuals (10–14). This more limited literature has shown that the severity of asthma exacerbation in obese adults—as measured by risk of
hospitalization—is no different (10, 12) or greater (11) than that normal-weight individuals. These earlier studies were potentially limited by small sample sizes, misclassification of the patients, and potential problems with generalizability. Despite its clinical relevance, the relationship between obesity and acute asthma morbidity remains to be elucidated.
In the present study, we investigated the impact of obesity on the risk of acute asthma outcomes using data from a 48-center observational study of emergency department (ED) patients with asthma exacerbation. We hypothesized that obese adults presenting to the ED with asthma exacerbation are at a higher risk of hospitalization compared with normal-weight adults.
Table 1. Characteristics of patients with asthma exacerbation, according to body mass index group Patient Characteristics Demographics Age, median (IQR), yr Female sex Race/ethnicity Non-Hispanic white Non-Hispanic black Hispanic ethnicity Others Current smoker Having primary care physician Health insurance* Private Public No insurance Median household income estimated from ZIP code, median (IQR) Chronic asthma factors Ever admitted for asthma Ever intubated for asthma Ever used systemic corticosteroids ED visit for asthma in past yr Seen by asthma specialist in past 12 mo Current Asthma medications Current use of oral corticosteroids Current use of inhaled corticosteroids Current use of long-acting b-agonist Current use of leukotriene modifiers Current use of omalizumab
Underweight (n = 16)
Normal Weight (n = 281)
Overweight (n = 323)
Obese (n = 607)
23 (20–37) 11 (69)
29 (22–41) 136 (48)
36 (25–46) 183 (57)
38 (29–46) 431 (71)
2 9 4 1 9 13
(13) (56) (25) (7) (57) (81)
57 149 52 23 93 142
(20) (53) (19) (8) (33) (51)
51 175 76 21 116 185
(16) (54) (24) (7) (36) (57)
115 334 109 49 180 434
(19) (55) (18) (8) (30) (71)
P Value
,0.001 ,0.001 0.65
0.23 ,0.001 0.001
6 (38) 3 (19) 7 (44) 28,615 (26,046–34,999)
78 (28) 93 (33) 97 (35) 33,902 (25,977–44,431)
6 (38)
108 (38)
1 (7)
96 (30) 116 (36) 99 (31) 31,679 (25,886–45,702) 34,336
196 (32) 254 (42) 134 (22) (26,279–46,846)
0.39
113 (35)
237 (39)
0.48
33 (12)
41 (13)
87 (14)
0.58
7 (44)
149 (53)
183 (57)
373 (61)
0.09
2 (13)
139 (49)
154 (48)
285 (47)
0.50
1 (7)
18 (6)
27 (8)
77 (13)
0.058
1 (7)
30 (11)
43 (13)
96 (16)
0.23
3 (19)
85 (30)
116 (36)
264 (43)
,0.001
3 (19)
59 (21)
82 (25)
184 (30)
0.02
1 (7)
32 (11)
24 (7)
90 (15)
0.01
0 (0)
0 (0)
0 (0)
9 (1)
0.03
Definition of abbreviations: ED = emergency department; IQR = interquartile range. Data are expressed as n (%) unless otherwise indicated. *Percentages are not equal 100 because of missing data.
1440
AnnalsATS Volume 11 Number 9 | November 2014
BRIEF COMMUNICATION Methods Study Design and Setting
We performed a multicenter chart review study to characterize adult ED patients with asthma exacerbation, as part of the Multicenter Airway Research Collaboration (MARC), a program of the Emergency Medicine Network (15). A total of 48 academic and community EDs across 23 U.S. states completed the study (see Table E1 in the online supplement). All patients were managed at the discretion of the treating physician. The institutional review board of each center approved the study. Selection of Participants
Using the billing code 493.xx (16), each site identified all ED visits with a primary ED or hospital discharge diagnosis of asthma during a 12-month period during January 1, 2011 to December 31, 2012. Inclusion criteria were ED visits made by adult patients aged 18 to 54 years and a history of
physician-diagnosed asthma before the index ED visit. We excluded patients aged 55 years or older because of potential difficulty distinguishing between asthma and chronic obstructive pulmonary disease (COPD). We excluded ED visits made by patients with a history of physiciandiagnosed COPD, transfer visits, repeat visits by the same individual, or visits not prompted largely by acute asthma. These criteria were the same as in our earlier research on this topic (1). Methods of Measurement
Onsite chart abstractors reviewed 40 ED charts randomly selected by the coordinating center at Massachusetts General Hospital. Two hospitals each examined an additional 40 charts to obtain a total of 2,000 charts. All abstractors participated in 1-hour web-conference training and then completed two practice charts, which were evaluated with a “criterion standard.” If an abstractor’s
accuracy was less than 80% per chart, the abstractor was retrained. Data abstraction was performed with a standardized form and included patients’ weight and height, demographics, chronic asthma factors, current asthma medications, and details of the current asthma exacerbation, including peak expiratory flow (PEF), ED management, ED length of stay (LOS), and ED disposition. PEF was used to assess acute asthma severity and was expressed as absolute value. Final changes in PEF from baseline were expressed as the relative change in absolute PEF value. Body mass index (BMI) was used as a measure of excess body fat (10–12). When the ED chart had no information on the patient’s weight or height, sites used values documented in non-ED records (e.g., primary care record) within 1 year before or after the index ED visit. Among the patients with recorded weight and height, 76% had the data within 1 month from the ED visit and 87% had the data within 6 months from the ED visit.
Table 2. Asthma exacerbation presentation and emergency department course, according to body mass index group Variables
ED presentations Duration of symptoms 30 kg/m2) (3). For the unadjusted analyses, we excluded patients with missing BMI. We compared patient characteristics, ED presentation, and ED course by BMI status using Chi-square, Fisher exact, or Kruskal-Wallis tests. To examine the association of BMI status with risk of hospitalization for asthma exacerbation, we constructed two regression models. First, we fitted an unadjusted model that included only BMI status as the independent variable. Second, we fitted a two-level hierarchical model with binomial response using random intercepts for the EDs to account for patient clustering at the ED level. We adjusted for 11 patient-level variables (i.e., age, sex, race/ ethnicity, smoking, insurance, median household income, primary care physician status, history of hospitalization for asthma, current use of systemic corticosteroids, concomitant medical disorders, and initial PEF at ED presentation). In the adjusted analyses, we conducted the multiple imputation with multivariate normal approach to account for the variables with missing data (17). We included both the dependent variable (hospitalization) and the primary exposure variable (BMI) in the imputation model and then excluded patients with imputed values for these variables from the analysis models. Detailed methods of the imputation method may be found in the online supplement. In sensitivity analyses, to address the effect of patients with a prolonged ED LOS, we repeated the model using a different definition of the outcome (i.e., hospitalization or ED LOS > 480 min). In addition, we stratified the analyses by severity of obesity (BMI, 30.0–34.9 vs. 35.01 kg/m2). All analyses were performed with SAS 9.3 (SAS Institute, Cary, NC) and Stata 12.0 software (StataCorp, College Station, TX).
patients missing weight or height, which precluded calculation of BMI. After this exclusion, the analytic cohort comprised 1,227 patients (61%) for the unadjusted analyses. The analytic and nonanalytic cohorts were similar in their ED presentation, such as initial respiratory rate and PEF (all P . 0.05; data not shown). However, patients in the analytic cohort were slightly younger (median, 32 vs. 35 yr) and more likely to be hospitalized (20 vs. 13%; both P , 0.05). Of the 1,227 patients in the analytic cohort, 323 patients (26%) were overweight, and 607 patients (50%) were obese (Table 1). Among the 607 obese patients, 364 patients (60%) were severely obese (BMI > 35 kg/m2). Patient characteristics differed across the BMI groups. Compared with normal-weight patients, overweight and obese patients were older and more likely to be women. Similarly, these overweight and obese patients were more likely to have a primary care physician and private or public health insurance. Although most markers of chronic asthma severity did not differ across groups, overweight and obese patients were more likely to have been on long-term control medications, such as inhaled corticosteroids and long-acting b-agonists, at the time of the ED visit. Table 2 summarizes the acute asthma presentation and ED course by BMI status. Although the respiratory rate at ED presentation was statistically different across groups, there were no clinically important differences. In addition, PEF at
Of 2,000 enrolled adults with asthma exacerbation from 48 EDs, we excluded 773 1442
Discussion This multicenter study of adults presenting to 48 EDs with asthma exacerbation demonstrated that one-half of this patient population was obese. In addition, we found that most markers for chronic severity and acute severity (e.g., vital signs and PEF at ED presentation) did not differ across the BMI groups. However, in comparison with normal-weight patients, obese patients with asthma exacerbation had a significantly higher risk of hospitalization. To our
.8 .6 .4 .2 95% of all subjects 0 0
Results
ED presentation and ED treatment did not differ across groups. Overall, there was a positive relationship between BMI and risk of hospitalization (Figure 1). Compared with normal-weight patients, the unadjusted risk of hospitalization was higher in patients who were overweight (11 vs. 18%; P = 0.03; Table 3) or obese (11 vs. 23%; P , 0.001). In the multivariable-adjusted model with multiple imputation (n = 1,204), the association lost statistical significance in overweight patients (OR, 1.56; 95% CI, 0.90–2.71; P = 0.11) but persisted in obese patients (OR, 1.69; 95% CI, 1.02–2.81; P = 0.04). The latter finding was driven by an even higher risk of hospitalization in severely obese patients (OR, 1.95; 95% CI, 1.13–3.34; P = 0.02; Table E3). In the sensitivity analysis with hospitalization or prolonged ED LOS as the dependent variable, the results did not change materially.
1 Probability of Hospitalization
Outcome Measure
10
20
30 40 50 60 70 Body Mass Index (kg/m2)
80
90
100
Figure 1. Unadjusted association of body mass index with the probability of hospitalization in adults presenting to the emergency department with asthma exacerbation. The fitting line represents locally weighted scatterplot smoothed (lowess) curve.
AnnalsATS Volume 11 Number 9 | November 2014
BRIEF COMMUNICATION Table 3. Unadjusted and multivariable associations of body mass index status with risk of hospitalization for asthma exacerbation Obesity Status
Normal weight Underweight Overweight Obese
Unadjusted Model*
Adjusted Model†
Sensitivity Analysis‡
OR (95% CI)
P Value
OR (95% CI)
P Value
OR (95% CI)
P Value
Reference 3.43 (1.11–10.5) 1.68 (1.05–2.68) 2.30 (1.52–3.49)
— 0.03 0.03 ,0.001
Reference 4.52 (1.27–16.1) 1.56 (0.90–2.71) 1.69 (1.02–2.81)
— 0.02 0.11 0.04
Reference 2.47 (0.72–8.55) 1.33 (0.85–2.07) 1.55 (1.02–2.33)
— 0.15 0.22 0.04
Definition of abbreviations: CI = confidence interval; OR = odds ratio. *Unadjusted model with hospitalization as the dependent variable (n = 1,204). † Hierarchical model with hospitalization as the dependent variable after multiple imputation (n = 1,204). ‡ Hierarchical model with hospitalization or emergency department length of stay > 480 minutes as the dependent variable after multiple imputation (n = 1,204).
knowledge, this is the largest study to have examined the association between BMI status and acute asthma morbidity in ED patients. Our data corroborate and build on previous reports on the association between obesity and risk of hospitalization, a finding of public health importance. Two novel findings are the dose-effect relationship between BMI and risk of hospitalization (Figure 1) and the identification of a subgroup at the highest risk (i.e., severely obese patients). Despite its seemingly obvious clinical and public health importance, few prior studies have examined the risk of hospitalization among obese children (13, 14) and adults (10–12) presenting to the ED with asthma exacerbation. Our previous multicenter study of 572 ED adult patients with asthma exacerbation in 2000 to 2001 found that the risk of hospitalization did not differ between underweight/normalweight and obese patients (10). Likewise, another study in a single hospital network system of 160 ED adult patients showed no difference in this outcome between nonobese (normal-weight/overweight) and obese patients (12). By contrast, a singlecenter study in Uruguay of 426 ED patients with severe asthma exacerbation demonstrated that overweight/obese adults had a higher risk of hospitalization (11). Differences in study design, patient population, classification of the BMI status, clinical setting, and statistical power—or possible combinations of these factors— may explain the differences across these studies. Consistent with the Uruguayan study, the present U.S.-based multicenter study with approximately three times larger sample size demonstrated that obese adults with asthma exacerbation had a higher risk of hospitalization than did normal-weight
adults. Interestingly, the unadjusted analysis demonstrated an increased risk of hospitalization in underweight patients. However, this finding should be interpreted with caution because of the low number of patients in this subgroup. Although clinicians may believe that patients with asthma with higher BMI should have increased risk of hospitalization through their higher chronic asthma severity and prevalence of comorbidities, we found that the association persisted even after adjusting for these factors. We suspect that the explanation is more complicated and likely multifactorial (4). The literature has linked obesity with decreased corticosteroid responsiveness, adipocytederived proinflammatory mediators, decrease in the expiratory reserve volume, and adiposity-related mechanical fat load (4, 18–23). All of these mechanisms may promote the acute severity and differential treatment response in this specific population. We acknowledge several potential limitations to the study. First, 39% of patients did not have BMI data, and this is a potential source of bias. However, the similarity in markers for acute asthma severity (e.g., respiratory rate and PEF at ED presentation) across the analytic and nonanalytic cohorts argues against a significant bias. In addition, in the adjusted model with multiple imputation recovering missing information, our inference remained clinically and statistically significant. In this study, we believe that the missing mechanism can be fully explained by the variables used in the imputation model. It is possible that normal-weight patients might have used healthcare less frequently, thereby leading to a higher probability of missing BMI. In this case,
Hasegawa, Tsugawa, Lopez, et al.: Obesity and Hospitalization in Acute Asthma
once we predict the missing data on BMI using the markers of healthcare use, the unobserved BMI values are no longer relevant to the missing data on this variable, and thus the missing at random assumption holds. Therefore, with the multiple imputation, the missing values of BMI would not have biased our estimates. Second, we used the weight and height values documented in non-ED records (e.g., primary care record) when the ED chart had no information. Thus, some of the recorded values may not have reflected those at the time of ED visit. Third, our observational data relied on medical record review for measurements of BMI and outcomes; therefore, error in data measurement is possible. However, prior studies demonstrated excellent interrater reliability in ED disposition (e.g., hospitalization) in patients with asthma exacerbation (k coefficient, 0.95) (24). Furthermore, we used a previously applied standard data collection system with uniform definitions and structured data forms (10). Fourth, in this multicenter observational study, we did not use standard criteria for hospitalization. Therefore, institutional variability in ED disposition is possible. However, the significant association between obesity and a higher risk of hospitalization persisted even after accounting for the influence of site. Finally, our sample consisted predominantly of urban, academic EDs. Therefore, these results may not be generalizable to rural or community medical centers. Yet, urban populations have disproportionately high asthma morbidity; this is the population for which targeted interventions are most urgently warranted. In sum, this large multicenter study showed that one-half of the adults presenting to the ED with asthma 1443
BRIEF COMMUNICATION exacerbation were obese. We also found that obese patients with asthma exacerbation are at a higher risk of hospitalization compared with normal-weight adults. This finding was driven by an even higher risk among severely obese patients. For researchers, these findings support further investigation of the mechanisms by which obesity affects acute asthma morbidity. In addition, the literature suggests that weight reduction in obese patients with asthma may improve asthma symptoms
and reduce the frequency of asthma exacerbations (25, 26). For clinicians, our finding underscores the importance of weight reduction measures in addition to high-quality longitudinal asthma care for obese patients. Furthermore, in the setting of obesity epidemic in the United States, the higher risk of hospitalization among obese patients is of interest to policy makers and payers. To mitigate asthma-related healthcare use (and expenditures) in an already stressed
References 1 Centers for Disease Control and Prevention. Asthma. 2012 [accessed 2014 June 17]. Available from: http://www.cdc.gov/nchs/fastats/ asthma.htm 2 HCUPNet. US Department of Health and Human Services. Agency for Healthcare Research and Quality. [accessed 2014 June 17]. Available from: http://hcupnet.ahrq.gov/ 3 Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of childhood and adult obesity in the United States, 2011-2012. JAMA 2014;311: 806–814. 4 Dixon AE, Holguin F, Sood A, Salome CM, Pratley RE, Beuther DA, Celedon ´ JC, Shore SA; American Thoracic Society Ad Hoc Subcommittee on Obesity and Lung Disease. An official American Thoracic Society Workshop report: obesity and asthma. Proc Am Thorac Soc 2010;7:325–335. 5 Camargo CA Jr, Weiss ST, Zhang S, Willett WC, Speizer FE. Prospective study of body mass index, weight change, and risk of adult-onset asthma in women. Arch Intern Med 1999;159: 2582–2588. 6 Mosen DM, Schatz M, Magid DJ, Camargo CA, Jr. The relationship between obesity and asthma severity and control in adults. J Allergy Clin Immunol 2008;122:507–511.e6. 7 Taylor B, Mannino D, Brown C, Crocker D, Twum-Baah N, Holguin F. Body mass index and asthma severity in the National Asthma Survey. Thorax 2008;63:14–20. 8 Quinto KB, Zuraw BL, Poon KY, Chen W, Schatz M, Christiansen SC. The association of obesity and asthma severity and control in children. J Allergy Clin Immunol 2011;128:964–969. 9 Schatz M, Zeiger RS, Zhang F, Chen W, Yang SJ, Camargo CA Jr. Overweight/obesity and risk of seasonal asthma exacerbations. J Allergy Clin Immunol Pract 2013;1:618–622. 10 Thomson CC, Clark S, Camargo CA Jr, Investigators M; MARC Investigators. Body mass index and asthma severity among adults presenting to the emergency department. Chest 2003;124:795–802. 11 Rodrigo GJ, Plaza V. Body mass index and response to emergency department treatment in adults with severe asthma exacerbations: a prospective cohort study. Chest 2007;132:1513–1519. 12 Yeh KH, Skowronski ME, Coreno AJ, Seitz RE, Villalba KD, DickeyWhite H, McFadden ER. Impact of obesity on the severity and therapeutic responsiveness of acute episodes of asthma. J Asthma 2011;48:546–552.
1444
healthcare system, our finding should encourage policy makers to develop and evaluate public health interventions that promote weight reduction in this highrisk population. n Author disclosures are available with the text of this article at www.atsjournals.org. Acknowledgment: The authors thank the MARC-36 study hospitals and research personnel for their ongoing dedication to asthma research (Table E1).
13 Carroll CL, Stoltz P, Raykov N, Smith SR, Zucker AR. Childhood overweight increases hospital admission rates for asthma. Pediatrics 2007;120:734–740. 14 Ginde AA, Santillan AA, Clark S, Camargo CA Jr. Body mass index and acute asthma severity among children presenting to the emergency department. Pediatr Allergy Immunol 2010;21:480–488. 15 Emergency Medicine Network. [updated 2014 Oct 23; accessed 2014 June 17]. Available from: http://www.emnet-usa.org/ 16 National Center for Health Statistics. International Classification of Diseases, Ninth Revision, Clinical Modification. [updated 2013 Jun 18; accessed 2014 June 17]. Available from: http://www.cdc.gov/ nchs/icd/icd9cm.htm 17 Little RJ, Rubin DB. Statistical analysis with missing data. New York: Wiley-Interscience; 2002. 18 Stream AR, Sutherland ER. Obesity and asthma disease phenotypes. Curr Opin Allergy Clin Immunol 2012;12:76–81. 19 Peters-Golden M, Swern A, Bird SS, Hustad CM, Grant E, Edelman JM. Influence of body mass index on the response to asthma controller agents. Eur Respir J 2006;27:495–503. 20 Sutherland ER, Lehman EB, Teodorescu M, Wechsler ME, National Heart, Lung, Blood Institute’s Asthma Clinical Research Network. Body mass index and phenotype in subjects with mild-to-moderate persistent asthma. J Allergy Clin Immunol 2009;123:1328–1334.e1. 21 Forno E, Lescher R, Strunk R, Weiss S, Fuhlbrigge A, Celedon ´ JC; Childhood Asthma Management Program Research Group. Decreased response to inhaled steroids in overweight and obese asthmatic children. J Allergy Clin Immunol 2011;127:741–749. 22 Gibson PG. Obesity and asthma. Ann Am Thorac Soc 2013;10: S138–S142. 23 Sideleva O, Suratt BT, Black KE, Tharp WG, Pratley RE, Forgione P, Dienz O, Irvin CG, Dixon AE. Obesity and asthma: an inflammatory disease of adipose tissue not the airway. Am J Respir Crit Care Med 2012;186:598–605. 24 Hasegawa K, Chiba T, Hagiwara Y, Watase H, Tsugawa Y, Brown DF, Camargo CA Jr; Japanese Emergency Medicine Network Investigators. Quality of care for acute asthma in emergency departments in Japan: a multicenter observational study. J Allergy Clin Immunol Pract 2013;1:509–515.e1–e3. 25 Adeniyi FB, Young T. Weight loss interventions for chronic asthma. Cochrane Database Syst Rev 2012;7:CD009339. 26 Eneli IU, Skybo T, Camargo CA Jr. Weight loss and asthma: a systematic review. Thorax 2008;63:671–676.
AnnalsATS Volume 11 Number 9 | November 2014
Department of Emergency Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts; 2Harvard Interfaculty Initiative in Health Policy, Cambridge, Massachusetts; 3Department of Emergency Medicine, Thomas Jefferson University Hospital, Philadelphia, Pennsylvania; and 4Department of Emergency Medicine, Baystate Medical Center, Springfield, Massachusetts
Abstract Rationale: Studies have linked obesity to incident asthma and worse chronic severity/control. However, the relationship between obesity and acute asthma morbidity remains unclear. Objectives: To determine whether obese adults presenting to the emergency department (ED) with asthma exacerbation are at higher risk of hospitalization compared with normal-weight adults. Methods: Multicenter chart review study of 48 EDs across 23 U.S. states. We identified ED patients aged 18 to 54 years with asthma exacerbation during 2011 to 2012. Primary outcome was hospitalization. Measurements and Main Results: The analytic cohort comprised 1,227 patients. Of these, 323 patients (27%) were overweight (body mass index [BMI], 25–29.9 kg/m2), and 607 (50%) were obese (BMI > 30 kg/m2). Among the 607 obese patients, 364 patients (60%) were severely obese (BMI > 35 kg/m2).
Several markers of chronic severity/control of asthma and acute severity did not differ across BMI groups. By contrast, compared with normal-weight patients, the risk of hospitalization was higher in patients who were overweight (11 vs. 18%; odds ratio [OR], 1.68; 95% confidence interval [CI], 1.05–2.68; P = 0.03) or obese (11 vs. 23%; OR, 2.30; 95% CI, 1.53–3.49; P , 0.001). In the adjusted analysis with multiple imputation, the association lost statistical significance in overweight patients (OR, 1.56; 95% CI, 0.90–2.71; P = 0.11) but persisted in obese patients (OR, 1.69; 95% CI, 1.02– 2.81; P = 0.04). The latter finding was driven by an even higher risk of hospitalization in severely obese patients (OR, 1.95; 95% CI, 1.13–3.34; P = 0.02). Conclusions: In this multicenter study of ED patients with asthma exacerbation, we found that obese adults were at a higher risk of hospitalization compared with normal-weight adults. Keywords: acute asthma; obesity; body mass index; hospitalizations; emergency department
(Received in original form June 19, 2014; accepted in final form August 29, 2014 ) The sponsor had no role in the conduct of the study; collection, management, analysis, or interpretation of the data; or preparation, review, or approval of the manuscript. Funded by a grant from Novartis Pharmaceuticals Corporation to Massachusetts General Hospital (primary investigator: C.A.C.). Author Contributions: K.H. was principally responsible for acquisition, collation, statistical analysis, and interpretation of the data, writing of the submitted article, and coordinating the submission process. Y.T. was involved in statistical analysis and interpretation of the data and substantial revision of the submitted article. B.L.L. and H.A.S. were involved in acquisition and interpretation of data and substantial revision of the article. A.F.S. was involved in the conception, hypothesis delineation, and design of the study, acquisition and interpretation of data, revision of the article, coordination of this study, and maintenance of the database from which these data were extracted. C.A.C. was involved in the conception, hypothesis delineation, and design of the study, supervision of statistical analysis, and substantial revision of the submitted article. Correspondence and requests for reprints should be addressed to Kohei Hasegawa, M.D., M.P.H., Department of Emergency Medicine, Massachusetts General Hospital, 326 Cambridge Street, Suite 410, Boston, MA 02114. E-mail: [email protected] This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org Ann Am Thorac Soc Vol 11, No 9, pp 1439–1444, Nov 2014 Copyright © 2014 by the American Thoracic Society DOI: 10.1513/AnnalsATS.201406-270BC Internet address: www.atsjournals.org
Asthma is an important public health problem in the United States. The prevalence remains at historically high levels, affecting 26 million Americans (1).
Additionally, the public health burden of asthma hospitalizations is significant: 385,000 hospitalizations in 2011, with an estimated direct cost of $2.3 billion
Hasegawa, Tsugawa, Lopez, et al.: Obesity and Hospitalization in Acute Asthma
annually (2). Parallel to the high prevalence of asthma, the United States is in the midst of an obesity epidemic, with 35% of U.S. adults obese in 2011 to 2012 (3). 1439
BRIEF COMMUNICATION Many studies have linked obesity to the etiology of asthma (4, 5) and to worse chronic severity (6–8) and control (4, 6, 8, 9). In contrast, fewer studies have examined the severity or clinical course of asthma exacerbations in obese individuals (10–14). This more limited literature has shown that the severity of asthma exacerbation in obese adults—as measured by risk of
hospitalization—is no different (10, 12) or greater (11) than that normal-weight individuals. These earlier studies were potentially limited by small sample sizes, misclassification of the patients, and potential problems with generalizability. Despite its clinical relevance, the relationship between obesity and acute asthma morbidity remains to be elucidated.
In the present study, we investigated the impact of obesity on the risk of acute asthma outcomes using data from a 48-center observational study of emergency department (ED) patients with asthma exacerbation. We hypothesized that obese adults presenting to the ED with asthma exacerbation are at a higher risk of hospitalization compared with normal-weight adults.
Table 1. Characteristics of patients with asthma exacerbation, according to body mass index group Patient Characteristics Demographics Age, median (IQR), yr Female sex Race/ethnicity Non-Hispanic white Non-Hispanic black Hispanic ethnicity Others Current smoker Having primary care physician Health insurance* Private Public No insurance Median household income estimated from ZIP code, median (IQR) Chronic asthma factors Ever admitted for asthma Ever intubated for asthma Ever used systemic corticosteroids ED visit for asthma in past yr Seen by asthma specialist in past 12 mo Current Asthma medications Current use of oral corticosteroids Current use of inhaled corticosteroids Current use of long-acting b-agonist Current use of leukotriene modifiers Current use of omalizumab
Underweight (n = 16)
Normal Weight (n = 281)
Overweight (n = 323)
Obese (n = 607)
23 (20–37) 11 (69)
29 (22–41) 136 (48)
36 (25–46) 183 (57)
38 (29–46) 431 (71)
2 9 4 1 9 13
(13) (56) (25) (7) (57) (81)
57 149 52 23 93 142
(20) (53) (19) (8) (33) (51)
51 175 76 21 116 185
(16) (54) (24) (7) (36) (57)
115 334 109 49 180 434
(19) (55) (18) (8) (30) (71)
P Value
,0.001 ,0.001 0.65
0.23 ,0.001 0.001
6 (38) 3 (19) 7 (44) 28,615 (26,046–34,999)
78 (28) 93 (33) 97 (35) 33,902 (25,977–44,431)
6 (38)
108 (38)
1 (7)
96 (30) 116 (36) 99 (31) 31,679 (25,886–45,702) 34,336
196 (32) 254 (42) 134 (22) (26,279–46,846)
0.39
113 (35)
237 (39)
0.48
33 (12)
41 (13)
87 (14)
0.58
7 (44)
149 (53)
183 (57)
373 (61)
0.09
2 (13)
139 (49)
154 (48)
285 (47)
0.50
1 (7)
18 (6)
27 (8)
77 (13)
0.058
1 (7)
30 (11)
43 (13)
96 (16)
0.23
3 (19)
85 (30)
116 (36)
264 (43)
,0.001
3 (19)
59 (21)
82 (25)
184 (30)
0.02
1 (7)
32 (11)
24 (7)
90 (15)
0.01
0 (0)
0 (0)
0 (0)
9 (1)
0.03
Definition of abbreviations: ED = emergency department; IQR = interquartile range. Data are expressed as n (%) unless otherwise indicated. *Percentages are not equal 100 because of missing data.
1440
AnnalsATS Volume 11 Number 9 | November 2014
BRIEF COMMUNICATION Methods Study Design and Setting
We performed a multicenter chart review study to characterize adult ED patients with asthma exacerbation, as part of the Multicenter Airway Research Collaboration (MARC), a program of the Emergency Medicine Network (15). A total of 48 academic and community EDs across 23 U.S. states completed the study (see Table E1 in the online supplement). All patients were managed at the discretion of the treating physician. The institutional review board of each center approved the study. Selection of Participants
Using the billing code 493.xx (16), each site identified all ED visits with a primary ED or hospital discharge diagnosis of asthma during a 12-month period during January 1, 2011 to December 31, 2012. Inclusion criteria were ED visits made by adult patients aged 18 to 54 years and a history of
physician-diagnosed asthma before the index ED visit. We excluded patients aged 55 years or older because of potential difficulty distinguishing between asthma and chronic obstructive pulmonary disease (COPD). We excluded ED visits made by patients with a history of physiciandiagnosed COPD, transfer visits, repeat visits by the same individual, or visits not prompted largely by acute asthma. These criteria were the same as in our earlier research on this topic (1). Methods of Measurement
Onsite chart abstractors reviewed 40 ED charts randomly selected by the coordinating center at Massachusetts General Hospital. Two hospitals each examined an additional 40 charts to obtain a total of 2,000 charts. All abstractors participated in 1-hour web-conference training and then completed two practice charts, which were evaluated with a “criterion standard.” If an abstractor’s
accuracy was less than 80% per chart, the abstractor was retrained. Data abstraction was performed with a standardized form and included patients’ weight and height, demographics, chronic asthma factors, current asthma medications, and details of the current asthma exacerbation, including peak expiratory flow (PEF), ED management, ED length of stay (LOS), and ED disposition. PEF was used to assess acute asthma severity and was expressed as absolute value. Final changes in PEF from baseline were expressed as the relative change in absolute PEF value. Body mass index (BMI) was used as a measure of excess body fat (10–12). When the ED chart had no information on the patient’s weight or height, sites used values documented in non-ED records (e.g., primary care record) within 1 year before or after the index ED visit. Among the patients with recorded weight and height, 76% had the data within 1 month from the ED visit and 87% had the data within 6 months from the ED visit.
Table 2. Asthma exacerbation presentation and emergency department course, according to body mass index group Variables
ED presentations Duration of symptoms 30 kg/m2) (3). For the unadjusted analyses, we excluded patients with missing BMI. We compared patient characteristics, ED presentation, and ED course by BMI status using Chi-square, Fisher exact, or Kruskal-Wallis tests. To examine the association of BMI status with risk of hospitalization for asthma exacerbation, we constructed two regression models. First, we fitted an unadjusted model that included only BMI status as the independent variable. Second, we fitted a two-level hierarchical model with binomial response using random intercepts for the EDs to account for patient clustering at the ED level. We adjusted for 11 patient-level variables (i.e., age, sex, race/ ethnicity, smoking, insurance, median household income, primary care physician status, history of hospitalization for asthma, current use of systemic corticosteroids, concomitant medical disorders, and initial PEF at ED presentation). In the adjusted analyses, we conducted the multiple imputation with multivariate normal approach to account for the variables with missing data (17). We included both the dependent variable (hospitalization) and the primary exposure variable (BMI) in the imputation model and then excluded patients with imputed values for these variables from the analysis models. Detailed methods of the imputation method may be found in the online supplement. In sensitivity analyses, to address the effect of patients with a prolonged ED LOS, we repeated the model using a different definition of the outcome (i.e., hospitalization or ED LOS > 480 min). In addition, we stratified the analyses by severity of obesity (BMI, 30.0–34.9 vs. 35.01 kg/m2). All analyses were performed with SAS 9.3 (SAS Institute, Cary, NC) and Stata 12.0 software (StataCorp, College Station, TX).
patients missing weight or height, which precluded calculation of BMI. After this exclusion, the analytic cohort comprised 1,227 patients (61%) for the unadjusted analyses. The analytic and nonanalytic cohorts were similar in their ED presentation, such as initial respiratory rate and PEF (all P . 0.05; data not shown). However, patients in the analytic cohort were slightly younger (median, 32 vs. 35 yr) and more likely to be hospitalized (20 vs. 13%; both P , 0.05). Of the 1,227 patients in the analytic cohort, 323 patients (26%) were overweight, and 607 patients (50%) were obese (Table 1). Among the 607 obese patients, 364 patients (60%) were severely obese (BMI > 35 kg/m2). Patient characteristics differed across the BMI groups. Compared with normal-weight patients, overweight and obese patients were older and more likely to be women. Similarly, these overweight and obese patients were more likely to have a primary care physician and private or public health insurance. Although most markers of chronic asthma severity did not differ across groups, overweight and obese patients were more likely to have been on long-term control medications, such as inhaled corticosteroids and long-acting b-agonists, at the time of the ED visit. Table 2 summarizes the acute asthma presentation and ED course by BMI status. Although the respiratory rate at ED presentation was statistically different across groups, there were no clinically important differences. In addition, PEF at
Of 2,000 enrolled adults with asthma exacerbation from 48 EDs, we excluded 773 1442
Discussion This multicenter study of adults presenting to 48 EDs with asthma exacerbation demonstrated that one-half of this patient population was obese. In addition, we found that most markers for chronic severity and acute severity (e.g., vital signs and PEF at ED presentation) did not differ across the BMI groups. However, in comparison with normal-weight patients, obese patients with asthma exacerbation had a significantly higher risk of hospitalization. To our
.8 .6 .4 .2 95% of all subjects 0 0
Results
ED presentation and ED treatment did not differ across groups. Overall, there was a positive relationship between BMI and risk of hospitalization (Figure 1). Compared with normal-weight patients, the unadjusted risk of hospitalization was higher in patients who were overweight (11 vs. 18%; P = 0.03; Table 3) or obese (11 vs. 23%; P , 0.001). In the multivariable-adjusted model with multiple imputation (n = 1,204), the association lost statistical significance in overweight patients (OR, 1.56; 95% CI, 0.90–2.71; P = 0.11) but persisted in obese patients (OR, 1.69; 95% CI, 1.02–2.81; P = 0.04). The latter finding was driven by an even higher risk of hospitalization in severely obese patients (OR, 1.95; 95% CI, 1.13–3.34; P = 0.02; Table E3). In the sensitivity analysis with hospitalization or prolonged ED LOS as the dependent variable, the results did not change materially.
1 Probability of Hospitalization
Outcome Measure
10
20
30 40 50 60 70 Body Mass Index (kg/m2)
80
90
100
Figure 1. Unadjusted association of body mass index with the probability of hospitalization in adults presenting to the emergency department with asthma exacerbation. The fitting line represents locally weighted scatterplot smoothed (lowess) curve.
AnnalsATS Volume 11 Number 9 | November 2014
BRIEF COMMUNICATION Table 3. Unadjusted and multivariable associations of body mass index status with risk of hospitalization for asthma exacerbation Obesity Status
Normal weight Underweight Overweight Obese
Unadjusted Model*
Adjusted Model†
Sensitivity Analysis‡
OR (95% CI)
P Value
OR (95% CI)
P Value
OR (95% CI)
P Value
Reference 3.43 (1.11–10.5) 1.68 (1.05–2.68) 2.30 (1.52–3.49)
— 0.03 0.03 ,0.001
Reference 4.52 (1.27–16.1) 1.56 (0.90–2.71) 1.69 (1.02–2.81)
— 0.02 0.11 0.04
Reference 2.47 (0.72–8.55) 1.33 (0.85–2.07) 1.55 (1.02–2.33)
— 0.15 0.22 0.04
Definition of abbreviations: CI = confidence interval; OR = odds ratio. *Unadjusted model with hospitalization as the dependent variable (n = 1,204). † Hierarchical model with hospitalization as the dependent variable after multiple imputation (n = 1,204). ‡ Hierarchical model with hospitalization or emergency department length of stay > 480 minutes as the dependent variable after multiple imputation (n = 1,204).
knowledge, this is the largest study to have examined the association between BMI status and acute asthma morbidity in ED patients. Our data corroborate and build on previous reports on the association between obesity and risk of hospitalization, a finding of public health importance. Two novel findings are the dose-effect relationship between BMI and risk of hospitalization (Figure 1) and the identification of a subgroup at the highest risk (i.e., severely obese patients). Despite its seemingly obvious clinical and public health importance, few prior studies have examined the risk of hospitalization among obese children (13, 14) and adults (10–12) presenting to the ED with asthma exacerbation. Our previous multicenter study of 572 ED adult patients with asthma exacerbation in 2000 to 2001 found that the risk of hospitalization did not differ between underweight/normalweight and obese patients (10). Likewise, another study in a single hospital network system of 160 ED adult patients showed no difference in this outcome between nonobese (normal-weight/overweight) and obese patients (12). By contrast, a singlecenter study in Uruguay of 426 ED patients with severe asthma exacerbation demonstrated that overweight/obese adults had a higher risk of hospitalization (11). Differences in study design, patient population, classification of the BMI status, clinical setting, and statistical power—or possible combinations of these factors— may explain the differences across these studies. Consistent with the Uruguayan study, the present U.S.-based multicenter study with approximately three times larger sample size demonstrated that obese adults with asthma exacerbation had a higher risk of hospitalization than did normal-weight
adults. Interestingly, the unadjusted analysis demonstrated an increased risk of hospitalization in underweight patients. However, this finding should be interpreted with caution because of the low number of patients in this subgroup. Although clinicians may believe that patients with asthma with higher BMI should have increased risk of hospitalization through their higher chronic asthma severity and prevalence of comorbidities, we found that the association persisted even after adjusting for these factors. We suspect that the explanation is more complicated and likely multifactorial (4). The literature has linked obesity with decreased corticosteroid responsiveness, adipocytederived proinflammatory mediators, decrease in the expiratory reserve volume, and adiposity-related mechanical fat load (4, 18–23). All of these mechanisms may promote the acute severity and differential treatment response in this specific population. We acknowledge several potential limitations to the study. First, 39% of patients did not have BMI data, and this is a potential source of bias. However, the similarity in markers for acute asthma severity (e.g., respiratory rate and PEF at ED presentation) across the analytic and nonanalytic cohorts argues against a significant bias. In addition, in the adjusted model with multiple imputation recovering missing information, our inference remained clinically and statistically significant. In this study, we believe that the missing mechanism can be fully explained by the variables used in the imputation model. It is possible that normal-weight patients might have used healthcare less frequently, thereby leading to a higher probability of missing BMI. In this case,
Hasegawa, Tsugawa, Lopez, et al.: Obesity and Hospitalization in Acute Asthma
once we predict the missing data on BMI using the markers of healthcare use, the unobserved BMI values are no longer relevant to the missing data on this variable, and thus the missing at random assumption holds. Therefore, with the multiple imputation, the missing values of BMI would not have biased our estimates. Second, we used the weight and height values documented in non-ED records (e.g., primary care record) when the ED chart had no information. Thus, some of the recorded values may not have reflected those at the time of ED visit. Third, our observational data relied on medical record review for measurements of BMI and outcomes; therefore, error in data measurement is possible. However, prior studies demonstrated excellent interrater reliability in ED disposition (e.g., hospitalization) in patients with asthma exacerbation (k coefficient, 0.95) (24). Furthermore, we used a previously applied standard data collection system with uniform definitions and structured data forms (10). Fourth, in this multicenter observational study, we did not use standard criteria for hospitalization. Therefore, institutional variability in ED disposition is possible. However, the significant association between obesity and a higher risk of hospitalization persisted even after accounting for the influence of site. Finally, our sample consisted predominantly of urban, academic EDs. Therefore, these results may not be generalizable to rural or community medical centers. Yet, urban populations have disproportionately high asthma morbidity; this is the population for which targeted interventions are most urgently warranted. In sum, this large multicenter study showed that one-half of the adults presenting to the ED with asthma 1443
BRIEF COMMUNICATION exacerbation were obese. We also found that obese patients with asthma exacerbation are at a higher risk of hospitalization compared with normal-weight adults. This finding was driven by an even higher risk among severely obese patients. For researchers, these findings support further investigation of the mechanisms by which obesity affects acute asthma morbidity. In addition, the literature suggests that weight reduction in obese patients with asthma may improve asthma symptoms
and reduce the frequency of asthma exacerbations (25, 26). For clinicians, our finding underscores the importance of weight reduction measures in addition to high-quality longitudinal asthma care for obese patients. Furthermore, in the setting of obesity epidemic in the United States, the higher risk of hospitalization among obese patients is of interest to policy makers and payers. To mitigate asthma-related healthcare use (and expenditures) in an already stressed
References 1 Centers for Disease Control and Prevention. Asthma. 2012 [accessed 2014 June 17]. Available from: http://www.cdc.gov/nchs/fastats/ asthma.htm 2 HCUPNet. US Department of Health and Human Services. Agency for Healthcare Research and Quality. [accessed 2014 June 17]. Available from: http://hcupnet.ahrq.gov/ 3 Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of childhood and adult obesity in the United States, 2011-2012. JAMA 2014;311: 806–814. 4 Dixon AE, Holguin F, Sood A, Salome CM, Pratley RE, Beuther DA, Celedon ´ JC, Shore SA; American Thoracic Society Ad Hoc Subcommittee on Obesity and Lung Disease. An official American Thoracic Society Workshop report: obesity and asthma. Proc Am Thorac Soc 2010;7:325–335. 5 Camargo CA Jr, Weiss ST, Zhang S, Willett WC, Speizer FE. Prospective study of body mass index, weight change, and risk of adult-onset asthma in women. Arch Intern Med 1999;159: 2582–2588. 6 Mosen DM, Schatz M, Magid DJ, Camargo CA, Jr. The relationship between obesity and asthma severity and control in adults. J Allergy Clin Immunol 2008;122:507–511.e6. 7 Taylor B, Mannino D, Brown C, Crocker D, Twum-Baah N, Holguin F. Body mass index and asthma severity in the National Asthma Survey. Thorax 2008;63:14–20. 8 Quinto KB, Zuraw BL, Poon KY, Chen W, Schatz M, Christiansen SC. The association of obesity and asthma severity and control in children. J Allergy Clin Immunol 2011;128:964–969. 9 Schatz M, Zeiger RS, Zhang F, Chen W, Yang SJ, Camargo CA Jr. Overweight/obesity and risk of seasonal asthma exacerbations. J Allergy Clin Immunol Pract 2013;1:618–622. 10 Thomson CC, Clark S, Camargo CA Jr, Investigators M; MARC Investigators. Body mass index and asthma severity among adults presenting to the emergency department. Chest 2003;124:795–802. 11 Rodrigo GJ, Plaza V. Body mass index and response to emergency department treatment in adults with severe asthma exacerbations: a prospective cohort study. Chest 2007;132:1513–1519. 12 Yeh KH, Skowronski ME, Coreno AJ, Seitz RE, Villalba KD, DickeyWhite H, McFadden ER. Impact of obesity on the severity and therapeutic responsiveness of acute episodes of asthma. J Asthma 2011;48:546–552.
1444
healthcare system, our finding should encourage policy makers to develop and evaluate public health interventions that promote weight reduction in this highrisk population. n Author disclosures are available with the text of this article at www.atsjournals.org. Acknowledgment: The authors thank the MARC-36 study hospitals and research personnel for their ongoing dedication to asthma research (Table E1).
13 Carroll CL, Stoltz P, Raykov N, Smith SR, Zucker AR. Childhood overweight increases hospital admission rates for asthma. Pediatrics 2007;120:734–740. 14 Ginde AA, Santillan AA, Clark S, Camargo CA Jr. Body mass index and acute asthma severity among children presenting to the emergency department. Pediatr Allergy Immunol 2010;21:480–488. 15 Emergency Medicine Network. [updated 2014 Oct 23; accessed 2014 June 17]. Available from: http://www.emnet-usa.org/ 16 National Center for Health Statistics. International Classification of Diseases, Ninth Revision, Clinical Modification. [updated 2013 Jun 18; accessed 2014 June 17]. Available from: http://www.cdc.gov/ nchs/icd/icd9cm.htm 17 Little RJ, Rubin DB. Statistical analysis with missing data. New York: Wiley-Interscience; 2002. 18 Stream AR, Sutherland ER. Obesity and asthma disease phenotypes. Curr Opin Allergy Clin Immunol 2012;12:76–81. 19 Peters-Golden M, Swern A, Bird SS, Hustad CM, Grant E, Edelman JM. Influence of body mass index on the response to asthma controller agents. Eur Respir J 2006;27:495–503. 20 Sutherland ER, Lehman EB, Teodorescu M, Wechsler ME, National Heart, Lung, Blood Institute’s Asthma Clinical Research Network. Body mass index and phenotype in subjects with mild-to-moderate persistent asthma. J Allergy Clin Immunol 2009;123:1328–1334.e1. 21 Forno E, Lescher R, Strunk R, Weiss S, Fuhlbrigge A, Celedon ´ JC; Childhood Asthma Management Program Research Group. Decreased response to inhaled steroids in overweight and obese asthmatic children. J Allergy Clin Immunol 2011;127:741–749. 22 Gibson PG. Obesity and asthma. Ann Am Thorac Soc 2013;10: S138–S142. 23 Sideleva O, Suratt BT, Black KE, Tharp WG, Pratley RE, Forgione P, Dienz O, Irvin CG, Dixon AE. Obesity and asthma: an inflammatory disease of adipose tissue not the airway. Am J Respir Crit Care Med 2012;186:598–605. 24 Hasegawa K, Chiba T, Hagiwara Y, Watase H, Tsugawa Y, Brown DF, Camargo CA Jr; Japanese Emergency Medicine Network Investigators. Quality of care for acute asthma in emergency departments in Japan: a multicenter observational study. J Allergy Clin Immunol Pract 2013;1:509–515.e1–e3. 25 Adeniyi FB, Young T. Weight loss interventions for chronic asthma. Cochrane Database Syst Rev 2012;7:CD009339. 26 Eneli IU, Skybo T, Camargo CA Jr. Weight loss and asthma: a systematic review. Thorax 2008;63:671–676.
AnnalsATS Volume 11 Number 9 | November 2014
