- Email: [email protected]
Understanding asthma-chronic obstructive pulmonary disease overlap syndrome
Accepted Manuscript Understanding Asthma-Chronic Obstructive Pulmonary Disease Overlap Syndrome Keele E. Wurst, Kaitlin Kelly-Reif, Greta Bushnell, Steven Pascoe, Neil Barnes PII:
S0954-6111(15)30068-8
DOI:
10.1016/j.rmed.2015.10.004
Reference:
YRMED 4796
To appear in:
Respiratory Medicine
Received Date: 6 May 2015 Revised Date:
27 August 2015
Accepted Date: 5 October 2015
Please cite this article as: Wurst KE, Kelly-Reif K, Bushnell G, Pascoe S, Barnes N, Understanding Asthma-Chronic Obstructive Pulmonary Disease Overlap Syndrome, Respiratory Medicine (2015), doi: 10.1016/j.rmed.2015.10.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Understanding Asthma-Chronic Obstructive Pulmonary Disease Overlap
RI PT
Syndrome
Keele E. Wurst,a Kaitlin Kelly-Reif,b Greta Bushnell,b Steven Pascoe,c Neil Barnesd
a
employed with GSK, Research Triangle Park, NC, USA) b
SC
PAREXEL International, Clinical Research Services, Research Triangle Park, NC USA (previously
c
GSK R&D, Research Triangle Park, NC, USA
d
M AN U
Department of Epidemiology, University of North Carolina, Chapel Hill, NC, USA
GSK, Stockley Park, Uxbridge, UK, William Harvey Institute and Barts and The London School of
Medicine and Dentistry
Keele E. Wurst PhD, RPh PAREXEL International
EP
Clinical Research Services
TE D
Corresponding author:
NC USA
AC C
Research Triangle Park
Email: [email protected]
Telephone: +1 919 483 7994
Key words (3 to 6): asthma, asthma-chronic obstructive pulmonary disease overlap syndrome, chronic obstructive pulmonary disease
1
ACCEPTED MANUSCRIPT
Abbreviations ACOS: asthma-chronic obstructive pulmonary disease overlap syndrome; BOLD: Burden of Obstructive
RI PT
Lung Disease; COPD: chronic obstructive pulmonary disease; COPDGene: Genetic Epidemiology of COPD; GEIRD: Gene-Environment Interactions in Respiratory Disease; GINA: Global Initiative for Asthma; GOLD: Global Initiative for Chronic Obstructive Lung Disease; FEV1: forced expiratory volume in one second;
SC
FVC: forced vital capacity; ICD: International Classification of Disease; IHCIS: Integrated Healthcare Information Services; LLN: lower limit of normal; NC BRFSS: North Carolina Behavioral Risk Factor
M AN U
Surveillance System; NHANES: National Health and Nutritional Examination Survey; PEF: peak expiratory flow; PLATINO: Latin American Project for the Investigation of Lung Disease; SGRQ: Saint George’s Respiratory Questionnaire; SPIROMICS: Subpopulations and intermediate outcome measures in COPD
AC C
EP
TE D
study; U-BIOPRED: Unbiased BIOmarkers in PREDiction of respiratory outcomes.
2
ACCEPTED MANUSCRIPT
Summary Asthma-chronic obstructive pulmonary disease overlap syndrome (ACOS) is a loosely-defined clinical entity referring to patients who exhibit characteristics of both asthma and chronic obstructive
RI PT
pulmonary disease (COPD). Clinical definitions and classifications for ACOS vary widely, which impacts our understanding of prevalence, diagnosis and treatment of the condition.
SC
This literature review was therefore conducted to characterize the prevalence of ACOS and the effect of different disease definitions on these estimates, as this has not previously been explored. From an
M AN U
analysis of English language literature published from 2000 to 2014, the estimated prevalence of ACOS ranges from 12.1% to 55.2% among patients with COPD and 13.3% to 61.0% among patients with asthma alone. This variability is linked to differences in COPD and asthma diagnostic criteria, disease ascertainment methods (spirometry-based versus clinical or symptom-based diagnoses and claims data),
TE D
and population characteristics including age, gender and smoking.
Understanding the reasons for differences in prevalence estimates of ACOS across the literature may
EP
help guide decision making on the most appropriate criteria for defining ACOS and aid investigators in
AC C
designing future ACOS clinical studies aimed at effective treatment.
3
ACCEPTED MANUSCRIPT
Introduction The diagnosis and differentiation of asthma from chronic obstructive pulmonary disease (COPD) in clinical practice is relatively straightforward in the majority of cases; however, some patients exhibit
RI PT
characteristics of both diseases. Where uncertainty exists regarding the correct diagnosis of asthma, COPD or both, this may represent a phenotype known as asthma-COPD overlap syndrome (ACOS).[1-4]
SC
COPD is highly prevalent in the global population of older adults (40 years of age and older) and has been associated with smoking and exposure to environmental tobacco smoke or fumes.[5] COPD is
M AN U
typically characterized by persistent airflow obstruction and chronic inflammation of the airways. Airway inflammation is also seen in asthma; however, there are distinct differences in the type of inflammatory cells seen in these two respiratory diseases. Biopsies reveal that inflammation in COPD is characterized predominantly by increases in CD8+ T-lymphocytes, neutrophils, and macrophages,[6-8] although increases in eosinophils have been observed in sputum at the time of exacerbation.[9] In contrast,
TE D
inflammation in asthma is commonly characterized by increases in CD4+ T-lymphocytes and eosinophils.[10] Airflow obstruction in COPD is defined as a post-bronchodilator measurement of forced
EP
expiratory volume in 1 second/forced vital capacity (FEV1/FVC) <0.70 or lower limit of normal (LLN), where LLN values are identified by selecting the lowest 2.5% of a normally distributed population.[5]
AC C
Asthma is also a chronic obstructive lung disease, but in mild and moderately severe asthma, airflow obstruction responds to treatment with inhaled corticosteroids and bronchodilators and is therefore not persistent and is reversible. Reversibility refers to the reduction of airflow obstruction after bronchodilator administration and is usually measured by comparing pre- and post-bronchodilator FEV1. An FEV1/FVC ratio >0.75–0.80, and a post-bronchodilator FEV1 increase of >12% and >0.200 L over the pre-bronchodilator measurements are usually indicators of asthma.[11] However, applying these definitions to differentiate between severe asthma of middle-age and COPD may lead to uncertainty
4
ACCEPTED MANUSCRIPT
owing to a number of factors. For example, many patients with COPD are responsive to albuterol (reversible) while many patients with a clinical diagnosis of asthma are not reversible on testing and approximately 25% of patients with COPD have never smoked.[12, 13] Furthermore, since the FEV1/FVC
RI PT
ratio declines with age, this definition may lead to over-diagnosis of COPD in the elderly and underdiagnosis in the young to middle-aged, which may be especially relevant when considering the possible diagnosis of nonreversible asthma.[5] In addition, there are some individuals with asthma who have
SC
neutrophilic inflammation that may be related to treatment with inhaled steroids[14] or an intrinsic part of their disease.[15, 16] Finally, the histopathological differentiation of asthma as an eosinophilic disease
M AN U
and COPD as a neutrophilic disease is also in question as the eosinophil is becoming an increasingly important marker of treatment response in COPD,[17, 18] which is possibly a result of a rise in the prevalence of ACOS.
TE D
While a variety of diagnostic criteria have been proposed for identifying ACOS, a specific formal definition of ACOS has yet to be determined. The importance of identifying patients with features of ACOS is underlined by recent work that has demonstrated that patients with ACOS have more
EP
symptoms, exacerbations, hospitalizations, lower health-related quality of life, and higher mortality than patients with COPD or asthma alone.[19-24] Recognizing that there is no generally agreed term or
AC C
definition for ACOS, the Global Initiative for Asthma (GINA) and the Global Initiative for Chronic Obstructive Lung Disease (GOLD) recently developed guidelines to help stimulate further research, to identify and characterize ACOS.[5, 11] Broadly, ACOS can be described as an obstructive airflow condition or a set of clinical characteristics where aspects of both asthma and COPD are present, such as FEV1/FVC <0.70 and evidence of airflow reversibility (post-bronchodilator FEV1 increase of >12% and >0.200 L). Developing diagnostic criteria for ACOS is especially difficult since the clinical presentation of asthma and COPD can appear similar with respect to symptomology and, as noted above, are
5
ACCEPTED MANUSCRIPT
sometimes challenging to differentiate.[1, 25] Whether ACOS is simply the coexistence of asthma and COPD or a distinct phenotype with related fundamental pathogenic mechanisms to asthma and COPD remains to be determined.[26] Furthermore, different studies classify ACOS as a sub-phenotype of
RI PT
COPD,[27, 28] whereas other studies treat ACOS as a phenotype distinct from either COPD or
asthma.[22, 29, 30] Proposed definitions for ACOS vary widely and include (1) patients with COPD who have any diagnosis of current or past asthma, (2) spirometric-based definitions such as patients with
SC
COPD who have consistent significant reversibility (FEV1/FVC <0.70 and a post-bronchodilator FEV1 increase of >12% and >0.200 L over the pre-bronchodilator measurement), or (3) patients with asthma
M AN U
who have persistent airflow obstruction.[3, 31] The absence of ACOS-specific pathological definitions and the variability of ACOS criteria across the literature present challenges for conducting studies to add to the available evidence.[3]
TE D
Aims
ACOS may represent a sizeable portion of the clinical patient population with airflow obstruction; however, the varying ACOS definitions and diagnostic criteria make this population difficult to
EP
characterize. The aim of this literature review was to characterize the prevalence of ACOS and examine
AC C
how different definitions of ACOS affect prevalence estimates, as this has not previously been explored.
Search strategy
Published studies and abstracts were identified through PubMed and Searchlight, respectively, from January 1, 2000 to February 28, 2014. Only studies written in English were retained. Relevant publications found from other references, but not identified in the original search, were added. The following key words were used to identify potentially relevant abstracts in PubMed: asthma, chronic obstructive pulmonary disease, COPD, asthma-chronic obstructive pulmonary disease, fixed airway
6
ACCEPTED MANUSCRIPT
obstruction, asthma COPD overlap, persistent airway obstruction, fixed airway obstruction risk factor, and prevalence. This search strategy produced 180 results, 71 of which were retained for further review. Thirty-one additional relevant papers identified through citations of systematically identified papers
RI PT
were also included in this analysis.
The following key words were used to identify potentially relevant conference abstracts in Searchlight:
SC
asthma, atopic asthma, bronchial asthma, bronchitic asthma, chronic obstructive pulmonary disease, chronic airflow obstruction, chronic obstructive airways disease, chronic obstructive lung disease, COPD,
M AN U
pulmonary disease, chronic obstructive, and overlap. This search strategy produced 23 conference abstracts, 18 of which were retained for further review.
Of all the articles retained for further assessment, 43 were considered relevant and included in this
TE D
review. Studies identified but not included in the review did not contain information on ACOS prevalence. Topics of the excluded articles included dysfunctional breathing, COPD subtypes excluding ACOS, smoking, treatment/management, air trapping, and airway obstruction. Studies that estimated
EP
the population prevalence of ACOS, the prevalence of ACOS in patients with COPD, and the prevalence of ACOS in patients with asthma were summarized. Variations were identified in the study populations
AC C
and in the definitions of asthma, COPD, and ACOS.
7
ACCEPTED MANUSCRIPT
Prevalence of ACOS Estimates of the population prevalence of ACOS ranged from 1.6% to 4.5% from studies in Italy, South Korea, Latin America, and the United States[22, 26, 30, 32, 33] (Table 1, Figure 1). Estimates of the
RI PT
population prevalence of ACOS varied across countries at least in part because of different COPD and asthma diagnostic criteria and different underlying distribution of risk factors, such as smoking, the age
AC C
EP
TE D
M AN U
SC
distribution in the source populations, and study samples.
8
ACCEPTED MANUSCRIPT
Table 1. COPD, asthma, and ACOS prevalence among total populations: differences in asthma definitions, spirometric criteria, COPD
Population Prevalence of COPD-only
Population Prevalence of Asthma-only
Population Prevalence of ACOS
5.3%
5.3%
11.7%
RI PT
definitions, and population demographics
ACOS Definition
Population Demographics (age: years)
Ever self-reported physician diagnosis of asthma and COPD
Age ≥25 General population, US Smokers and nonsmokers
Report of wheezing in the last 12 months and post-bronchodilator increase in FEV1 or FVC of 0.200 L and 12% Self-reported physician diagnosed asthma, ever
Meeting both criteria of asthma and COPD
Age >40 Population in 5 Latin American cities Smokers and nonsmokers
Meeting both criteria of asthma and COPD
Wheezing for past year and FEV1/FVC≥0.70
Wheezing and FEV1/FVC<0.70
Age >40 Population in 5 Latin American cities Smokers and nonsmokers Korean patient population Smokers and nonsmokers
Asthma Definition
2.7%
Self-reported physician diagnosis of COPD, ever
Self-reported physician diagnosed, asthma ever
1.7%
1.8%
FEV1/FVC<0.70
10.6%
8.9%
2.9%
FEV1/FVC<0.70
7.8%
6.1%
2.4%
FEV1/FVC<0.70 and no wheezing in past year
8.2% (7.5–9)
a
1.6% (1.3–2)
a
M AN U
TE D
a
Self-reported physician diagnosis of COPD, ever
Self-reported physician diagnosed asthma, ever
Ever self-reported physician diagnosis of asthma and COPD
Age 20–44 General population, Italy Smokers and nonsmokers
EP
3.3% (2.8–3.8)
SC
COPD Definition
Self-reported physician diagnosed asthma, ever
Ever self-reported physician diagnosis of asthma and COPD
Age 45–64 General population, Italy Smokers and nonsmokers
5.7% (4.7–6.7)a
4.9% (4–5.9)a
2.1% (1.5–2.8)a
13.3% (11.1–15.5)a
2.9 % (1.8–4)a
4.5 %(3.2–5.9)a
Self-reported physician diagnosis of COPD, ever
Self-reported physician diagnosed asthma, ever
Ever self-reported physician diagnosis of asthma and COPD
Age 65–84 General population, Italy Smokers and nonsmokers
5.6%
12.1%
2.4%
Self-reported physician diagnosis of COPD, ever
Self-reported physician diagnosed asthma, ever
Ever physician diagnosis of asthma and COPD
Age ≥18 General population, North Carolina Smokers and nonsmokers
AC C
Self-reported physician diagnosis of COPD, ever
Author, Year Data Study Sample Size Diaz-Guzman et al. 2011[32] NHANES III n = 15,203 Menezes et al. 2014 [22] PLATINO n=5,044 Menezes et al. 2014[22] PLATINO n=5,044 Lee et al. 2013[33] 4th Korean NHANES n=9,369 de Marco et al. 2013[26] GEIRD n=5,163 de Marco et al. 2013[26] GEIRD n=2,167 de Marco et al. 2013[26] GEIRD n=1,030 Pleasants et al. 2014[30] NC BRFSS n=24,073
9
ACCEPTED MANUSCRIPT
ACOS: asthma-COPD overlap syndrome; COPD: chronic obstructive pulmonary disease; FEV1: forced expiratory volume in 1 second; FVC: forced vital capacity; GEIRD: GeneEnvironment Interactions in Respiratory Disease; ICD: International Classification of Diseases; NC BRFSS: North Carolina Behavioral Risk Factor Surveillance System; NHANES:
EP
TE D
M AN U
SC
Prevalence (95% confidence interval).
AC C
a
RI PT
National Health and Nutrition Examination Survey; PLATINO: Latin American Project for the Investigation of Lung Disease.
10
ACCEPTED MANUSCRIPT
ACOS prevalence among patients with COPD or asthma On the basis of the results of the literature review, the prevalence of ACOS among patients with COPD is estimated to range from 12.1% to 55.2% (Table 2)[19-22, 24, 26-28, 30, 33-42] and the prevalence of
RI PT
ACOS among patients with asthma is estimated to range from 13.3% to 61.0% (Table 3).[19, 21, 22, 26, 30, 32, 33, 37, 38, 41, 43-45] The wide variations in prevalence estimates are most likely due to the method by which ACOS cases were classified, but differences in population inclusion criteria, in the data
SC
source, and in the definitions of asthma and COPD including airflow reversibility criteria and spirometric versus clinical diagnoses can also influence the prevalence of ACOS. In general, the prevalence of COPD
M AN U
alone is difficult to accurately estimate due to variations in COPD diagnostic criteria, thus, identifying the
AC C
EP
TE D
ACOS population is even more complicated.[46]
11
ACCEPTED MANUSCRIPT
COPD Definition
Asthma Definition
ACOS Definition
12.1%
FEV1/FVC <0.70
Self-reported physician-diagnosed before the age of 40
Asthma diagnosis among patients with COPD
12.6%
FEV1/FVC <0.70, FEV1<80% of predicted, GOLD Stage II+
Self-reported physician-diagnosed before the age of 40
Asthma diagnosis among patients with COPD
13.0%
FEV1/FVC <0.70, GOLD Stage II+
Self-reported physician-diagnosed before the age of 40
Asthma diagnosis among patients with COPD
13.0%
FEV1/FVC <0.70
Report of wheezing in the last 12 months and post-bronchodilator increase in FEV1 or FVC of 0.200 L and 12%
Asthma diagnosis among patients with COPD
16.7%
FEV1/FVC <0.70
Self-reported physician diagnosis, current
Asthma diagnosis among patients with COPD
22.8%
FEV1/FVC <0.70
Self-report medical-diagnosed asthma, prior
Asthma diagnosis among patients with COPD
23.5%
FEV1/FVC <0.70, FEV1<80% of predicted
Self-reported physician diagnosed, ever
Asthma diagnosis among patients with COPD
23.6%
FEV1/FVC <0.70 and no wheezing in past year
Wheezing and FEV1/FVC<0.70
Korean patient population Smokers and nonsmokers
25.0%
Self-reported physician diagnosis of COPD, ever
Self-reported physician diagnosed asthma, ever
Ever self-reported physician diagnosis of asthma and COPD
26.0%
ICD-10 codes for COPD diagnoses (ICD-10:J41-J44)
ICD-10 codes for asthma diagnoses (ICD-10:J45-J46)
Diagnosis of both asthma and COPD based on ICD-10 codes at different times
Age 65–84 General population, Italy Smokers and nonsmokers Age >34 Persons with a primary and/or secondary diagnosis of asthma and/or COPD based on ICD-10 codes among the entire Finnish
SC
M AN U
TE D
AC C
Wheezing for past year and FEV1/FVC ≥0.70
Demographics of Population with COPD (age: years)
RI PT
Proportion of COPD Patients with ACOS
EP
Table 2. ACOS prevalence among patients with COPD: variation in study definition and details of general population with COPD
Age ≥40 Spain multicenter study Smokers only Age 45–80 US multicenter study Current and former smokers Age 45–80 US multicenter study Current and former smokers Age >40 Population in 5 Latin American cities Smokers and nonsmokers Age 40–80 US, multicenter Among smokers only Age ≥40 Population in 5 Latin American cities Smokers and nonsmokers Age ≥40 Multi-country Never smokers only
Author, Year Data Study Sample Size Izquierdo-Alonso et al. 2013[27] n=331 Hardin et al. 2014[20] COPDGene n=3,570 Hardin et al. 2011[34] COPDGene n=1,059 Menezes et al. 2014[22] PLATINO n=5,044 Putcha et al. 2013[24] SPIROMICS n=581 Talamo et al. 2007[36] PLATINO n=5,303 Lamprecht et al. 2011a[37] BOLD Study n=4,291 Lee et al. 2013[33] th 4 Korean National Health and Nutritional Survey n=9,369 de Marco et al. 2013[26] GEIRD n=1,030 Andersen et al. 2013[19] Hospitalization records n=105,122
12
Proportion of COPD Patients with ACOS
COPD Definition
Asthma Definition
ACOS Definition
26.4%
FEV1/FVC <0.70 or FEV1/FVC <88% of predicted
27.0%
Self-reported physician diagnosis of COPD, ever
Post-bronchodilator increase in FEV1 of ≥12%, a bronchodilator response of ≥15% or diurnal variation of ≥20% in PEF recording, moderate-to-severe bronchial hyperactivity, or a decrease in FEV1 of ≥15% in the exercise test Self-reported physician diagnosed asthma, ever
Ever self-reported physician diagnosis of asthma and COPD
Age 45–64 General population, Italy Smokers and nonsmokers
ICD-9 codes 491, 492, and 496
Experience of asthmatic symptoms (i.e., wheezing, cough, and chest tightness worsening at night) ICD-9 code 493
Diagnosis of COPD and report of experiencing asthmatic symptoms Diagnosis of asthma and COPD within 12 months
Hospital-based patient sample with stable COPD (2007-2009), Japan Age ≥40 US patients with medical and pharmacy benefits
de Marco et al. 2013[26] GEIRD n=2,167 Kitaguchi et al. 2012[39] n=63
27.0%
FEV1/FVC <70% and FEV1 <80%
30.2%
32.9%
FEV1/FVC <0.70 COPD Medical Codes
Current or previous asthma diagnosis
Asthma diagnosis among patients with COPD
33.0%
Self-reported physician diagnosis of COPD, ever
Self-reported physician diagnosed asthma, ever
Ever self-reported physician diagnosis of asthma and COPD
Age ≥40 UK patients with newly diagnosed COPD Smokers and nonsmokers Age 20–44 General population, Italy Smokers and nonsmokers
41.4%
Self-reported physician diagnosis of COPD, ever
Self-reported physician diagnosed, ever
Diagnosis of asthma and COPD ever
43.0%
ICD-9 codes 491, 492, and 496
Diagnosis of asthma and COPD within 30 months
49.0%
FEV1/FVC <0.70 or FEV1/FVC <88% of predicted reference value
54.6%
ICD-10 codes J42–J44, except J430 Use of more than one drug for COPD at least twice per year
TE D
M AN U
SC
RI PT
Author, Year Data Study Sample Size
Diagnosis of both asthma and COPD based on spirometric criteria
Demographics of Population with COPD (age: years) population 134 volunteers Finnish patients with asthma and COPD
EP
ACCEPTED MANUSCRIPT
AC C
ICD-9 code 493
Post-bronchodilator increase in FEV1 of ≥12%, a bronchodilator response of ≥15% or diurnal variation of ≥20% in PEF recording, moderate to severe bronchial hyperreactivity, or a decrease in FEV1 of ≥15% in the exercise test ICD-10 codes J45-J46 Use of more than one drug for asthma at least twice per year
Age ≥18 General population, US Smokers and nonsmokers Age 40–64 US Medicaid patients
Diagnosis of both asthma and COPD based on spirometric criteria
Age 18–75 Finnish hospital-based population, identified from ICD-10 codes J44–J46 Smokers and nonsmokers
Diagnosis of asthma and COPD within 12 months
Age ≥40 Korean patient population
Iwamoto et al. 2014[38] Sample of patients from FinnCADStudy n=134
Blanchette et al. 2009[40] US IHCIS Claims n=24,935 Wurst et al. 2014[35] General practice research database n=7,881 de Marco et al. 2013[26] GEIRD n=5,163 Pleasants et al. 2014[30] NC BRFSS n=24,073 Shaya et al. 2008[41] Medicaid claims n=9,131 Kauppi et al. 2011[21] Medical records n=1,546
Rhee et al. 2014[42] Insurance data n=185,147
13
ACCEPTED MANUSCRIPT
COPD Definition
Asthma Definition
ACOS Definition
Demographics of Population with COPD (age: years)
55.2%
FEV1/FVC <0.70
Post-bronchodilator increase in FEV ≥15%, peak flow variability ≥20% during 1 week of testing, physician diagnosis of asthma in conjunction with current symptoms or inhaler use in the last 12 months
Diagnosis of asthma within last 12 months and current COPD
Age >50 Random population based participants, New Zealand Smokers and nonsmokers
RI PT
Proportion of COPD Patients with ACOS
Author, Year Data Study Sample Size Marsh et al. 2008[28] Wellington respiratory survey subset n=469
SC
ACOS: asthma-COPD overlap syndrome; COPD: chronic obstructive pulmonary disease; BOLD: Burden of Obstructive Lung Disease; COPDGene: Genetic Epidemiology of COPD; FEV1: forced expiratory volume in one second; FVC: forced vital capacity; GEIRD: Gene-Environment Interactions in Respiratory Disease; GOLD, Global Initiative for Chronic
M AN U
Obstructive Lung Disease; ICD: International Classification of Diseases; IHCIS: Integrated Healthcare Information Services; NC BRFSS: North Carolina Behavioral Risk Factor Surveillance System; PEF: peak expiratory flow; PLATINO: Latin American Project for the Investigation of Lung Disease; SPIROMICS: Subpopulations and intermediate outcome measures in COPD study. a
Lamprecht et al 2011[37]: Of the patients with a physician diagnosis of asthma ‘ever’, 30.4% and 17.7% were considered GOLD stage II+ (FEV1/FVC <0.7 and FEV1 <80% predicted)
AC C
EP
TE D
and GOLD stage I (FEV1/FVC <0.7 and FEV1 ≥80% predicted), respectively.
14
ACCEPTED MANUSCRIPT
Table 3. ACOS prevalence among patients with asthma: variation in study definition and details of general population of patients with asthma COPD Definition
Asthma Definition
ACOS Definition
Demographics of Population with Asthma (age: years)
13.3%
FEV1/FVC <0.70 and FEV1 <80% predicted
Self-reported physician diagnosed, asthma ever
16.0%
Self-reported physician diagnosis of COPD, ever
Self-reported physician diagnosed asthma, ever
Evidence of obstruction (FEV/FVC <0.70 and FEV1 <80% predicted, and ever selfreported physician diagnosis of asthma) Ever self-reported physician diagnosis of asthma and COPD
17.0%
FEV1/FVC <0.70 or FEV1/FVC <88% of predicted
18.0%
FEV1/FVC <0.70
Post-bronchodilator increase in FEV1 of ≥12%, a bronchodilator response of ≥15% or diurnal variation of ≥20% in PEF recording, moderate to severe bronchial hyper reactivity, or a decrease in FEV1 of ≥15% in the exercise test. Physician diagnosis of asthma
24.9%
FEV1/FVC <0.70, FEV1<80% of predicted
27.5%
Airway obstruction where FEV1/FVC ratio < predicted LLN
27.8%
Self-reported physician diagnosis of COPD, ever
28.2%
FEV1/FVC <0.70 and no wheezing in past year
Wheezing for past year and FEV1/FVC≥0.70
Wheezing and FEV1/FVC<0.70
29.0%
FEV1/FVC <0.70
Physician diagnosis of asthma
Meeting the criteria for asthma and COPD
Patients with asthma ages 40–60 Former or nonsmokers
29.8%
ICD-10 codes for COPD
ICD-10 codes for asthma
Diagnosis of both asthma and
Age ≥34
RI PT
Proportion of Asthma Patients with ACOS
Age ≥25 General Population, US Smokers and nonsmokers
Author, Year Data Study Sample Size Diaz-Guzman et al. 2011[32] NHANES III n=15,203 de Marco et al. 2013[26] GEIRD n=5,163 Kauppi et al. 2011[21] Hospital medical records n=1,546
Meeting the criteria for asthma and COPD
Patients with asthma ages 18–40 Former or nonsmokers
Self-reported physician diagnosed, ever
Meeting the criteria for asthma and COPD
Diagnosis of bronchial asthma between 2009 and 2011
Meeting the criteria for asthma and COPD
Age ≥40 Multi-country Never smokers only Age 18–88 years Sample of US patients with spirometry data
Self-reported physician diagnosed asthma, ever
Ever physician diagnosis of asthma and COPD
Kalberg et al. 2005[43] Clinical trials database (1998-2002) n=6,497 Lamprecht et al. 2011[37] BOLD study n=4,291 Banga et al. 2013[44] Tertiary care center n=1,482 Pleasants et al. 2014[30] NC BRFSS n=24,073
SC
Age 20–44 General population, Italy Smokers and nonsmokers Age 18–75 Finnish hospital-based study population identified from ICD-10 codes J44–J46
AC C
EP
TE D
M AN U
Diagnosis of both asthma and COPD based on spirometric criteria
Ages ≥18 Smokers and nonsmokers General population, North Carolina, US Korean patient population Smokers and nonsmokers
Lee JH et al. 2013[33] 4th Korean NHANES n=9,369 Kalberg et al. 2005[43] Clinical trials database (1998-2002) n=3,346 Andersen et al. 2013[19]
15
ACCEPTED MANUSCRIPT
diagnoses (ICD-10:J41-J44)
diagnoses (ICD-10:J45-J46)
COPD based on ICD-10 codes at different times
30.0%
Self-reported physician diagnosis of COPD, ever
Self-reported physician diagnosed asthma, ever
Ever self-reported physician diagnosis of asthma and COPD
30.4%
FEV1/FVC <0.70 or FEV1/FVC <88% of predicted
Diagnosis of both asthma and COPD based on spirometric criteria
43.0%
FEV1/FVC <0.70
Post-bronchodilator increase in FEV1 of ≥12%, a bronchodilator response of ≥15% or diurnal variation of ≥20% in PEF recording, moderate-to-severe bronchial hyperactivity, or a decrease in FEV1 of ≥15% in the exercise test Physician diagnosis of asthma
Meeting the criteria for asthma and COPD
Patients with asthma age >60 US population Former or nonsmokers
45.9%
ICD-9 Codes 491, 492, and 496
ICD-9 Code 493
Diagnosis of asthma and COPD within 30 months
Age 40–64 US Medicaid patients
46.0%
Post-bronchodilator FEV1 or FEV1/FVC <75% of predicted with a TLC>75% of predicted
Meeting criteria for asthma and COPD
Average age 55.3 (SD 11.7) Former or nonsmokers
51.4%
FEV1/FVC <0.70
Wheezing history, FEV1/FVC <0.70 and post-bronchodilator FEV1 increase of 12%, PC20, diurnal PEF variation/tapering med Report of wheezing in the last 12 months and postbronchodilator increase in FEV1 or FVC of 0.200 L and 12%
COPD diagnosis among patients with asthma
61.0%
Self-reported physician diagnosis of COPD, ever
Age >40 Population in 5 Latin American cities Smokers and nonsmokers Age 65–84 General population, Italy Smokers and nonsmokers
RI PT
SC
M AN U
TE D
EP
AC C
Self-reported physician diagnosed asthma, ever
Persons with a primary and/or secondary diagnosis of asthma and/or COPD based on ICD-10 codes among the entire Finnish population Age 45–64 General population, Italy Smokers and nonsmokers 134 volunteers Finnish patients with asthma and COPD
Ever self-reported physician diagnosis of asthma and COPD
Hospitalization records n=105,122
de Marco et al. 2013[26] GEIRD n=2,167 Iwamoto et al. 2014[38] Sample of patients from FinnCADStudy n=134
Kalberg et al. 2005[43] Clinical trials database (1998-2002) n=601 Shaya et al. 2008[41] Medicaid claims n=9,131 Wagener, 2013[45] U-BIOPRED study cohort n=148 Menezes et al 2014[22] PLATINO n=5,044 de Marco et al. 2013[26] GEIRD n=1,030
ACOS: asthma-COPD overlap syndrome; COPD: chronic obstructive pulmonary disease; BOLD: Burden of Obstructive Lung Disease; FEV1: forced expiratory volume in one second; FVC: forced vital capacity; GEIRD: Gene-Environment Interactions in Respiratory Disease; ICD: International Classification of Diseases; LLN, lower limit of normal; NC BRFSS: North Carolina Behavioral Risk Factor Surveillance System; NHANES: National Health and Nutrition Examination Survey; PEF: peak expiratory flow; PLATINO: Latin American Project for the Investigation of Lung Disease; U-BIOPRED: Unbiased BIOmarkers in PREDiction of respiratory outcomes; SD: standard deviation.
16
ACCEPTED MANUSCRIPT
Spirometry-based versus clinical or symptom-based diagnoses In some studies an asthma diagnosis is based on spirometry-confirmed reversibility. The investigators from the Latin American Project for the Investigation of Lung Disease (PLATINO) study in Latin America
RI PT
noted that two different definitions of asthma resulted in varied ACOS prevalence estimates in the same population. The prevalence of ACOS based on spirometry and the presence of reversibility was
SC
estimated to be 1.8% compared with 2.9% using physician diagnosis only.[22]
Spirometry-defined asthma
M AN U
The variation in ACOS prevalence estimates among patients with COPD is highlighted by the reversibility criteria in patients with COPD set forth in different medical systems internationally. In a Finnish medical records cohort study, the authors stated that reversibility seen in patients with COPD is considered to be asthma, so definitions of COPD and asthma among the Finnish population will sometimes differ
TE D
compared with other populations.[19, 21] Analysis of medical records and hospitalization records in Finland estimated that between 30% and 49% of patients with COPD also have ACOS[19, 21]; here diagnostic criteria of asthma appeared to be based on spirometric measurements, including spirometric
EP
criteria for reversibility.[19]
AC C
In other countries where reversibility among patients with COPD is not necessarily regarded as asthma, it is difficult to distinguish between COPD with asthma and COPD alone with airway reversibility. If ACOS is defined on the basis of obstruction and reversibility, this phenotype may include patients with COPD who have reversibility but no other asthma features. Evidence now suggests that 12–42% of patients with COPD may demonstrate post-bronchodilator reversibility; however, reversibility may vary over time and thus it may not be a stable phenotype.[47, 48]
17
ACCEPTED MANUSCRIPT
Spirometry-defined COPD When using spirometric criteria for COPD classification, the estimates of ACOS among patients with asthma ranged from 25.0% to 51.4% (Table 3). The diagnostic criteria for COPD also influence estimates
RI PT
of both overall COPD prevalence and ACOS prevalence. Marsh et al.[28] estimated an ACOS prevalence of 55.2% in patients with COPD aged 50 years and older using spirometric criteria. The authors
estimated the prevalence of ACOS using two different criteria for COPD, (1) FEV1/FVC <0.70 and (2)
SC
FEV1/FVC less than the LLN.[28] They concluded that when less than the LLN criterion was used in COPD diagnosis rather than the FEV1/FVC <0.70 criterion, one-third fewer patients presented with COPD.
M AN U
Furthermore, Marsh et al. concluded that fewer patients with COPD also presented with the ‘asthma phenotype’ (reversibility, peak flow variability, or physician diagnosis in addition to COPD) when the LLN criterion was used instead of the FEV1/FVC <0.70 criterion.[28]
TE D
A Norwegian study compared spirometric results against physician diagnoses in 376 patients aged 40 years or older.[29] The study reported that among patients previously diagnosed with only asthma by a physician, 17.1% could be re-diagnosed with COPD because of obstruction using spirometric criteria.[29]
EP
Diaz-Guzman et al. reported that 13.3% of patients with asthma in the US National Health and Nutrition Examination Survey (NHANES) III had evidence of obstruction and thus may also be classified as COPD or
AC C
ACOS.[32]
Self-reported physician diagnosis In the Continuing to Confront COPD survey, international prevalence estimates of ACOS among patients with COPD in 12 countries ranged from 12% to 54%.[49] Asthma was defined as self-reported physician diagnosis and COPD as self-reported physician diagnosis or symptom-based chronic bronchitis. The
18
ACCEPTED MANUSCRIPT
broad range of estimates highlights the differences in underlying asthma population characteristics, diagnosis practices, and healthcare systems across countries.
RI PT
ACOS defined in claims data
ACOS prevalence estimates are generally higher using claims data than those using spirometric
definitions or self-reported physician diagnoses. It is unknown whether the medical codes are based on
SC
reversibility criteria or other criteria including symptoms such as wheezing. An analysis of US claims data using International Classification of Disease (ICD)-9 codes estimated that 30–43% of the population over
M AN U
40 years of age with a diagnosis code for COPD also had an asthma diagnosis code within 12–30 months.[40, 41] Estimates from claims data depend on the duration of time allowed between COPD and asthma diagnostic codes and whether more than one asthma code is required, as some patients may receive an asthma diagnosis before their COPD diagnosis and thus could be misclassified as having both
TE D
diseases. Patients with more utilizations (for any reason) also have more opportunities to receive dual diagnoses. Additionally, as was previously mentioned, it may not be clear how the physician diagnosed
EP
COPD or asthma (i.e., if spirometry was used and if reversibility was considered).
Asthma definition: past versus current asthma
AC C
Another source of variation in ACOS prevalence estimates is the definition of asthma with regard to timing of the diagnosis, for example, ‘ever’ diagnosed with asthma, ‘prior’ asthma diagnosis (usually before the age of 40), or ‘current’ asthma diagnosis. The ACOS literature lacks consensus on whether the ACOS population only includes those with the concurrence of existing asthma and clinical characteristics of COPD or if the ACOS population includes patients with COPD and a history of asthma. The GINA guidelines include both.[11]
19
ACCEPTED MANUSCRIPT
In the subpopulations and intermediate outcome measures in the COPD study (SPIROMICS), restricting asthma diagnostic criteria to a current asthma diagnosis resulted in an ACOS prevalence estimate of 16.7% among smokers over the age of 40 with COPD.[24] A study of self-reported, physician-diagnosed
RI PT
asthma among smokers before the age of 40 with no current diagnosis reported a prevalence of 12.1% with COPD.[27] A UK study of ACOS that defined asthma as a current or former diagnosis reported a prevalence of 32.9% in smokers and nonsmokers with COPD.[35] All three of these studies used
SC
spirometric criteria for the evaluation of COPD, and used either self-reported physician diagnosis or medical codes for asthma identification. Comparing these studies, the inclusion of current or past
M AN U
asthma (i.e., ‘ever’ asthma) resulted in higher estimates of ACOS prevalence than studies with asthma defined as ‘current’ asthma or as only ‘past’ asthma.
Ford and Mannino [50] utilized data from the US NHANES III (1988–1994) to estimate the prevalence of
TE D
‘ever’ asthma (7.8%) and ‘current’ asthma (5.2%) in people aged 20–74 years. Taken together, these estimates suggest that 67% of people to ever report asthma will currently report asthma; applying that proportion to the ACOS population prevalence estimate of 2.7% from Diaz-Guzman et al., who defined
EP
asthma as ‘ever’ reported, the prevalence of ACOS decreases to 1.8%, demonstrating the change in
AC C
estimate when using ‘current’ versus ‘ever’ asthma.[32]
Comparisons among definitions Differences between studies such as sample size, source population, and age inclusion criteria make it difficult to conclude that using spirometric diagnosis is solely responsible for the difference in estimates.[28, 30] Generally, there are four different classification schemes for defining ACOS: (1) a reported physician diagnosis of asthma and a reported physician diagnosis of COPD at any point in a patient’s life, (2) a reported physician diagnosis of asthma and spirometry-defined COPD, (3) both
20
ACCEPTED MANUSCRIPT
spirometry-defined asthma and spirometry-defined COPD, and (4) ICD codes to determine both asthma and COPD diagnoses among patients. There is considerable variability in prevalence estimates of ACOS
with asthma[19, 21, 22, 26, 30, 32, 33, 37, 38, 41, 43-45] (Figure 3).
RI PT
for each of the definitions in patients with COPD[19-22, 24, 26-28, 30, 33-42] (Figure 2) and in patients
The ACOS prevalence estimate ranged from 25.0% to 41.4% when ACOS was defined as a self-report of a
SC
physician diagnosis of both asthma and COPD (1).[26, 30] When ACOS was defined as a reported
physician diagnosis of asthma and spirometry-defined COPD, the ACOS prevalence estimates in patients
M AN U
with COPD ranged from 12.1% to 32.9% (2).[20, 24, 27, 34-37, 39] Prevalence estimates for ACOS in patients with COPD from varying geographic regions using spirometry-defined asthma and spirometrydefined COPD (3) produced the widest range of all subgroups at 13.0–55.2%.[21, 22, 28, 33, 38] In studies that used ICD codes for both the COPD and asthma diagnoses (4),[19, 40-42] the range of ACOS
TE D
prevalence estimates in patients with COPD was generally higher than when ACOS was defined as selfreported asthma and spirometry-defined COPD (26.0–54.6% versus 12.1–32.9%, respectively, Figure
EP
2).[20, 24, 27, 34-37, 39]
The same high variability also exists when examining ACOS prevalence among patients with asthma,
AC C
again, with variability depending on the asthma and COPD definitions used (Table 3, Figure 3). When ACOS was defined by the combination of self-reported asthma and spirometry-defined COPD (2), the ACOS prevalence estimates ranged from 13.3% to 43.0%.[32, 37, 43, 44] When defining ACOS as a selfreported physician diagnosis of COPD and a self-reported physician asthma diagnosis (1), the ACOS prevalence estimates had a wider range (16.0–61.0%)[26, 30] than the ACOS estimates using spirometric criteria for both (3) asthma and COPD (17.0–51.4%).[21, 22, 33, 38, 45] Two disparate prevalence
21
ACCEPTED MANUSCRIPT
estimates (29.8% and 45.9%) of ACOS among patients with asthma were obtained from the studies that used ICD codes (4).[19, 41]
RI PT
Differences in ACOS prevalence estimates due to age of the study cohort
As ACOS is more prevalent in the population over 40 years of age and has a relatively low prevalence in younger age groups, the age of the study population can affect the estimated ACOS prevalence. For
SC
example, in a study from Finland 0.4% of patients with ACOS were ≤40 years of age and 78.2% of
patients with ACOS were over the age of 55 years.[21] In a second study performed in Italy, 1.6% of
M AN U
patients aged 20–44 and 2.1% of patients aged 45–64 years had ACOS, compared with 4.5% of patients aged 65–84 years.[26] Thus, inclusion of younger patients in a study can reduce the estimated ACOS prevalence.
TE D
Differences in estimates due to inclusion of smoking
Despite differences in COPD definitions across studies (SPIROMICS, Genetic Epidemiology of COPD [COPDGene] and Burden of Obstructive Lung Disease [BOLD]), smoking seems to be associated with an
EP
increase in ACOS prevalence estimates. The COPDGene and SPIROMICS studies examined the ACOS prevalence among smokers with COPD and found the prevalence of ACOS to be 12.6% and 16.7%,
AC C
respectively.[20, 24] However, studies estimating the prevalence of ACOS in nonsmokers are limited. Nonsmokers and never smokers with COPD may represent an important component of the ACOS population and provide information on ACOS population characteristics and risk factors. Nonsmokers (fewer than 10 pack-years) who cannot achieve adequate asthma control may often represent a portion of patients with ACOS, as they are likely to have undiagnosed COPD symptoms.[3, 51] Furthermore, COPD among never smokers is often under-diagnosed, in part because there is little understanding of the risk factors of COPD among never smokers.[13]
22
ACCEPTED MANUSCRIPT
Target populations for future research In order to understand the phenotypes better of asthma, COPD, and ACOS, future studies should place emphasis on nonsmokers with COPD and assess whether ACOS should be defined as COPD with a history
RI PT
of asthma, COPD concurrent with asthma, or both. It would also be clinically important to determine if smokers who receive an early diagnosis of asthma (before the age of 35 years) have a greater risk of developing COPD than nonsmokers with an early diagnosis of asthma. Furthermore, development of
SC
distinct phenotypes and biomarkers should be prioritized not only to improve the diagnosis and
definition of ACOS, but also to understand the pathogenesis of ACOS. Our review of ACOS literature
M AN U
shows that patients may have identical physiology in terms of FEV1/FVC ratios and reversibility, but these patients may have very different underlying pathology or different treatment responses to medication. In order to make progress in the treatment and prevention of ACOS, we need to move from an emphasis on using physiology to predict treatment responses to understanding the underlying
need to be established.
EP
Conclusions
TE D
pathology of the disease. Additionally, comprehensive therapeutic interventions for patients with ACOS
Efforts to estimate the prevalence of ACOS on the basis of physiologic measures, such as FEV1, FEV1/FVC
AC C
ratio and degree of reversibility, are misleading because these criteria do not always clearly separate asthma and COPD, and therefore, do not help in defining the ACOS population. We recommend that at a minimum, future investigations on ACOS include a detailed smoking history, the basis of the diagnosis of asthma, the degree of reversibility and whether this was measured on or off treatment so that a uniform dataset is available to provide a basis for comparison of results among studies and guide decision making on the most appropriate criteria for defining ACOS.
23
ACCEPTED MANUSCRIPT
Role of the Funding Source The work presented here, including the conduct of the literature review and analysis, was supported by
RI PT
GSK.
Acknowledgments
Editorial support (provided by Joann Hettasch, PhD, and Natasha Thomas, PhD, Fishawack Indicia Ltd)
SC
was funded by GSK.
M AN U
Author Contributions
All authors contributed to the writing of the paper and participated in the review and interpretation of
AC C
EP
TE D
the data. All authors read and approved the final manuscript.
24
ACCEPTED MANUSCRIPT
References [1]
B. J. Carolan, E. R. Sutherland, Clinical phenotypes of chronic obstructive pulmonary disease and
asthma: recent advances, J. Allergy. Clin. Immunol. 131 (2013) 627. S. R. Kim, Y. K. Rhee, Overlap between asthma and COPD: where the two diseases converge,
Allergy. Asthma. Immunol. Res. 2 (2010) 209. [3]
RI PT
[2]
S. Louie, A. A. Zeki, M. Schivo et al, The asthma-chronic obstructive pulmonary disease overlap
[4]
SC
syndrome: pharmacotherapeutic considerations, Expert. Rev. Clin. Pharmacol. 6 (2013) 197.
M. O. Nakawah, C. Hawkins, F. Barbandi, Asthma, chronic obstructive pulmonary disease
[5]
M AN U
(COPD), and the overlap syndrome, J. Am. Board. Fam. Med. 26 (2013) 470.
Global Initiative for Chronic Obstructive Lung Disease (GOLD), Global strategy for the diagnosis,
management, and prevention of chronic obstructive pulmonary disease, 2014, Available at http://www.goldcopd.org/uploads/users/files/GOLD_Report_2014_Jan23.pdf . T. C. O'Shaughnessy, T. W. Ansari, N. C. Barnes et al, Inflammation in bronchial biopsies of
TE D
[6]
subjects with chronic bronchitis: inverse relationship of CD8+ T lymphocytes with FEV1, Am. J. Respir. Crit. Care. Med. 155 (1997) 852.
M. Saetta, A. Di Stefano, G. Turato et al, CD8+ T-lymphocytes in peripheral airways of smokers
EP
[7]
with chronic obstructive pulmonary disease, Am. J. Respir. Crit. Care. Med. 157 (1998) 822. D. Stanescu, A. Sanna, C. Veriter et al, Airways obstruction, chronic expectoration, and rapid
AC C
[8]
decline of FEV1 in smokers are associated with increased levels of sputum neutrophils, Thorax. 51 (1996) 267. [9]
M. Bafadhel, S. McKenna, S. Terry et al, Acute exacerbations of chronic obstructive pulmonary
disease: identification of biologic clusters and their biomarkers, Am. J. Respir. Crit. Care. Med. 184 (2011) 662.
25
ACCEPTED MANUSCRIPT
[10]
L. M. Fabbri, M. Romagnoli, L. Corbetta et al, Differences in airway inflammation in patients with
fixed airflow obstruction due to asthma or chronic obstructive pulmonary disease, Am. J. Respir. Crit. Care. Med. 167 (2003) 418. Global Initiative for Asthma, From the Global Strategy for Asthma Management and Prevention,
RI PT
[11]
2015, Available at http://www.ginasthma.org/local/uploads/files/GINA_Report_2015.pdf. [12]
C. E. Behrendt, Mild and moderate-to-severe COPD in nonsmokers: distinct demographic
[13]
SC
profiles, Chest. 128 (2005) 1239.
S. S. Salvi and P. J. Barnes, Chronic obstructive pulmonary disease in non-smokers, Lancet. 374
[14]
D. C. Cowan, J. O. Cowan, R. Palmay et al, Effects of steroid therapy on inflammatory cell
subtypes in asthma, Thorax. 65 (2010) 384. [15]
M AN U
(2009) 733.
P. G. Gibson, J. L. Simpson, N. Saltos, Heterogeneity of airway inflammation in persistent asthma
[16]
TE D
: evidence of neutrophilic inflammation and increased sputum interleukin-8, Chest. 119 (2001) 1329. A. Jatakanon, C. Uasuf, W. Maziak et al., Neutrophilic inflammation in severe persistent asthma,
Am. J. Respir. Crit. Care. Med. 160 (1999) 1532.
C. E. Brightling, S. McKenna, B. Hargadon et al, Sputum eosinophilia and the short term response
EP
[17]
to inhaled mometasone in chronic obstructive pulmonary disease, Thorax. 60 (2005) 193. R. Leigh, M. M. Pizzichini, M. M. Morris et al, Stable COPD: predicting benefit from high-dose
AC C
[18]
inhaled corticosteroid treatment, Eur. Respir. J. 27 (2006) 964. [19]
H. Andersen, P. Lampela, A. Nevanlinna et al, High hospital burden in overlap syndrome of
asthma and COPD, Clin. Respir. J. 7 (2013) 342. [20]
M. Hardin, M. Cho, M. L. McDonald et al, The clinical and genetic features of COPD-asthma
overlap syndrome, Eur. Respir. J. 44 (2014) 341.
26
ACCEPTED MANUSCRIPT
[21]
P. Kauppi, H. Kupiainen, A. Lindqvist et al, Overlap syndrome of asthma and COPD predicts low
quality of life, J. Asthma. 48 (2011) 279. [22]
A. M. Menezes, M. Montes de Oca, R. Perez-Padilla et al, Increased risk of exacerbation and
[23]
RI PT
hospitalization in subjects with an overlap phenotype: COPD-asthma, Chest. 145 (2014) 297.
M. Miravitlles, J. B. Soriano, J. Ancochea et al., Characterisation of the overlap COPD-asthma
phenotype. Focus on physical activity and health status, Respir. Med. 107 (2013) 1053.
N. Putcha, M. B. Drummond, C. H. Martinez et al, Self-reported asthma is associated with worse
SC
[24]
health outcomes in COPD, Am. J. Respir. Crit. Care. Med. (2013) A3947.
A. A. Zeki, M. Schivo, A. Chan et al., The asthma-COPD overlap syndrome: a common clinical
M AN U
[25]
problem in the elderly, J. Allergy. 2011 (2011) 861926. [26]
R. de Marco, G. Pesce, A. Marcon et al, The coexistence of asthma and chronic obstructive
pulmonary disease (COPD): prevalence and risk factors in young, middle-aged and elderly people from
[27]
TE D
the general population, PloS. One. 8 (2013) e62985.
J. L. Izquierdo-Alonso, J. M. Rodriguez-Gonzalezmoro, P. de Lucas-Ramos et al, Prevalence and
characteristics of three clinical phenotypes of chronic obstructive pulmonary disease (COPD), Respir.
[28]
EP
Med. 107 (2013) 724.
S. E. Marsh, J. Travers, M. Weatherall et al, Proportional classifications of COPD phenotypes,
[29]
AC C
Thorax. 63 (2008) 761.
H. Melbye, E. Drivenes, L. G. Dalbak et al, Asthma, chronic obstructive pulmonary disease, or
both? Diagnostic labeling and spirometry in primary care patients aged 40 years or more, Int. J .COPD. 6 (2011) 597. [30]
R. A. Pleasants, J. A. Ohar, J. B. Croft et al, Chronic obstructive pulmonary disease and asthma-
patient characteristics and health impairment, COPD. 11 (2014) 256.
27
ACCEPTED MANUSCRIPT
[31]
J. J. Soler-Cataluna, B. Cosio, J. L. Izquierdo et al, Consensus document on the overlap phenotype
COPD-asthma in COPD, Arch. Bronconeumol. 48 (2012) 331. E. Diaz-Guzman, M. Khosravi, D. M. Mannino, Asthma, chronic obstructive pulmonary disease,
and mortality in the U.S. population, COPD. 8 (2011) 400. [33]
K. A. Kong, J. H. Lee, J. W. Chung et al, Health-related quality of life in airway diseases, Am. J.
Respir. Crit. Care. Med. 187 (2013) A5554.
M. Hardin, E. K. Silverman, R. G. Barr et al, The clinical features of the overlap between COPD
SC
[34]
and asthma, Respir. Res. 12 (2011) 127.
K. E. Wurst, A. Shukla, H. Muellerova et al, Respiratory pharmacotherapy use in patients newly
M AN U
[35]
RI PT
[32]
diagnosed with chronic obstructive pulmonary disease in a primary care setting in the UK: a retrospective cohort study, COPD. 11 (2014) 521. [36]
C. Talamo, M. M. de Oca, R. Halbert et al., Diagnostic labeling of COPD in five Latin American
[37]
TE D
cities, Chest. 131 (2007) 60.
B. Lamprecht, M. A. McBurnie, W. M. Vollmer et al, COPD in never smokers: results from the
population-based burden of obstructive lung disease study, Chest. 139 (2011) 752. H. Iwamoto, J. Gao, J. Koskela et al, Differences in plasma and sputum biomarkers between
EP
[38]
COPD and COPD-asthma overlap, Eur. Respir. J. 43 (2014) 421. Y. Kitaguchi, Y. Komatsu, K. Fujimoto et al, Sputum eosinophilia can predict responsiveness to
AC C
[39]
inhaled corticosteroid treatment in patients with overlap syndrome of COPD and asthma, Int. J. COPD. 7 (2012) 283. [40]
C. M. Blanchette, M. Broder, C. Ory et al., Cost and utilization of COPD and asthma among
insured adults in the US, Curr. Med. Res. Opin. 25 (2009) 1385. [41]
F. T. Shaya, D. Dongyi, M. O. Akazawa et al., Burden of concomitant asthma and COPD in a
medicaid population, Chest. 134 (2008) 14.
28
ACCEPTED MANUSCRIPT
[42]
C. K. Rhee, H. K. Yoon, K. H. Yoo et al., Medical utilization and cost in patients with overlap
syndrome of chronic obstructive pulmonary disease and asthma, COPD. 11 (2014) 163. [43]
C. Kalberg, W. Sense, K. Knobil, Fixed airway obstruction in patients with asthma: data from a
[44]
RI PT
large clinical trials database, Proc. Am. Thor. Soc. 2 (abstracts issue) (2005) B95.
A. Banga, K. McCarthy, B.M. Pichurko, Predictors of airway obstruction on spirometry among
patients with bronchial asthma presenting to a tertiary care center, Am. J. Respir. Crit. Care. Med187
[45]
A. H. Wagener, D. Gibeon, X. Yang et al, Associated factors persistent airflow limitation in
M AN U
asthma in U-BIOPRED Eur, Respir. J. 42 (Suppl 52) (2013) P3142. [46]
J. Chang, Z. Mosenifar, Differentiating COPD from asthma in clinical practice, J. Intensive. Care.
Med. 22 (2007) 300. [47]
SC
(2013) A5009.
P. M. Calverley, P. S. Burge, S. Spencer et al, Bronchodilator reversibility testing in chronic
[48]
TE D
obstructive pulmonary disease, Thorax. 58 (2003) 659.
P. Albert, A. Agusti, L. Edwards et al, Bronchodilator responsiveness as a phenotypic
characteristic of established chronic obstructive pulmonary disease, Thorax. 67 (2012) 701. S. H. Landis, H. Muellerova, D. M. Mannino et al, Continuing to Confront COPD International
EP
[49]
Patient Survey: methods, COPD prevalence, and disease burden in 2012-2013, Int. J. COPD. 9 (2014) 597. E. S. Ford, D. M. Mannino, Time trends in obesity among adults with asthma in the United
AC C
[50]
States: findings from three national survey, J. Asthma. 42 (2005) 91. [51]
E. D. Bateman, C. Feldman, J. O'Brien et al, Guideline for the management of chronic obstructive
pulmonary disease (COPD): 2004 revision S. Afr. Med. J. 94 (2004) 559.
29
ACCEPTED MANUSCRIPT
Figure legends
Figure 1. COPD, asthma, and ACOS prevalence among total populations (corresponding to Table 1)
RI PT
ACOS: asthma-COPD overlap syndrome; COPD: chronic obstructive pulmonary disease. Figure 2. ACOS prevalence among patients with COPD (corresponding to Table 2)
Asthma - Diagnosis/COPD - Spirometry: studies with a reported physician diagnosis of asthma and
spirometry defined COPD; Asthma - Spirometry/COPD - Spirometry: studies with spirometry defined
SC
asthma and spirometry defined COPD; Asthma - Diagnosis/COPD - Diagnosis: studies with a reported
M AN U
physician diagnosis of asthma and a reported physician diagnosis of COPD at any point in a patient’s life; Asthma - ICD Code/COPD - ICD Code: studies with ICD code classification for both asthma and COPD. ACOS: asthma-COPD overlap syndrome; COPD: chronic obstructive pulmonary disease; ICD, International Classification of Diseases.
Figure 3. ACOS prevalence among patients with asthma (corresponding to Table 3)
TE D
Asthma - Diagnosis/COPD - Spirometry: studies with a reported physician diagnosis of asthma and spirometry defined COPD; Asthma - Spirometry/COPD - Spirometry: studies with spirometry defined asthma and spirometry defined COPD; Asthma - Diagnosis/COPD - Diagnosis: studies with a reported
EP
physician diagnosis of asthma and a reported physician diagnosis of COPD at any point in a patient’s life; Asthma - ICD Code/COPD - ICD Code: studies with ICD code classification for both asthma and COPD.
AC C
ACOS: asthma - COPD overlap syndrome; COPD: chronic obstructive pulmonary disease; ICD, International Classification of Diseases.
30
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
11.7
12.0
13.3 12.1
EP
14.0
10.6
AC C
10.0
8.0
6.0
8.2 7.8
6.1 5.7
5.6
5.3 5.3 4.9 4.5
4.0
3.3
2.9
2.7
2.9 2.4
2.0
2.4
2.1
1.7 1.8
1.6
Diaz-Guzman et al. 2011
Menezes et al. 2014
Menezes et al. 2014
Lee et al. 2013
de Marco et al. 2013
de Marco et al. 2013
de Marco et al. 2013
ACOS
COPD-only
Asthma-only
ACOS
Asthma-only
COPD-only
ACOS
Asthma-only
COPD-only
ACOS
COPD-only
Asthma-only
ACOS
Asthma-only
COPD-only
ACOS
Asthma-only
COPD-only
ACOS
Asthma-only
COPD-only
ACOS
Asthma-only
0.0
COPD-only
Proportion (%)
8.9
Pleasants et al. 2014
Izq
ui e
Al
o-
on
so et Ha al . rdi 20 ne 13 ta Ha l. 2 rdi 01 ne 4 Me ta ne l. ze 20 se 11 ta Pu l. 20 tch 14 ae ta Ta l . lam 20 13 oe La t mp al. rec 20 07 ht et al. Le 20 ee 11 de ta Ma l . 20 rco 13 et An al . de 20 rse 13 ne Iw t a am l. 2 oto 01 3 de et al . Ma 20 rco 14 et Ki al. tag uc 20 hi 13 Bla et al . nc he 20 tte 12 et a l. 2 W urs 00 te 9 de ta Ma l. 2 rco 01 4 et Ple al . as 20 an 13 ts et al. Sh 20 ay 14 ae t al. Ka up 20 pi 08 et a l. 2 Rh 01 ee 1 et a Ma l. 2 rsh 01 4 et al. 20 08
rd
Proportion (%) 50.0
40.0
EP
60.0
30.0
22.8% 23.5% 23.6%
AC C
Asthma - Diagnosis / COPD - Spirometry
Asthma - Spirometry / COPD - Spirometry
Asthma - Diagnosis / COPD - Diagnosis
Asthma - ICD Code / COPD - ICD Code
20.0
12.1% 12.6%
TE D M AN U SC
RI PT
ACCEPTED MANUSCRIPT
54.6% 55.2%
49.0%
41.4% 43.0%
32.9% 33.0%
26.0% 26.4% 27.0% 27.0%
30.2%
25.0%
16.7%
13.0% 13.0%
10.0
0.0
Di
ne ta de l. 2 Ma 01 rco 1 et al. 20 Ka 13 up pi et al. Ka 20 l be 11 rg et La al . mp 20 rec 05 ht et al. Ba 20 ng 11 ae ta Ple l. as 20 an 13 ts et al. 20 14 Le ee ta l. 2 Ka 01 l be 3 rg et al . An 20 de 05 rse ne ta de l. 2 Ma 01 rco 3 et al . Iw 20 am 13 oto et al. Ka 20 l be 14 rg et al . 20 Sh 05 ay ae ta W l. 2 ag 00 en 8 er et Me al. 20 ne 13 ze s et al . de Ma 20 14 rco et al. 20 13
ma
-G uz
az
Proportion (%)
40.0
30.0
24.9%
AC C
60.0
50.0
EP
70.0 Asthma - Spirometry / COPD - Spirometry
Asthma - Diagnosis / COPD - Diagnosis
TE D M AN U SC
RI PT
ACCEPTED MANUSCRIPT
Asthma - Diagnosis / COPD - Spirometry
20.0
16.0%
17.0% 61.0%
Asthma - ICD Code / COPD - ICD Code
51.4%
43.0% 45.9% 46.0%
30.0% 30.4% 29.0% 29.8% 27.5% 27.8% 28.2%
18.0%
13.3%
10.0
0.0
ACCEPTED MANUSCRIPT Highlights Prevalence of ACOS varies widely (12–61%) among patients with COPD or asthma
•
The variability is linked to differences in COPD and asthma diagnostic criteria
•
Other factors linked to variability include age and gender of the study population
•
Disease burden in patients with ACOS is greater than in asthma or COPD alone
AC C
EP
TE D
M AN U
SC
RI PT
•
ACCEPTED MANUSCRIPT
KW was an employee of GSK at the time the study was performed and is currently employed by PAREXEL. SP and NB are employees of GSK and own stocks/shares in GSK.
AC C
EP
TE D
M AN U
SC
RI PT
KK-R and GB hold graduate research assistant positions at The University of North Carolina.
S0954-6111(15)30068-8
DOI:
10.1016/j.rmed.2015.10.004
Reference:
YRMED 4796
To appear in:
Respiratory Medicine
Received Date: 6 May 2015 Revised Date:
27 August 2015
Accepted Date: 5 October 2015
Please cite this article as: Wurst KE, Kelly-Reif K, Bushnell G, Pascoe S, Barnes N, Understanding Asthma-Chronic Obstructive Pulmonary Disease Overlap Syndrome, Respiratory Medicine (2015), doi: 10.1016/j.rmed.2015.10.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Understanding Asthma-Chronic Obstructive Pulmonary Disease Overlap
RI PT
Syndrome
Keele E. Wurst,a Kaitlin Kelly-Reif,b Greta Bushnell,b Steven Pascoe,c Neil Barnesd
a
employed with GSK, Research Triangle Park, NC, USA) b
SC
PAREXEL International, Clinical Research Services, Research Triangle Park, NC USA (previously
c
GSK R&D, Research Triangle Park, NC, USA
d
M AN U
Department of Epidemiology, University of North Carolina, Chapel Hill, NC, USA
GSK, Stockley Park, Uxbridge, UK, William Harvey Institute and Barts and The London School of
Medicine and Dentistry
Keele E. Wurst PhD, RPh PAREXEL International
EP
Clinical Research Services
TE D
Corresponding author:
NC USA
AC C
Research Triangle Park
Email: [email protected]
Telephone: +1 919 483 7994
Key words (3 to 6): asthma, asthma-chronic obstructive pulmonary disease overlap syndrome, chronic obstructive pulmonary disease
1
ACCEPTED MANUSCRIPT
Abbreviations ACOS: asthma-chronic obstructive pulmonary disease overlap syndrome; BOLD: Burden of Obstructive
RI PT
Lung Disease; COPD: chronic obstructive pulmonary disease; COPDGene: Genetic Epidemiology of COPD; GEIRD: Gene-Environment Interactions in Respiratory Disease; GINA: Global Initiative for Asthma; GOLD: Global Initiative for Chronic Obstructive Lung Disease; FEV1: forced expiratory volume in one second;
SC
FVC: forced vital capacity; ICD: International Classification of Disease; IHCIS: Integrated Healthcare Information Services; LLN: lower limit of normal; NC BRFSS: North Carolina Behavioral Risk Factor
M AN U
Surveillance System; NHANES: National Health and Nutritional Examination Survey; PEF: peak expiratory flow; PLATINO: Latin American Project for the Investigation of Lung Disease; SGRQ: Saint George’s Respiratory Questionnaire; SPIROMICS: Subpopulations and intermediate outcome measures in COPD
AC C
EP
TE D
study; U-BIOPRED: Unbiased BIOmarkers in PREDiction of respiratory outcomes.
2
ACCEPTED MANUSCRIPT
Summary Asthma-chronic obstructive pulmonary disease overlap syndrome (ACOS) is a loosely-defined clinical entity referring to patients who exhibit characteristics of both asthma and chronic obstructive
RI PT
pulmonary disease (COPD). Clinical definitions and classifications for ACOS vary widely, which impacts our understanding of prevalence, diagnosis and treatment of the condition.
SC
This literature review was therefore conducted to characterize the prevalence of ACOS and the effect of different disease definitions on these estimates, as this has not previously been explored. From an
M AN U
analysis of English language literature published from 2000 to 2014, the estimated prevalence of ACOS ranges from 12.1% to 55.2% among patients with COPD and 13.3% to 61.0% among patients with asthma alone. This variability is linked to differences in COPD and asthma diagnostic criteria, disease ascertainment methods (spirometry-based versus clinical or symptom-based diagnoses and claims data),
TE D
and population characteristics including age, gender and smoking.
Understanding the reasons for differences in prevalence estimates of ACOS across the literature may
EP
help guide decision making on the most appropriate criteria for defining ACOS and aid investigators in
AC C
designing future ACOS clinical studies aimed at effective treatment.
3
ACCEPTED MANUSCRIPT
Introduction The diagnosis and differentiation of asthma from chronic obstructive pulmonary disease (COPD) in clinical practice is relatively straightforward in the majority of cases; however, some patients exhibit
RI PT
characteristics of both diseases. Where uncertainty exists regarding the correct diagnosis of asthma, COPD or both, this may represent a phenotype known as asthma-COPD overlap syndrome (ACOS).[1-4]
SC
COPD is highly prevalent in the global population of older adults (40 years of age and older) and has been associated with smoking and exposure to environmental tobacco smoke or fumes.[5] COPD is
M AN U
typically characterized by persistent airflow obstruction and chronic inflammation of the airways. Airway inflammation is also seen in asthma; however, there are distinct differences in the type of inflammatory cells seen in these two respiratory diseases. Biopsies reveal that inflammation in COPD is characterized predominantly by increases in CD8+ T-lymphocytes, neutrophils, and macrophages,[6-8] although increases in eosinophils have been observed in sputum at the time of exacerbation.[9] In contrast,
TE D
inflammation in asthma is commonly characterized by increases in CD4+ T-lymphocytes and eosinophils.[10] Airflow obstruction in COPD is defined as a post-bronchodilator measurement of forced
EP
expiratory volume in 1 second/forced vital capacity (FEV1/FVC) <0.70 or lower limit of normal (LLN), where LLN values are identified by selecting the lowest 2.5% of a normally distributed population.[5]
AC C
Asthma is also a chronic obstructive lung disease, but in mild and moderately severe asthma, airflow obstruction responds to treatment with inhaled corticosteroids and bronchodilators and is therefore not persistent and is reversible. Reversibility refers to the reduction of airflow obstruction after bronchodilator administration and is usually measured by comparing pre- and post-bronchodilator FEV1. An FEV1/FVC ratio >0.75–0.80, and a post-bronchodilator FEV1 increase of >12% and >0.200 L over the pre-bronchodilator measurements are usually indicators of asthma.[11] However, applying these definitions to differentiate between severe asthma of middle-age and COPD may lead to uncertainty
4
ACCEPTED MANUSCRIPT
owing to a number of factors. For example, many patients with COPD are responsive to albuterol (reversible) while many patients with a clinical diagnosis of asthma are not reversible on testing and approximately 25% of patients with COPD have never smoked.[12, 13] Furthermore, since the FEV1/FVC
RI PT
ratio declines with age, this definition may lead to over-diagnosis of COPD in the elderly and underdiagnosis in the young to middle-aged, which may be especially relevant when considering the possible diagnosis of nonreversible asthma.[5] In addition, there are some individuals with asthma who have
SC
neutrophilic inflammation that may be related to treatment with inhaled steroids[14] or an intrinsic part of their disease.[15, 16] Finally, the histopathological differentiation of asthma as an eosinophilic disease
M AN U
and COPD as a neutrophilic disease is also in question as the eosinophil is becoming an increasingly important marker of treatment response in COPD,[17, 18] which is possibly a result of a rise in the prevalence of ACOS.
TE D
While a variety of diagnostic criteria have been proposed for identifying ACOS, a specific formal definition of ACOS has yet to be determined. The importance of identifying patients with features of ACOS is underlined by recent work that has demonstrated that patients with ACOS have more
EP
symptoms, exacerbations, hospitalizations, lower health-related quality of life, and higher mortality than patients with COPD or asthma alone.[19-24] Recognizing that there is no generally agreed term or
AC C
definition for ACOS, the Global Initiative for Asthma (GINA) and the Global Initiative for Chronic Obstructive Lung Disease (GOLD) recently developed guidelines to help stimulate further research, to identify and characterize ACOS.[5, 11] Broadly, ACOS can be described as an obstructive airflow condition or a set of clinical characteristics where aspects of both asthma and COPD are present, such as FEV1/FVC <0.70 and evidence of airflow reversibility (post-bronchodilator FEV1 increase of >12% and >0.200 L). Developing diagnostic criteria for ACOS is especially difficult since the clinical presentation of asthma and COPD can appear similar with respect to symptomology and, as noted above, are
5
ACCEPTED MANUSCRIPT
sometimes challenging to differentiate.[1, 25] Whether ACOS is simply the coexistence of asthma and COPD or a distinct phenotype with related fundamental pathogenic mechanisms to asthma and COPD remains to be determined.[26] Furthermore, different studies classify ACOS as a sub-phenotype of
RI PT
COPD,[27, 28] whereas other studies treat ACOS as a phenotype distinct from either COPD or
asthma.[22, 29, 30] Proposed definitions for ACOS vary widely and include (1) patients with COPD who have any diagnosis of current or past asthma, (2) spirometric-based definitions such as patients with
SC
COPD who have consistent significant reversibility (FEV1/FVC <0.70 and a post-bronchodilator FEV1 increase of >12% and >0.200 L over the pre-bronchodilator measurement), or (3) patients with asthma
M AN U
who have persistent airflow obstruction.[3, 31] The absence of ACOS-specific pathological definitions and the variability of ACOS criteria across the literature present challenges for conducting studies to add to the available evidence.[3]
TE D
Aims
ACOS may represent a sizeable portion of the clinical patient population with airflow obstruction; however, the varying ACOS definitions and diagnostic criteria make this population difficult to
EP
characterize. The aim of this literature review was to characterize the prevalence of ACOS and examine
AC C
how different definitions of ACOS affect prevalence estimates, as this has not previously been explored.
Search strategy
Published studies and abstracts were identified through PubMed and Searchlight, respectively, from January 1, 2000 to February 28, 2014. Only studies written in English were retained. Relevant publications found from other references, but not identified in the original search, were added. The following key words were used to identify potentially relevant abstracts in PubMed: asthma, chronic obstructive pulmonary disease, COPD, asthma-chronic obstructive pulmonary disease, fixed airway
6
ACCEPTED MANUSCRIPT
obstruction, asthma COPD overlap, persistent airway obstruction, fixed airway obstruction risk factor, and prevalence. This search strategy produced 180 results, 71 of which were retained for further review. Thirty-one additional relevant papers identified through citations of systematically identified papers
RI PT
were also included in this analysis.
The following key words were used to identify potentially relevant conference abstracts in Searchlight:
SC
asthma, atopic asthma, bronchial asthma, bronchitic asthma, chronic obstructive pulmonary disease, chronic airflow obstruction, chronic obstructive airways disease, chronic obstructive lung disease, COPD,
M AN U
pulmonary disease, chronic obstructive, and overlap. This search strategy produced 23 conference abstracts, 18 of which were retained for further review.
Of all the articles retained for further assessment, 43 were considered relevant and included in this
TE D
review. Studies identified but not included in the review did not contain information on ACOS prevalence. Topics of the excluded articles included dysfunctional breathing, COPD subtypes excluding ACOS, smoking, treatment/management, air trapping, and airway obstruction. Studies that estimated
EP
the population prevalence of ACOS, the prevalence of ACOS in patients with COPD, and the prevalence of ACOS in patients with asthma were summarized. Variations were identified in the study populations
AC C
and in the definitions of asthma, COPD, and ACOS.
7
ACCEPTED MANUSCRIPT
Prevalence of ACOS Estimates of the population prevalence of ACOS ranged from 1.6% to 4.5% from studies in Italy, South Korea, Latin America, and the United States[22, 26, 30, 32, 33] (Table 1, Figure 1). Estimates of the
RI PT
population prevalence of ACOS varied across countries at least in part because of different COPD and asthma diagnostic criteria and different underlying distribution of risk factors, such as smoking, the age
AC C
EP
TE D
M AN U
SC
distribution in the source populations, and study samples.
8
ACCEPTED MANUSCRIPT
Table 1. COPD, asthma, and ACOS prevalence among total populations: differences in asthma definitions, spirometric criteria, COPD
Population Prevalence of COPD-only
Population Prevalence of Asthma-only
Population Prevalence of ACOS
5.3%
5.3%
11.7%
RI PT
definitions, and population demographics
ACOS Definition
Population Demographics (age: years)
Ever self-reported physician diagnosis of asthma and COPD
Age ≥25 General population, US Smokers and nonsmokers
Report of wheezing in the last 12 months and post-bronchodilator increase in FEV1 or FVC of 0.200 L and 12% Self-reported physician diagnosed asthma, ever
Meeting both criteria of asthma and COPD
Age >40 Population in 5 Latin American cities Smokers and nonsmokers
Meeting both criteria of asthma and COPD
Wheezing for past year and FEV1/FVC≥0.70
Wheezing and FEV1/FVC<0.70
Age >40 Population in 5 Latin American cities Smokers and nonsmokers Korean patient population Smokers and nonsmokers
Asthma Definition
2.7%
Self-reported physician diagnosis of COPD, ever
Self-reported physician diagnosed, asthma ever
1.7%
1.8%
FEV1/FVC<0.70
10.6%
8.9%
2.9%
FEV1/FVC<0.70
7.8%
6.1%
2.4%
FEV1/FVC<0.70 and no wheezing in past year
8.2% (7.5–9)
a
1.6% (1.3–2)
a
M AN U
TE D
a
Self-reported physician diagnosis of COPD, ever
Self-reported physician diagnosed asthma, ever
Ever self-reported physician diagnosis of asthma and COPD
Age 20–44 General population, Italy Smokers and nonsmokers
EP
3.3% (2.8–3.8)
SC
COPD Definition
Self-reported physician diagnosed asthma, ever
Ever self-reported physician diagnosis of asthma and COPD
Age 45–64 General population, Italy Smokers and nonsmokers
5.7% (4.7–6.7)a
4.9% (4–5.9)a
2.1% (1.5–2.8)a
13.3% (11.1–15.5)a
2.9 % (1.8–4)a
4.5 %(3.2–5.9)a
Self-reported physician diagnosis of COPD, ever
Self-reported physician diagnosed asthma, ever
Ever self-reported physician diagnosis of asthma and COPD
Age 65–84 General population, Italy Smokers and nonsmokers
5.6%
12.1%
2.4%
Self-reported physician diagnosis of COPD, ever
Self-reported physician diagnosed asthma, ever
Ever physician diagnosis of asthma and COPD
Age ≥18 General population, North Carolina Smokers and nonsmokers
AC C
Self-reported physician diagnosis of COPD, ever
Author, Year Data Study Sample Size Diaz-Guzman et al. 2011[32] NHANES III n = 15,203 Menezes et al. 2014 [22] PLATINO n=5,044 Menezes et al. 2014[22] PLATINO n=5,044 Lee et al. 2013[33] 4th Korean NHANES n=9,369 de Marco et al. 2013[26] GEIRD n=5,163 de Marco et al. 2013[26] GEIRD n=2,167 de Marco et al. 2013[26] GEIRD n=1,030 Pleasants et al. 2014[30] NC BRFSS n=24,073
9
ACCEPTED MANUSCRIPT
ACOS: asthma-COPD overlap syndrome; COPD: chronic obstructive pulmonary disease; FEV1: forced expiratory volume in 1 second; FVC: forced vital capacity; GEIRD: GeneEnvironment Interactions in Respiratory Disease; ICD: International Classification of Diseases; NC BRFSS: North Carolina Behavioral Risk Factor Surveillance System; NHANES:
EP
TE D
M AN U
SC
Prevalence (95% confidence interval).
AC C
a
RI PT
National Health and Nutrition Examination Survey; PLATINO: Latin American Project for the Investigation of Lung Disease.
10
ACCEPTED MANUSCRIPT
ACOS prevalence among patients with COPD or asthma On the basis of the results of the literature review, the prevalence of ACOS among patients with COPD is estimated to range from 12.1% to 55.2% (Table 2)[19-22, 24, 26-28, 30, 33-42] and the prevalence of
RI PT
ACOS among patients with asthma is estimated to range from 13.3% to 61.0% (Table 3).[19, 21, 22, 26, 30, 32, 33, 37, 38, 41, 43-45] The wide variations in prevalence estimates are most likely due to the method by which ACOS cases were classified, but differences in population inclusion criteria, in the data
SC
source, and in the definitions of asthma and COPD including airflow reversibility criteria and spirometric versus clinical diagnoses can also influence the prevalence of ACOS. In general, the prevalence of COPD
M AN U
alone is difficult to accurately estimate due to variations in COPD diagnostic criteria, thus, identifying the
AC C
EP
TE D
ACOS population is even more complicated.[46]
11
ACCEPTED MANUSCRIPT
COPD Definition
Asthma Definition
ACOS Definition
12.1%
FEV1/FVC <0.70
Self-reported physician-diagnosed before the age of 40
Asthma diagnosis among patients with COPD
12.6%
FEV1/FVC <0.70, FEV1<80% of predicted, GOLD Stage II+
Self-reported physician-diagnosed before the age of 40
Asthma diagnosis among patients with COPD
13.0%
FEV1/FVC <0.70, GOLD Stage II+
Self-reported physician-diagnosed before the age of 40
Asthma diagnosis among patients with COPD
13.0%
FEV1/FVC <0.70
Report of wheezing in the last 12 months and post-bronchodilator increase in FEV1 or FVC of 0.200 L and 12%
Asthma diagnosis among patients with COPD
16.7%
FEV1/FVC <0.70
Self-reported physician diagnosis, current
Asthma diagnosis among patients with COPD
22.8%
FEV1/FVC <0.70
Self-report medical-diagnosed asthma, prior
Asthma diagnosis among patients with COPD
23.5%
FEV1/FVC <0.70, FEV1<80% of predicted
Self-reported physician diagnosed, ever
Asthma diagnosis among patients with COPD
23.6%
FEV1/FVC <0.70 and no wheezing in past year
Wheezing and FEV1/FVC<0.70
Korean patient population Smokers and nonsmokers
25.0%
Self-reported physician diagnosis of COPD, ever
Self-reported physician diagnosed asthma, ever
Ever self-reported physician diagnosis of asthma and COPD
26.0%
ICD-10 codes for COPD diagnoses (ICD-10:J41-J44)
ICD-10 codes for asthma diagnoses (ICD-10:J45-J46)
Diagnosis of both asthma and COPD based on ICD-10 codes at different times
Age 65–84 General population, Italy Smokers and nonsmokers Age >34 Persons with a primary and/or secondary diagnosis of asthma and/or COPD based on ICD-10 codes among the entire Finnish
SC
M AN U
TE D
AC C
Wheezing for past year and FEV1/FVC ≥0.70
Demographics of Population with COPD (age: years)
RI PT
Proportion of COPD Patients with ACOS
EP
Table 2. ACOS prevalence among patients with COPD: variation in study definition and details of general population with COPD
Age ≥40 Spain multicenter study Smokers only Age 45–80 US multicenter study Current and former smokers Age 45–80 US multicenter study Current and former smokers Age >40 Population in 5 Latin American cities Smokers and nonsmokers Age 40–80 US, multicenter Among smokers only Age ≥40 Population in 5 Latin American cities Smokers and nonsmokers Age ≥40 Multi-country Never smokers only
Author, Year Data Study Sample Size Izquierdo-Alonso et al. 2013[27] n=331 Hardin et al. 2014[20] COPDGene n=3,570 Hardin et al. 2011[34] COPDGene n=1,059 Menezes et al. 2014[22] PLATINO n=5,044 Putcha et al. 2013[24] SPIROMICS n=581 Talamo et al. 2007[36] PLATINO n=5,303 Lamprecht et al. 2011a[37] BOLD Study n=4,291 Lee et al. 2013[33] th 4 Korean National Health and Nutritional Survey n=9,369 de Marco et al. 2013[26] GEIRD n=1,030 Andersen et al. 2013[19] Hospitalization records n=105,122
12
Proportion of COPD Patients with ACOS
COPD Definition
Asthma Definition
ACOS Definition
26.4%
FEV1/FVC <0.70 or FEV1/FVC <88% of predicted
27.0%
Self-reported physician diagnosis of COPD, ever
Post-bronchodilator increase in FEV1 of ≥12%, a bronchodilator response of ≥15% or diurnal variation of ≥20% in PEF recording, moderate-to-severe bronchial hyperactivity, or a decrease in FEV1 of ≥15% in the exercise test Self-reported physician diagnosed asthma, ever
Ever self-reported physician diagnosis of asthma and COPD
Age 45–64 General population, Italy Smokers and nonsmokers
ICD-9 codes 491, 492, and 496
Experience of asthmatic symptoms (i.e., wheezing, cough, and chest tightness worsening at night) ICD-9 code 493
Diagnosis of COPD and report of experiencing asthmatic symptoms Diagnosis of asthma and COPD within 12 months
Hospital-based patient sample with stable COPD (2007-2009), Japan Age ≥40 US patients with medical and pharmacy benefits
de Marco et al. 2013[26] GEIRD n=2,167 Kitaguchi et al. 2012[39] n=63
27.0%
FEV1/FVC <70% and FEV1 <80%
30.2%
32.9%
FEV1/FVC <0.70 COPD Medical Codes
Current or previous asthma diagnosis
Asthma diagnosis among patients with COPD
33.0%
Self-reported physician diagnosis of COPD, ever
Self-reported physician diagnosed asthma, ever
Ever self-reported physician diagnosis of asthma and COPD
Age ≥40 UK patients with newly diagnosed COPD Smokers and nonsmokers Age 20–44 General population, Italy Smokers and nonsmokers
41.4%
Self-reported physician diagnosis of COPD, ever
Self-reported physician diagnosed, ever
Diagnosis of asthma and COPD ever
43.0%
ICD-9 codes 491, 492, and 496
Diagnosis of asthma and COPD within 30 months
49.0%
FEV1/FVC <0.70 or FEV1/FVC <88% of predicted reference value
54.6%
ICD-10 codes J42–J44, except J430 Use of more than one drug for COPD at least twice per year
TE D
M AN U
SC
RI PT
Author, Year Data Study Sample Size
Diagnosis of both asthma and COPD based on spirometric criteria
Demographics of Population with COPD (age: years) population 134 volunteers Finnish patients with asthma and COPD
EP
ACCEPTED MANUSCRIPT
AC C
ICD-9 code 493
Post-bronchodilator increase in FEV1 of ≥12%, a bronchodilator response of ≥15% or diurnal variation of ≥20% in PEF recording, moderate to severe bronchial hyperreactivity, or a decrease in FEV1 of ≥15% in the exercise test ICD-10 codes J45-J46 Use of more than one drug for asthma at least twice per year
Age ≥18 General population, US Smokers and nonsmokers Age 40–64 US Medicaid patients
Diagnosis of both asthma and COPD based on spirometric criteria
Age 18–75 Finnish hospital-based population, identified from ICD-10 codes J44–J46 Smokers and nonsmokers
Diagnosis of asthma and COPD within 12 months
Age ≥40 Korean patient population
Iwamoto et al. 2014[38] Sample of patients from FinnCADStudy n=134
Blanchette et al. 2009[40] US IHCIS Claims n=24,935 Wurst et al. 2014[35] General practice research database n=7,881 de Marco et al. 2013[26] GEIRD n=5,163 Pleasants et al. 2014[30] NC BRFSS n=24,073 Shaya et al. 2008[41] Medicaid claims n=9,131 Kauppi et al. 2011[21] Medical records n=1,546
Rhee et al. 2014[42] Insurance data n=185,147
13
ACCEPTED MANUSCRIPT
COPD Definition
Asthma Definition
ACOS Definition
Demographics of Population with COPD (age: years)
55.2%
FEV1/FVC <0.70
Post-bronchodilator increase in FEV ≥15%, peak flow variability ≥20% during 1 week of testing, physician diagnosis of asthma in conjunction with current symptoms or inhaler use in the last 12 months
Diagnosis of asthma within last 12 months and current COPD
Age >50 Random population based participants, New Zealand Smokers and nonsmokers
RI PT
Proportion of COPD Patients with ACOS
Author, Year Data Study Sample Size Marsh et al. 2008[28] Wellington respiratory survey subset n=469
SC
ACOS: asthma-COPD overlap syndrome; COPD: chronic obstructive pulmonary disease; BOLD: Burden of Obstructive Lung Disease; COPDGene: Genetic Epidemiology of COPD; FEV1: forced expiratory volume in one second; FVC: forced vital capacity; GEIRD: Gene-Environment Interactions in Respiratory Disease; GOLD, Global Initiative for Chronic
M AN U
Obstructive Lung Disease; ICD: International Classification of Diseases; IHCIS: Integrated Healthcare Information Services; NC BRFSS: North Carolina Behavioral Risk Factor Surveillance System; PEF: peak expiratory flow; PLATINO: Latin American Project for the Investigation of Lung Disease; SPIROMICS: Subpopulations and intermediate outcome measures in COPD study. a
Lamprecht et al 2011[37]: Of the patients with a physician diagnosis of asthma ‘ever’, 30.4% and 17.7% were considered GOLD stage II+ (FEV1/FVC <0.7 and FEV1 <80% predicted)
AC C
EP
TE D
and GOLD stage I (FEV1/FVC <0.7 and FEV1 ≥80% predicted), respectively.
14
ACCEPTED MANUSCRIPT
Table 3. ACOS prevalence among patients with asthma: variation in study definition and details of general population of patients with asthma COPD Definition
Asthma Definition
ACOS Definition
Demographics of Population with Asthma (age: years)
13.3%
FEV1/FVC <0.70 and FEV1 <80% predicted
Self-reported physician diagnosed, asthma ever
16.0%
Self-reported physician diagnosis of COPD, ever
Self-reported physician diagnosed asthma, ever
Evidence of obstruction (FEV/FVC <0.70 and FEV1 <80% predicted, and ever selfreported physician diagnosis of asthma) Ever self-reported physician diagnosis of asthma and COPD
17.0%
FEV1/FVC <0.70 or FEV1/FVC <88% of predicted
18.0%
FEV1/FVC <0.70
Post-bronchodilator increase in FEV1 of ≥12%, a bronchodilator response of ≥15% or diurnal variation of ≥20% in PEF recording, moderate to severe bronchial hyper reactivity, or a decrease in FEV1 of ≥15% in the exercise test. Physician diagnosis of asthma
24.9%
FEV1/FVC <0.70, FEV1<80% of predicted
27.5%
Airway obstruction where FEV1/FVC ratio < predicted LLN
27.8%
Self-reported physician diagnosis of COPD, ever
28.2%
FEV1/FVC <0.70 and no wheezing in past year
Wheezing for past year and FEV1/FVC≥0.70
Wheezing and FEV1/FVC<0.70
29.0%
FEV1/FVC <0.70
Physician diagnosis of asthma
Meeting the criteria for asthma and COPD
Patients with asthma ages 40–60 Former or nonsmokers
29.8%
ICD-10 codes for COPD
ICD-10 codes for asthma
Diagnosis of both asthma and
Age ≥34
RI PT
Proportion of Asthma Patients with ACOS
Age ≥25 General Population, US Smokers and nonsmokers
Author, Year Data Study Sample Size Diaz-Guzman et al. 2011[32] NHANES III n=15,203 de Marco et al. 2013[26] GEIRD n=5,163 Kauppi et al. 2011[21] Hospital medical records n=1,546
Meeting the criteria for asthma and COPD
Patients with asthma ages 18–40 Former or nonsmokers
Self-reported physician diagnosed, ever
Meeting the criteria for asthma and COPD
Diagnosis of bronchial asthma between 2009 and 2011
Meeting the criteria for asthma and COPD
Age ≥40 Multi-country Never smokers only Age 18–88 years Sample of US patients with spirometry data
Self-reported physician diagnosed asthma, ever
Ever physician diagnosis of asthma and COPD
Kalberg et al. 2005[43] Clinical trials database (1998-2002) n=6,497 Lamprecht et al. 2011[37] BOLD study n=4,291 Banga et al. 2013[44] Tertiary care center n=1,482 Pleasants et al. 2014[30] NC BRFSS n=24,073
SC
Age 20–44 General population, Italy Smokers and nonsmokers Age 18–75 Finnish hospital-based study population identified from ICD-10 codes J44–J46
AC C
EP
TE D
M AN U
Diagnosis of both asthma and COPD based on spirometric criteria
Ages ≥18 Smokers and nonsmokers General population, North Carolina, US Korean patient population Smokers and nonsmokers
Lee JH et al. 2013[33] 4th Korean NHANES n=9,369 Kalberg et al. 2005[43] Clinical trials database (1998-2002) n=3,346 Andersen et al. 2013[19]
15
ACCEPTED MANUSCRIPT
diagnoses (ICD-10:J41-J44)
diagnoses (ICD-10:J45-J46)
COPD based on ICD-10 codes at different times
30.0%
Self-reported physician diagnosis of COPD, ever
Self-reported physician diagnosed asthma, ever
Ever self-reported physician diagnosis of asthma and COPD
30.4%
FEV1/FVC <0.70 or FEV1/FVC <88% of predicted
Diagnosis of both asthma and COPD based on spirometric criteria
43.0%
FEV1/FVC <0.70
Post-bronchodilator increase in FEV1 of ≥12%, a bronchodilator response of ≥15% or diurnal variation of ≥20% in PEF recording, moderate-to-severe bronchial hyperactivity, or a decrease in FEV1 of ≥15% in the exercise test Physician diagnosis of asthma
Meeting the criteria for asthma and COPD
Patients with asthma age >60 US population Former or nonsmokers
45.9%
ICD-9 Codes 491, 492, and 496
ICD-9 Code 493
Diagnosis of asthma and COPD within 30 months
Age 40–64 US Medicaid patients
46.0%
Post-bronchodilator FEV1 or FEV1/FVC <75% of predicted with a TLC>75% of predicted
Meeting criteria for asthma and COPD
Average age 55.3 (SD 11.7) Former or nonsmokers
51.4%
FEV1/FVC <0.70
Wheezing history, FEV1/FVC <0.70 and post-bronchodilator FEV1 increase of 12%, PC20, diurnal PEF variation/tapering med Report of wheezing in the last 12 months and postbronchodilator increase in FEV1 or FVC of 0.200 L and 12%
COPD diagnosis among patients with asthma
61.0%
Self-reported physician diagnosis of COPD, ever
Age >40 Population in 5 Latin American cities Smokers and nonsmokers Age 65–84 General population, Italy Smokers and nonsmokers
RI PT
SC
M AN U
TE D
EP
AC C
Self-reported physician diagnosed asthma, ever
Persons with a primary and/or secondary diagnosis of asthma and/or COPD based on ICD-10 codes among the entire Finnish population Age 45–64 General population, Italy Smokers and nonsmokers 134 volunteers Finnish patients with asthma and COPD
Ever self-reported physician diagnosis of asthma and COPD
Hospitalization records n=105,122
de Marco et al. 2013[26] GEIRD n=2,167 Iwamoto et al. 2014[38] Sample of patients from FinnCADStudy n=134
Kalberg et al. 2005[43] Clinical trials database (1998-2002) n=601 Shaya et al. 2008[41] Medicaid claims n=9,131 Wagener, 2013[45] U-BIOPRED study cohort n=148 Menezes et al 2014[22] PLATINO n=5,044 de Marco et al. 2013[26] GEIRD n=1,030
ACOS: asthma-COPD overlap syndrome; COPD: chronic obstructive pulmonary disease; BOLD: Burden of Obstructive Lung Disease; FEV1: forced expiratory volume in one second; FVC: forced vital capacity; GEIRD: Gene-Environment Interactions in Respiratory Disease; ICD: International Classification of Diseases; LLN, lower limit of normal; NC BRFSS: North Carolina Behavioral Risk Factor Surveillance System; NHANES: National Health and Nutrition Examination Survey; PEF: peak expiratory flow; PLATINO: Latin American Project for the Investigation of Lung Disease; U-BIOPRED: Unbiased BIOmarkers in PREDiction of respiratory outcomes; SD: standard deviation.
16
ACCEPTED MANUSCRIPT
Spirometry-based versus clinical or symptom-based diagnoses In some studies an asthma diagnosis is based on spirometry-confirmed reversibility. The investigators from the Latin American Project for the Investigation of Lung Disease (PLATINO) study in Latin America
RI PT
noted that two different definitions of asthma resulted in varied ACOS prevalence estimates in the same population. The prevalence of ACOS based on spirometry and the presence of reversibility was
SC
estimated to be 1.8% compared with 2.9% using physician diagnosis only.[22]
Spirometry-defined asthma
M AN U
The variation in ACOS prevalence estimates among patients with COPD is highlighted by the reversibility criteria in patients with COPD set forth in different medical systems internationally. In a Finnish medical records cohort study, the authors stated that reversibility seen in patients with COPD is considered to be asthma, so definitions of COPD and asthma among the Finnish population will sometimes differ
TE D
compared with other populations.[19, 21] Analysis of medical records and hospitalization records in Finland estimated that between 30% and 49% of patients with COPD also have ACOS[19, 21]; here diagnostic criteria of asthma appeared to be based on spirometric measurements, including spirometric
EP
criteria for reversibility.[19]
AC C
In other countries where reversibility among patients with COPD is not necessarily regarded as asthma, it is difficult to distinguish between COPD with asthma and COPD alone with airway reversibility. If ACOS is defined on the basis of obstruction and reversibility, this phenotype may include patients with COPD who have reversibility but no other asthma features. Evidence now suggests that 12–42% of patients with COPD may demonstrate post-bronchodilator reversibility; however, reversibility may vary over time and thus it may not be a stable phenotype.[47, 48]
17
ACCEPTED MANUSCRIPT
Spirometry-defined COPD When using spirometric criteria for COPD classification, the estimates of ACOS among patients with asthma ranged from 25.0% to 51.4% (Table 3). The diagnostic criteria for COPD also influence estimates
RI PT
of both overall COPD prevalence and ACOS prevalence. Marsh et al.[28] estimated an ACOS prevalence of 55.2% in patients with COPD aged 50 years and older using spirometric criteria. The authors
estimated the prevalence of ACOS using two different criteria for COPD, (1) FEV1/FVC <0.70 and (2)
SC
FEV1/FVC less than the LLN.[28] They concluded that when less than the LLN criterion was used in COPD diagnosis rather than the FEV1/FVC <0.70 criterion, one-third fewer patients presented with COPD.
M AN U
Furthermore, Marsh et al. concluded that fewer patients with COPD also presented with the ‘asthma phenotype’ (reversibility, peak flow variability, or physician diagnosis in addition to COPD) when the LLN criterion was used instead of the FEV1/FVC <0.70 criterion.[28]
TE D
A Norwegian study compared spirometric results against physician diagnoses in 376 patients aged 40 years or older.[29] The study reported that among patients previously diagnosed with only asthma by a physician, 17.1% could be re-diagnosed with COPD because of obstruction using spirometric criteria.[29]
EP
Diaz-Guzman et al. reported that 13.3% of patients with asthma in the US National Health and Nutrition Examination Survey (NHANES) III had evidence of obstruction and thus may also be classified as COPD or
AC C
ACOS.[32]
Self-reported physician diagnosis In the Continuing to Confront COPD survey, international prevalence estimates of ACOS among patients with COPD in 12 countries ranged from 12% to 54%.[49] Asthma was defined as self-reported physician diagnosis and COPD as self-reported physician diagnosis or symptom-based chronic bronchitis. The
18
ACCEPTED MANUSCRIPT
broad range of estimates highlights the differences in underlying asthma population characteristics, diagnosis practices, and healthcare systems across countries.
RI PT
ACOS defined in claims data
ACOS prevalence estimates are generally higher using claims data than those using spirometric
definitions or self-reported physician diagnoses. It is unknown whether the medical codes are based on
SC
reversibility criteria or other criteria including symptoms such as wheezing. An analysis of US claims data using International Classification of Disease (ICD)-9 codes estimated that 30–43% of the population over
M AN U
40 years of age with a diagnosis code for COPD also had an asthma diagnosis code within 12–30 months.[40, 41] Estimates from claims data depend on the duration of time allowed between COPD and asthma diagnostic codes and whether more than one asthma code is required, as some patients may receive an asthma diagnosis before their COPD diagnosis and thus could be misclassified as having both
TE D
diseases. Patients with more utilizations (for any reason) also have more opportunities to receive dual diagnoses. Additionally, as was previously mentioned, it may not be clear how the physician diagnosed
EP
COPD or asthma (i.e., if spirometry was used and if reversibility was considered).
Asthma definition: past versus current asthma
AC C
Another source of variation in ACOS prevalence estimates is the definition of asthma with regard to timing of the diagnosis, for example, ‘ever’ diagnosed with asthma, ‘prior’ asthma diagnosis (usually before the age of 40), or ‘current’ asthma diagnosis. The ACOS literature lacks consensus on whether the ACOS population only includes those with the concurrence of existing asthma and clinical characteristics of COPD or if the ACOS population includes patients with COPD and a history of asthma. The GINA guidelines include both.[11]
19
ACCEPTED MANUSCRIPT
In the subpopulations and intermediate outcome measures in the COPD study (SPIROMICS), restricting asthma diagnostic criteria to a current asthma diagnosis resulted in an ACOS prevalence estimate of 16.7% among smokers over the age of 40 with COPD.[24] A study of self-reported, physician-diagnosed
RI PT
asthma among smokers before the age of 40 with no current diagnosis reported a prevalence of 12.1% with COPD.[27] A UK study of ACOS that defined asthma as a current or former diagnosis reported a prevalence of 32.9% in smokers and nonsmokers with COPD.[35] All three of these studies used
SC
spirometric criteria for the evaluation of COPD, and used either self-reported physician diagnosis or medical codes for asthma identification. Comparing these studies, the inclusion of current or past
M AN U
asthma (i.e., ‘ever’ asthma) resulted in higher estimates of ACOS prevalence than studies with asthma defined as ‘current’ asthma or as only ‘past’ asthma.
Ford and Mannino [50] utilized data from the US NHANES III (1988–1994) to estimate the prevalence of
TE D
‘ever’ asthma (7.8%) and ‘current’ asthma (5.2%) in people aged 20–74 years. Taken together, these estimates suggest that 67% of people to ever report asthma will currently report asthma; applying that proportion to the ACOS population prevalence estimate of 2.7% from Diaz-Guzman et al., who defined
EP
asthma as ‘ever’ reported, the prevalence of ACOS decreases to 1.8%, demonstrating the change in
AC C
estimate when using ‘current’ versus ‘ever’ asthma.[32]
Comparisons among definitions Differences between studies such as sample size, source population, and age inclusion criteria make it difficult to conclude that using spirometric diagnosis is solely responsible for the difference in estimates.[28, 30] Generally, there are four different classification schemes for defining ACOS: (1) a reported physician diagnosis of asthma and a reported physician diagnosis of COPD at any point in a patient’s life, (2) a reported physician diagnosis of asthma and spirometry-defined COPD, (3) both
20
ACCEPTED MANUSCRIPT
spirometry-defined asthma and spirometry-defined COPD, and (4) ICD codes to determine both asthma and COPD diagnoses among patients. There is considerable variability in prevalence estimates of ACOS
with asthma[19, 21, 22, 26, 30, 32, 33, 37, 38, 41, 43-45] (Figure 3).
RI PT
for each of the definitions in patients with COPD[19-22, 24, 26-28, 30, 33-42] (Figure 2) and in patients
The ACOS prevalence estimate ranged from 25.0% to 41.4% when ACOS was defined as a self-report of a
SC
physician diagnosis of both asthma and COPD (1).[26, 30] When ACOS was defined as a reported
physician diagnosis of asthma and spirometry-defined COPD, the ACOS prevalence estimates in patients
M AN U
with COPD ranged from 12.1% to 32.9% (2).[20, 24, 27, 34-37, 39] Prevalence estimates for ACOS in patients with COPD from varying geographic regions using spirometry-defined asthma and spirometrydefined COPD (3) produced the widest range of all subgroups at 13.0–55.2%.[21, 22, 28, 33, 38] In studies that used ICD codes for both the COPD and asthma diagnoses (4),[19, 40-42] the range of ACOS
TE D
prevalence estimates in patients with COPD was generally higher than when ACOS was defined as selfreported asthma and spirometry-defined COPD (26.0–54.6% versus 12.1–32.9%, respectively, Figure
EP
2).[20, 24, 27, 34-37, 39]
The same high variability also exists when examining ACOS prevalence among patients with asthma,
AC C
again, with variability depending on the asthma and COPD definitions used (Table 3, Figure 3). When ACOS was defined by the combination of self-reported asthma and spirometry-defined COPD (2), the ACOS prevalence estimates ranged from 13.3% to 43.0%.[32, 37, 43, 44] When defining ACOS as a selfreported physician diagnosis of COPD and a self-reported physician asthma diagnosis (1), the ACOS prevalence estimates had a wider range (16.0–61.0%)[26, 30] than the ACOS estimates using spirometric criteria for both (3) asthma and COPD (17.0–51.4%).[21, 22, 33, 38, 45] Two disparate prevalence
21
ACCEPTED MANUSCRIPT
estimates (29.8% and 45.9%) of ACOS among patients with asthma were obtained from the studies that used ICD codes (4).[19, 41]
RI PT
Differences in ACOS prevalence estimates due to age of the study cohort
As ACOS is more prevalent in the population over 40 years of age and has a relatively low prevalence in younger age groups, the age of the study population can affect the estimated ACOS prevalence. For
SC
example, in a study from Finland 0.4% of patients with ACOS were ≤40 years of age and 78.2% of
patients with ACOS were over the age of 55 years.[21] In a second study performed in Italy, 1.6% of
M AN U
patients aged 20–44 and 2.1% of patients aged 45–64 years had ACOS, compared with 4.5% of patients aged 65–84 years.[26] Thus, inclusion of younger patients in a study can reduce the estimated ACOS prevalence.
TE D
Differences in estimates due to inclusion of smoking
Despite differences in COPD definitions across studies (SPIROMICS, Genetic Epidemiology of COPD [COPDGene] and Burden of Obstructive Lung Disease [BOLD]), smoking seems to be associated with an
EP
increase in ACOS prevalence estimates. The COPDGene and SPIROMICS studies examined the ACOS prevalence among smokers with COPD and found the prevalence of ACOS to be 12.6% and 16.7%,
AC C
respectively.[20, 24] However, studies estimating the prevalence of ACOS in nonsmokers are limited. Nonsmokers and never smokers with COPD may represent an important component of the ACOS population and provide information on ACOS population characteristics and risk factors. Nonsmokers (fewer than 10 pack-years) who cannot achieve adequate asthma control may often represent a portion of patients with ACOS, as they are likely to have undiagnosed COPD symptoms.[3, 51] Furthermore, COPD among never smokers is often under-diagnosed, in part because there is little understanding of the risk factors of COPD among never smokers.[13]
22
ACCEPTED MANUSCRIPT
Target populations for future research In order to understand the phenotypes better of asthma, COPD, and ACOS, future studies should place emphasis on nonsmokers with COPD and assess whether ACOS should be defined as COPD with a history
RI PT
of asthma, COPD concurrent with asthma, or both. It would also be clinically important to determine if smokers who receive an early diagnosis of asthma (before the age of 35 years) have a greater risk of developing COPD than nonsmokers with an early diagnosis of asthma. Furthermore, development of
SC
distinct phenotypes and biomarkers should be prioritized not only to improve the diagnosis and
definition of ACOS, but also to understand the pathogenesis of ACOS. Our review of ACOS literature
M AN U
shows that patients may have identical physiology in terms of FEV1/FVC ratios and reversibility, but these patients may have very different underlying pathology or different treatment responses to medication. In order to make progress in the treatment and prevention of ACOS, we need to move from an emphasis on using physiology to predict treatment responses to understanding the underlying
need to be established.
EP
Conclusions
TE D
pathology of the disease. Additionally, comprehensive therapeutic interventions for patients with ACOS
Efforts to estimate the prevalence of ACOS on the basis of physiologic measures, such as FEV1, FEV1/FVC
AC C
ratio and degree of reversibility, are misleading because these criteria do not always clearly separate asthma and COPD, and therefore, do not help in defining the ACOS population. We recommend that at a minimum, future investigations on ACOS include a detailed smoking history, the basis of the diagnosis of asthma, the degree of reversibility and whether this was measured on or off treatment so that a uniform dataset is available to provide a basis for comparison of results among studies and guide decision making on the most appropriate criteria for defining ACOS.
23
ACCEPTED MANUSCRIPT
Role of the Funding Source The work presented here, including the conduct of the literature review and analysis, was supported by
RI PT
GSK.
Acknowledgments
Editorial support (provided by Joann Hettasch, PhD, and Natasha Thomas, PhD, Fishawack Indicia Ltd)
SC
was funded by GSK.
M AN U
Author Contributions
All authors contributed to the writing of the paper and participated in the review and interpretation of
AC C
EP
TE D
the data. All authors read and approved the final manuscript.
24
ACCEPTED MANUSCRIPT
References [1]
B. J. Carolan, E. R. Sutherland, Clinical phenotypes of chronic obstructive pulmonary disease and
asthma: recent advances, J. Allergy. Clin. Immunol. 131 (2013) 627. S. R. Kim, Y. K. Rhee, Overlap between asthma and COPD: where the two diseases converge,
Allergy. Asthma. Immunol. Res. 2 (2010) 209. [3]
RI PT
[2]
S. Louie, A. A. Zeki, M. Schivo et al, The asthma-chronic obstructive pulmonary disease overlap
[4]
SC
syndrome: pharmacotherapeutic considerations, Expert. Rev. Clin. Pharmacol. 6 (2013) 197.
M. O. Nakawah, C. Hawkins, F. Barbandi, Asthma, chronic obstructive pulmonary disease
[5]
M AN U
(COPD), and the overlap syndrome, J. Am. Board. Fam. Med. 26 (2013) 470.
Global Initiative for Chronic Obstructive Lung Disease (GOLD), Global strategy for the diagnosis,
management, and prevention of chronic obstructive pulmonary disease, 2014, Available at http://www.goldcopd.org/uploads/users/files/GOLD_Report_2014_Jan23.pdf . T. C. O'Shaughnessy, T. W. Ansari, N. C. Barnes et al, Inflammation in bronchial biopsies of
TE D
[6]
subjects with chronic bronchitis: inverse relationship of CD8+ T lymphocytes with FEV1, Am. J. Respir. Crit. Care. Med. 155 (1997) 852.
M. Saetta, A. Di Stefano, G. Turato et al, CD8+ T-lymphocytes in peripheral airways of smokers
EP
[7]
with chronic obstructive pulmonary disease, Am. J. Respir. Crit. Care. Med. 157 (1998) 822. D. Stanescu, A. Sanna, C. Veriter et al, Airways obstruction, chronic expectoration, and rapid
AC C
[8]
decline of FEV1 in smokers are associated with increased levels of sputum neutrophils, Thorax. 51 (1996) 267. [9]
M. Bafadhel, S. McKenna, S. Terry et al, Acute exacerbations of chronic obstructive pulmonary
disease: identification of biologic clusters and their biomarkers, Am. J. Respir. Crit. Care. Med. 184 (2011) 662.
25
ACCEPTED MANUSCRIPT
[10]
L. M. Fabbri, M. Romagnoli, L. Corbetta et al, Differences in airway inflammation in patients with
fixed airflow obstruction due to asthma or chronic obstructive pulmonary disease, Am. J. Respir. Crit. Care. Med. 167 (2003) 418. Global Initiative for Asthma, From the Global Strategy for Asthma Management and Prevention,
RI PT
[11]
2015, Available at http://www.ginasthma.org/local/uploads/files/GINA_Report_2015.pdf. [12]
C. E. Behrendt, Mild and moderate-to-severe COPD in nonsmokers: distinct demographic
[13]
SC
profiles, Chest. 128 (2005) 1239.
S. S. Salvi and P. J. Barnes, Chronic obstructive pulmonary disease in non-smokers, Lancet. 374
[14]
D. C. Cowan, J. O. Cowan, R. Palmay et al, Effects of steroid therapy on inflammatory cell
subtypes in asthma, Thorax. 65 (2010) 384. [15]
M AN U
(2009) 733.
P. G. Gibson, J. L. Simpson, N. Saltos, Heterogeneity of airway inflammation in persistent asthma
[16]
TE D
: evidence of neutrophilic inflammation and increased sputum interleukin-8, Chest. 119 (2001) 1329. A. Jatakanon, C. Uasuf, W. Maziak et al., Neutrophilic inflammation in severe persistent asthma,
Am. J. Respir. Crit. Care. Med. 160 (1999) 1532.
C. E. Brightling, S. McKenna, B. Hargadon et al, Sputum eosinophilia and the short term response
EP
[17]
to inhaled mometasone in chronic obstructive pulmonary disease, Thorax. 60 (2005) 193. R. Leigh, M. M. Pizzichini, M. M. Morris et al, Stable COPD: predicting benefit from high-dose
AC C
[18]
inhaled corticosteroid treatment, Eur. Respir. J. 27 (2006) 964. [19]
H. Andersen, P. Lampela, A. Nevanlinna et al, High hospital burden in overlap syndrome of
asthma and COPD, Clin. Respir. J. 7 (2013) 342. [20]
M. Hardin, M. Cho, M. L. McDonald et al, The clinical and genetic features of COPD-asthma
overlap syndrome, Eur. Respir. J. 44 (2014) 341.
26
ACCEPTED MANUSCRIPT
[21]
P. Kauppi, H. Kupiainen, A. Lindqvist et al, Overlap syndrome of asthma and COPD predicts low
quality of life, J. Asthma. 48 (2011) 279. [22]
A. M. Menezes, M. Montes de Oca, R. Perez-Padilla et al, Increased risk of exacerbation and
[23]
RI PT
hospitalization in subjects with an overlap phenotype: COPD-asthma, Chest. 145 (2014) 297.
M. Miravitlles, J. B. Soriano, J. Ancochea et al., Characterisation of the overlap COPD-asthma
phenotype. Focus on physical activity and health status, Respir. Med. 107 (2013) 1053.
N. Putcha, M. B. Drummond, C. H. Martinez et al, Self-reported asthma is associated with worse
SC
[24]
health outcomes in COPD, Am. J. Respir. Crit. Care. Med. (2013) A3947.
A. A. Zeki, M. Schivo, A. Chan et al., The asthma-COPD overlap syndrome: a common clinical
M AN U
[25]
problem in the elderly, J. Allergy. 2011 (2011) 861926. [26]
R. de Marco, G. Pesce, A. Marcon et al, The coexistence of asthma and chronic obstructive
pulmonary disease (COPD): prevalence and risk factors in young, middle-aged and elderly people from
[27]
TE D
the general population, PloS. One. 8 (2013) e62985.
J. L. Izquierdo-Alonso, J. M. Rodriguez-Gonzalezmoro, P. de Lucas-Ramos et al, Prevalence and
characteristics of three clinical phenotypes of chronic obstructive pulmonary disease (COPD), Respir.
[28]
EP
Med. 107 (2013) 724.
S. E. Marsh, J. Travers, M. Weatherall et al, Proportional classifications of COPD phenotypes,
[29]
AC C
Thorax. 63 (2008) 761.
H. Melbye, E. Drivenes, L. G. Dalbak et al, Asthma, chronic obstructive pulmonary disease, or
both? Diagnostic labeling and spirometry in primary care patients aged 40 years or more, Int. J .COPD. 6 (2011) 597. [30]
R. A. Pleasants, J. A. Ohar, J. B. Croft et al, Chronic obstructive pulmonary disease and asthma-
patient characteristics and health impairment, COPD. 11 (2014) 256.
27
ACCEPTED MANUSCRIPT
[31]
J. J. Soler-Cataluna, B. Cosio, J. L. Izquierdo et al, Consensus document on the overlap phenotype
COPD-asthma in COPD, Arch. Bronconeumol. 48 (2012) 331. E. Diaz-Guzman, M. Khosravi, D. M. Mannino, Asthma, chronic obstructive pulmonary disease,
and mortality in the U.S. population, COPD. 8 (2011) 400. [33]
K. A. Kong, J. H. Lee, J. W. Chung et al, Health-related quality of life in airway diseases, Am. J.
Respir. Crit. Care. Med. 187 (2013) A5554.
M. Hardin, E. K. Silverman, R. G. Barr et al, The clinical features of the overlap between COPD
SC
[34]
and asthma, Respir. Res. 12 (2011) 127.
K. E. Wurst, A. Shukla, H. Muellerova et al, Respiratory pharmacotherapy use in patients newly
M AN U
[35]
RI PT
[32]
diagnosed with chronic obstructive pulmonary disease in a primary care setting in the UK: a retrospective cohort study, COPD. 11 (2014) 521. [36]
C. Talamo, M. M. de Oca, R. Halbert et al., Diagnostic labeling of COPD in five Latin American
[37]
TE D
cities, Chest. 131 (2007) 60.
B. Lamprecht, M. A. McBurnie, W. M. Vollmer et al, COPD in never smokers: results from the
population-based burden of obstructive lung disease study, Chest. 139 (2011) 752. H. Iwamoto, J. Gao, J. Koskela et al, Differences in plasma and sputum biomarkers between
EP
[38]
COPD and COPD-asthma overlap, Eur. Respir. J. 43 (2014) 421. Y. Kitaguchi, Y. Komatsu, K. Fujimoto et al, Sputum eosinophilia can predict responsiveness to
AC C
[39]
inhaled corticosteroid treatment in patients with overlap syndrome of COPD and asthma, Int. J. COPD. 7 (2012) 283. [40]
C. M. Blanchette, M. Broder, C. Ory et al., Cost and utilization of COPD and asthma among
insured adults in the US, Curr. Med. Res. Opin. 25 (2009) 1385. [41]
F. T. Shaya, D. Dongyi, M. O. Akazawa et al., Burden of concomitant asthma and COPD in a
medicaid population, Chest. 134 (2008) 14.
28
ACCEPTED MANUSCRIPT
[42]
C. K. Rhee, H. K. Yoon, K. H. Yoo et al., Medical utilization and cost in patients with overlap
syndrome of chronic obstructive pulmonary disease and asthma, COPD. 11 (2014) 163. [43]
C. Kalberg, W. Sense, K. Knobil, Fixed airway obstruction in patients with asthma: data from a
[44]
RI PT
large clinical trials database, Proc. Am. Thor. Soc. 2 (abstracts issue) (2005) B95.
A. Banga, K. McCarthy, B.M. Pichurko, Predictors of airway obstruction on spirometry among
patients with bronchial asthma presenting to a tertiary care center, Am. J. Respir. Crit. Care. Med187
[45]
A. H. Wagener, D. Gibeon, X. Yang et al, Associated factors persistent airflow limitation in
M AN U
asthma in U-BIOPRED Eur, Respir. J. 42 (Suppl 52) (2013) P3142. [46]
J. Chang, Z. Mosenifar, Differentiating COPD from asthma in clinical practice, J. Intensive. Care.
Med. 22 (2007) 300. [47]
SC
(2013) A5009.
P. M. Calverley, P. S. Burge, S. Spencer et al, Bronchodilator reversibility testing in chronic
[48]
TE D
obstructive pulmonary disease, Thorax. 58 (2003) 659.
P. Albert, A. Agusti, L. Edwards et al, Bronchodilator responsiveness as a phenotypic
characteristic of established chronic obstructive pulmonary disease, Thorax. 67 (2012) 701. S. H. Landis, H. Muellerova, D. M. Mannino et al, Continuing to Confront COPD International
EP
[49]
Patient Survey: methods, COPD prevalence, and disease burden in 2012-2013, Int. J. COPD. 9 (2014) 597. E. S. Ford, D. M. Mannino, Time trends in obesity among adults with asthma in the United
AC C
[50]
States: findings from three national survey, J. Asthma. 42 (2005) 91. [51]
E. D. Bateman, C. Feldman, J. O'Brien et al, Guideline for the management of chronic obstructive
pulmonary disease (COPD): 2004 revision S. Afr. Med. J. 94 (2004) 559.
29
ACCEPTED MANUSCRIPT
Figure legends
Figure 1. COPD, asthma, and ACOS prevalence among total populations (corresponding to Table 1)
RI PT
ACOS: asthma-COPD overlap syndrome; COPD: chronic obstructive pulmonary disease. Figure 2. ACOS prevalence among patients with COPD (corresponding to Table 2)
Asthma - Diagnosis/COPD - Spirometry: studies with a reported physician diagnosis of asthma and
spirometry defined COPD; Asthma - Spirometry/COPD - Spirometry: studies with spirometry defined
SC
asthma and spirometry defined COPD; Asthma - Diagnosis/COPD - Diagnosis: studies with a reported
M AN U
physician diagnosis of asthma and a reported physician diagnosis of COPD at any point in a patient’s life; Asthma - ICD Code/COPD - ICD Code: studies with ICD code classification for both asthma and COPD. ACOS: asthma-COPD overlap syndrome; COPD: chronic obstructive pulmonary disease; ICD, International Classification of Diseases.
Figure 3. ACOS prevalence among patients with asthma (corresponding to Table 3)
TE D
Asthma - Diagnosis/COPD - Spirometry: studies with a reported physician diagnosis of asthma and spirometry defined COPD; Asthma - Spirometry/COPD - Spirometry: studies with spirometry defined asthma and spirometry defined COPD; Asthma - Diagnosis/COPD - Diagnosis: studies with a reported
EP
physician diagnosis of asthma and a reported physician diagnosis of COPD at any point in a patient’s life; Asthma - ICD Code/COPD - ICD Code: studies with ICD code classification for both asthma and COPD.
AC C
ACOS: asthma - COPD overlap syndrome; COPD: chronic obstructive pulmonary disease; ICD, International Classification of Diseases.
30
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
11.7
12.0
13.3 12.1
EP
14.0
10.6
AC C
10.0
8.0
6.0
8.2 7.8
6.1 5.7
5.6
5.3 5.3 4.9 4.5
4.0
3.3
2.9
2.7
2.9 2.4
2.0
2.4
2.1
1.7 1.8
1.6
Diaz-Guzman et al. 2011
Menezes et al. 2014
Menezes et al. 2014
Lee et al. 2013
de Marco et al. 2013
de Marco et al. 2013
de Marco et al. 2013
ACOS
COPD-only
Asthma-only
ACOS
Asthma-only
COPD-only
ACOS
Asthma-only
COPD-only
ACOS
COPD-only
Asthma-only
ACOS
Asthma-only
COPD-only
ACOS
Asthma-only
COPD-only
ACOS
Asthma-only
COPD-only
ACOS
Asthma-only
0.0
COPD-only
Proportion (%)
8.9
Pleasants et al. 2014
Izq
ui e
Al
o-
on
so et Ha al . rdi 20 ne 13 ta Ha l. 2 rdi 01 ne 4 Me ta ne l. ze 20 se 11 ta Pu l. 20 tch 14 ae ta Ta l . lam 20 13 oe La t mp al. rec 20 07 ht et al. Le 20 ee 11 de ta Ma l . 20 rco 13 et An al . de 20 rse 13 ne Iw t a am l. 2 oto 01 3 de et al . Ma 20 rco 14 et Ki al. tag uc 20 hi 13 Bla et al . nc he 20 tte 12 et a l. 2 W urs 00 te 9 de ta Ma l. 2 rco 01 4 et Ple al . as 20 an 13 ts et al. Sh 20 ay 14 ae t al. Ka up 20 pi 08 et a l. 2 Rh 01 ee 1 et a Ma l. 2 rsh 01 4 et al. 20 08
rd
Proportion (%) 50.0
40.0
EP
60.0
30.0
22.8% 23.5% 23.6%
AC C
Asthma - Diagnosis / COPD - Spirometry
Asthma - Spirometry / COPD - Spirometry
Asthma - Diagnosis / COPD - Diagnosis
Asthma - ICD Code / COPD - ICD Code
20.0
12.1% 12.6%
TE D M AN U SC
RI PT
ACCEPTED MANUSCRIPT
54.6% 55.2%
49.0%
41.4% 43.0%
32.9% 33.0%
26.0% 26.4% 27.0% 27.0%
30.2%
25.0%
16.7%
13.0% 13.0%
10.0
0.0
Di
ne ta de l. 2 Ma 01 rco 1 et al. 20 Ka 13 up pi et al. Ka 20 l be 11 rg et La al . mp 20 rec 05 ht et al. Ba 20 ng 11 ae ta Ple l. as 20 an 13 ts et al. 20 14 Le ee ta l. 2 Ka 01 l be 3 rg et al . An 20 de 05 rse ne ta de l. 2 Ma 01 rco 3 et al . Iw 20 am 13 oto et al. Ka 20 l be 14 rg et al . 20 Sh 05 ay ae ta W l. 2 ag 00 en 8 er et Me al. 20 ne 13 ze s et al . de Ma 20 14 rco et al. 20 13
ma
-G uz
az
Proportion (%)
40.0
30.0
24.9%
AC C
60.0
50.0
EP
70.0 Asthma - Spirometry / COPD - Spirometry
Asthma - Diagnosis / COPD - Diagnosis
TE D M AN U SC
RI PT
ACCEPTED MANUSCRIPT
Asthma - Diagnosis / COPD - Spirometry
20.0
16.0%
17.0% 61.0%
Asthma - ICD Code / COPD - ICD Code
51.4%
43.0% 45.9% 46.0%
30.0% 30.4% 29.0% 29.8% 27.5% 27.8% 28.2%
18.0%
13.3%
10.0
0.0
ACCEPTED MANUSCRIPT Highlights Prevalence of ACOS varies widely (12–61%) among patients with COPD or asthma
•
The variability is linked to differences in COPD and asthma diagnostic criteria
•
Other factors linked to variability include age and gender of the study population
•
Disease burden in patients with ACOS is greater than in asthma or COPD alone
AC C
EP
TE D
M AN U
SC
RI PT
•
ACCEPTED MANUSCRIPT
KW was an employee of GSK at the time the study was performed and is currently employed by PAREXEL. SP and NB are employees of GSK and own stocks/shares in GSK.
AC C
EP
TE D
M AN U
SC
RI PT
KK-R and GB hold graduate research assistant positions at The University of North Carolina.