Undiagnosed Chronic Obstructive Pulmonary Disease ... - ATS Journals

&get_box_var;

ORIGINAL ARTICLE

Undiagnosed Chronic Obstructive Pulmonary Disease Contributes to the Burden of Health Care Use Data from the CanCOLD Study Laura E. Labonte´ 1, Wan C. Tan2, Pei Z. Li1, Palmina Mancino1, Shawn D. Aaron3, Andrea Benedetti1,4, Kenneth R. Chapman5, Robert Cowie6, J. Mark FitzGerald2, Paul Hernandez7, François Maltais8, Darcy D. Marciniuk9, Dennis O’Donnell10, Don Sin2, and Jean Bourbeau1; for the Canadian Respiratory Research Network and the CanCOLD Collaborative Research Group* 1 Respiratory Epidemiology and Clinical Research Unit, Research Institute of the McGill University Health Centre, and 4Department of Epidemiology, Biostatistics and Occupational Health, McGill University, Montreal, Quebec, Canada; 2University of British Columbia, Vancouver, British Columbia, Canada; 3Ottawa Hospital Research Institute, Ottawa, Ontario, Canada; 5University of Toronto, Toronto, Ontario, Canada; 6University of Calgary, Calgary, Alberta, Canada; 7Dalhousie University, Halifax, Nova Scotia, Canada; 8Institut Universitaire de Cardiologie et de Pneumologie de Quebec, ´ Universite´ Laval, Quebec, ´ Quebec, ´ Canada; 9University of Saskatchewan, Saskatoon, Saskatchewan, Canada; and 10Queens University, Kingston, Ontario, Canada

Abstract Rationale: Chronic obstructive pulmonary disease (COPD) remains

undiagnosed in many individuals with persistent airflow limitation. These individuals may be susceptible to exacerbation-like respiratory events that consume health care resources. Objectives: To compare exacerbation-like respiratory events, event

prevalence, and differences in the odds of using medication and/or health services between subjects with diagnosed and undiagnosed COPD. Methods: Subjects sampled from the general population

participating in the CanCOLD (Canadian Cohort Obstructive Lung Disease) study, with at least 12 months of exacerbation-event follow-up who were classified as having physician-diagnosed or undiagnosed COPD were assessed. Exacerbation-like respiratory events were captured using a questionnaire administered every 3 months.

Measurements and Main Results: A total of 355 subjects were undiagnosed and 150 were diagnosed with COPD. Undiagnosed subjects were less symptomatic and functionally impaired, had been prescribed fewer respiratory medications, and had better health status. The incidence of reported exacerbation-like events was higher in diagnosed subjects and increased in both groups with the severity of airflow obstruction. Although subjects with diagnosed COPD were more often prescribed medication for exacerbation events, health service use for exacerbation events was similar in both groups. Conclusions: Most subjects with COPD in Canada remain

undiagnosed. These subjects are less symptomatic and impaired, which may partly explain lack of diagnosis. Although patients with undiagnosed COPD experience fewer exacerbations than those with diagnosed COPD, they use a similar amount of health services for exacerbation events; thus, the overall health system burden of exacerbations in those with undiagnosed COPD is considerable. Keywords: undiagnosed COPD; exacerbation; burden

( Received in original form September 11, 2015; accepted in final form January 29, 2016 ) *A complete list of members may be found before the beginning of the REFERENCES. The Canadian Cohort Obstructive Lung Disease (CanCOLD) study is currently funded by the Canadian Respiratory Research Network (industry partners: Astra Zeneca Canada Ltd., Boehringer Ingelheim Canada Ltd., GlaxoSmithKline Canada Ltd., and Novartis). Researchers at RI-MUHC Montreal and Icapture Centre Vancouver lead the project. Previous funding partners are the CIHR (CIHR/Rx&D Collaborative Research Program Operating Grants- 93326) and the Respiratory Health Network of the FRSQ (industry partners: Almirall, Merck Nycomed, Pfizer Canada Ltd., and Theratechnologies). The funders had no role in the study design, data collection and analysis, or preparation of the manuscript. Author Contributions: L.E.L. contributed to the conception and implementation of the study, the analysis of the data, and the writing of the manuscript. J.B. contributed to the study conception and design, implementation and acquisition of the data, and the writing and revision of the article. W.C.T., S.D.A., A.B., K.R.C., R.C., J.M.F., P.H., F.M., D.D.M., D.O., and D.S. contributed to the acquisition of the data and the revision of the article. P.Z.L. and P.M. contributed to the analysis and interpretation of the data. All authors approved the final version of the manuscript. Correspondence and requests for reprints should be addressed to Jean Bourbeau, M.D., Center for Innovative Medicine, McGill University Health Centre, 1001 Decarie Boulevard, Room C047371.5, Montreal, PQ, H4A 3J1 Canada. E-mail: [email protected] This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org Am J Respir Crit Care Med Vol 194, Iss 3, pp 285–298, Aug 1, 2016 Copyright © 2016 by the American Thoracic Society Originally Published in Press as DOI: 10.1164/rccm.201509-1795OC on February 2, 2016 Internet address: www.atsjournals.org

Labonte, ´ Tan, Li, et al.: Health Care Burden of Undiagnosed COPD

285

ORIGINAL ARTICLE

At a Glance Commentary Scientific Knowledge on the Subject: Chronic obstructive

pulmonary disease (COPD) exacerbations represent an important determinant of the overall burden of COPD. The incidence and impact of exacerbation events in persons with undiagnosed COPD within the general population is unknown. What This Study Adds to the Field: This study shows for the first

time that despite experiencing fewer exacerbations, health care use to treat exacerbation-like events in undiagnosed individuals with COPD is similar to that of diagnosed individuals. Consequently, COPD exacerbation events contribute much more than previously thought to the overall burden of COPD. The natural course of chronic obstructive pulmonary disease (COPD) is perturbed by episodes of acute symptom worsening, known as exacerbations (1). These events are associated with accelerated lung function decline (2), impaired health status (3, 4), increased hospitalization (5), and increased mortality (6); as a result, they contribute greatly to the increasing cost and burden of COPD (7). Exacerbations are most commonly caused by viral and/or bacterial infections, and become more frequent as disease progresses (2, 8, 9). The longitudinal ECLIPSE cohort study showed that, among patients with COPD, there is a stable “exacerbation-susceptibility” phenotype (10). Accordingly, the best way of identifying subjects susceptible to exacerbations is through their exacerbation history, where frequent exacerbations predict risk of future events. To date, however, most studies on exacerbations have focused on diagnosed and treated patients. As a result, there is a lack of knowledge and understanding of exacerbations in people with undiagnosed COPD, who may have milder and less symptomatic disease, but may still be affected by the consequences of these acute respiratory events. Hill and colleagues (11) found that in the Canadian primary care setting, approximately 21% of patients aged 40 years and older with at least 20 years of 286

smoking history had spirometry consistent with COPD, but among these, only 30% were previously diagnosed. Patients with chronic airflow limitation uninformed about their disease and exacerbation management may nonetheless experience recurrent exacerbation-like events leading to increasing emergency department visits and hospitalizations (12). Little is known about subjects with undiagnosed COPD in the general population and their potential burden on the health care system. A recent study by Gershon and colleagues (13) using CanCOLD (Canadian Cohort Obstructive Lung Disease study) linked to the health administrative databases, limited to subjects from three sites in Ontario, has shown that overdiagnosed but not undiagnosed COPD seems to place significant burden on the health care system. However, it is unknown whether subjects with undiagnosed COPD and who exacerbate experience poor outcomes compared with those with physiciandiagnosed COPD and exacerbations. Being able to recognize characteristics associated with an inherent susceptibility to exacerbations (10) in undiagnosed subjects could help design and implement new approaches to identify these individuals for early disease management to reduce exacerbations and related complications, and providing new insight into why these individuals remain undiagnosed. In our study, we evaluated subjects aged 40 years and older participating in the population-based CanCOLD study who had spirometrically defined airflow obstruction to compare the demographic and clinical characteristics of the two subgroups, with or without a prior self-reported physician diagnosis of COPD. The primary objective of the study was to compare the incidence of symptom-based and event-based exacerbationlike respiratory events in these subgroups. Lastly, we determined the clinical factors, patient-reported outcomes, medication, and health service use associated with exacerbationlike events in these subgroups. Some of the results of this study have been previously reported in the form of abstracts (14, 15).

Methods Study Design and Subjects

Subjects were sampled from the general population to participate in the CanCOLD

study. The CanCOLD study was built on the cross-sectional population-based COLD study, which use the study protocol from the global Burden of Obstructive Lung Disease initiative (16). The details of the study design and protocol of the CanCOLD study have been previously published (17). For the purpose of this analysis, we selected from the entire CanCOLD cohort a subset of subjects with at least 12 months of exacerbation follow-up using an exacerbation questionnaire. All subjects had spirometrically defined COPD at study entry; however, only a minority had received a previous physician diagnosis of COPD. All patients provided written informed consent, and the study was approved by the relevant ethics and review boards. Full subject assessments were performed every 18 months (for full details, see Reference 17). Additionally, a questionnaire administered by telephone interview or online every 3 months was used to capture exacerbation-like respiratory events. Physician Diagnosed and Undiagnosed COPD

Subjects with spirometrically defined COPD who reported having received a previous physician-diagnosis of COPD, chronic bronchitis, or emphysema on entering the CanCOLD study were identified as “diagnosed” subjects, and subjects with spirometrically defined COPD, who had not received a diagnosis of COPD before entry in the CanCOLD study, were identified as having “undiagnosed” COPD. Exacerbation Definition and Method of Measurement

Exacerbation-like respiratory events were the main outcome for this analysis. In the CanCOLD study, two different operational definitions were used (Table 1). The definition could be symptom-based, requiring a change in at least one major symptom (dyspnea, sputum purulence, sputum volume), or event-based, requiring change of at least one major symptom and use of antibiotics and/or systemic corticosteroids or health services. The purpose of using both definitions was to capture exacerbation-like respiratory events, with varying levels of severity (receiving treatment or not). The self-administered online exacerbation questionnaire included the

American Journal of Respiratory and Critical Care Medicine Volume 194 Number 3 | August 1 2016

ORIGINAL ARTICLE Table 1. Exacerbation Definitions and Methods of Capture Recall period COPD exacerbation definition

3 mo Symptom-based: The presence of at least one major symptom (increased dyspnea, increased sputum volume, or increased sputum purulence) for at least 48 h

Cohort Method to capture exacerbations Question asked to capture exacerbations

CanCOLD Telephone interview/online questionnaire administered every 3 mo (recall of last 3 mo)

Event-based: The presence of at least one major symptom (increased dyspnea, increased sputum volume, or increased sputum purulence) for at least 48 h and use of antibiotics or corticosteroids or health care use (ED or doctor visit, hospitalization)

In the last 3 mo (or since this questionnaire was last administered) have you experienced an episode with new or changes in any respiratory symptoms (i.e., cough, phlegm, wheeze, breathlessness) that became worse for at least 2 d? If yes, subject completes the following sections: 1. Respiratory symptoms 2. Medication (due to respiratory symptoms) 3. Work (affected by respiratory symptoms) 4. Health care visits (due to respiratory symptoms)

Definition of abbreviations: CanCOLD = Canadian Cohort Obstructive Lung Disease; COPD = chronic obstructive pulmonary disease; ED = emergency department.

first question of the telephone interview and was used to identify subjects who experienced an exacerbation-like respiratory event. To ensure consistency, if a subject did experience an event, they were contacted and administered the entire questionnaire via telephone interview. This questionnaire included questions on respiratory symptoms and their duration, medication use, impact on work (absenteeism and presenteeism), and health care resource use. It was developed to minimize recall bias, based on previously used questionnaires that measured exacerbations in large cohorts (see Figure E1 in the online supplement) (18–21). If subjects reported having experienced at least one major symptom that persisted at least 2 days, this was identified as an exacerbation-like respiratory event, which was then classified as being either symptom-based or event-based using criteria outlined in Table 1. A second exacerbation-like event was considered to have occurred when subjects experienced at least 3 days without any symptoms (back to baseline) between events. Statistical Analyses

Descriptive statistics are shown as counts and percentages for categorical data and means and SD for continuous variables. When comparing subgroups, we used Student’s t tests (normal continuous variables) or Wilcoxon signed rank test (not normal continuous variables) or chi-square test (categorical variables), two-tailed in

nature. We considered a P value of 0.05 or less to be significant. Only spirometric data that fulfilled the American Thoracic Society acceptability and repeatability criteria were used for analyses (22). To determine factors associated with reporting exacerbation-like events, determinants, and impact/outcomes of susceptibility to exacerbation-like respiratory events, multivariable logistic regression analyses were used. Adjusted odd ratios and 95% confidence intervals were calculated adjusting for sex, age, body mass index, and smoking history. Statistically significant factors were included in a stepwise multiple variable model. Analyses were performed using statistical software (Statistical Analysis Software, version 9.1; SAS Institute, Cary, NC).

Results From a total of 1,532 CanCOLD participants, there were 712 individuals with COPD, including 505 subjects with COPD with 12 months of exacerbation-like respiratory event follow-up and 207 in whom 12 months of follow-up was not completed; their characteristics are presented in Table 2. The 505 subjects with COPD who are the subject of this report included 150 patients with physician-diagnosed COPD and 355 individuals with undiagnosed COPD; that is, two-thirds of subjects with spirometrically defined COPD had not been previously diagnosed by a physician.


Characteristics of Subjects with Undiagnosed and Diagnosed COPD at Study Entry

Table 3 shows demographic, lung function, disease status, and clinical characteristics at study entry for all subjects with COPD with 12 months of follow-up and for those reporting exacerbations during prospective follow-up according to whether they were diagnosed or undiagnosed with COPD. Compared with diagnosed subjects with COPD, undiagnosed subjects included greater percentage of men and fewer eversmokers. They also had lower lifetime exposures to tobacco smoking, better lung function and less severe COPD, fewer comorbidities, and were prescribed fewer respiratory medications. Among subjects who reported exacerbation-like respiratory events during follow-up, undiagnosed subjects with COPD tended to be less dyspneic, have better health status, better physical status, and less anxiety than diagnosed subjects. Conclusions regarding the comparison of diagnosed and undiagnosed subjects were similar whether or not subjects who reported receiving a previous physician diagnosis of asthma (n = 162) were included in the analysis (see Table E2). Incidence of Exacerbation-like Respiratory Events

The overall rate of exacerbation-like respiratory events reported over 12 months of follow-up is shown in Table 4. Figure 1 shows the proportion of subjects who 287

ORIGINAL ARTICLE Table 2. Characteristics at Study Entry of Patients with COPD with 12 Months of Exacerbation-like Respiratory Event Follow-up Completed versus Not Completed

Subjects, n (%) Demographic characteristics Age, yr, mean (SD) Male sex, n (%) Ever-smokers, n (%) Cigarette smoker pack-years, mean (SD) Lung function and disease status Post-bronchodilator spirometry, mean (SD) FEV1, L FEV1, % predicted FEV1/FVC, % GOLD classification (fixed FEV1/FVC), n (%)† GOLD 1 GOLD 21 Clinical characteristics Self-reported comorbidities, n (%) CVD Angina Myocardial infarction Diabetes Musculoskeletal Reflux/heartburn Allergy Generalized anxiety disorder Major depression Physician-diagnosed COPD Physician-diagnosed asthma Respiratory medications reported in the past 12 mo, n (%) SABD LABA or LAMA ICS alone ICS combined with LABA/LAMA Any above medications Patient-reported outcomes, mean (SD) BOD score MMRC dyspnea scale, 1–5 CAT score SGRQ total score COPD-specific module total score SF-36V2 Physical component scale Mental component scale HADS Anxiety score Depression score CHAMPS caloric expenditure per week Moderate and greater intensity, >3 MET (kcal) All activities

Total Subjects with COPD

Did Not Complete 12-Month Follow-up

Completed 12-Month Follow-up

712

207 (29.1)

505 (70.9)

67.5 129 160 24.4

67.4 297 358 21.9

67.5 426 518 22.6

(10.1) (59.8) (72.8) (24.9)

(10.1) (62.3) (77.3) (26.2)

P Value*

(10.1) (58.8) (70.9) (24.3)

0.868 0.386 0.081 0.232

2.3 (0.8) 82.3 (19.2) 61.1 (8.1)

2.3 (0.8) 82.9 (18.4) 61.6 (7.7)

2.3 (0.8) 82.1 (19.6) 60.9 (8.2)

0.998 0.674 0.269

399 (56.0) 313 (44.0)

120 (58.0) 87 (42.0)

279 (55.2) 226 (44.8)

0.506 0.506

185 34 30 69 241 168 282 25 48 196 224

(26.0) (4.8) (4.2) (9.7) (33.8) (23.6) (39.6) (3.5) (6.7) (27.5) (31.5)

48 10 7 17 67 47 72 4 10 46 62

(23.2) (4.8) (3.4) (8.2) (32.4) (22.7) (34.8) (1.9) (4.8) (22.2) (30.0)

137 24 23 52 174 121 210 21 38 150 162

(27.1) (4.8) (4.6) (10.3) (34.5) (24.0) (41.6) (4.2) (7.5) (29.7) (32.1)

0.276 0.964 0.479 0.393 0.593 0.72 0.092 0.143 0.193 0.042 0.579

47 16 58 132 253

(6.6) (2.2) (8.1) (18.5) (35.5)

10 6 20 38 74

(4.8) (2.9) (9.7) (18.4) (35.7)

37 10 38 94 179

(7.3) (2.0) (7.5) (18.6) (35.4)

0.223 0.453 0.344 0.936 0.939

0.5 1.5 7.8 15.3 9.7

(0.9) (0.7) (6.5) (15.1) (8.5)

0.5 1.5 8.4 16.2 10.6

(0.8) (0.8) (7.1) (16.2) (9.1)

0.5 1.6 7.5 14.9 9.4

(0.9) (0.7) (6.3) (14.6) (8.2)

0.695 0.617 0.179 0.432 0.119

50.5 (8.9) 50.1 (9.3)

50.3 (9.0) 49.4 (9.8)

50.5 (8.9) 50.4 (9.1)

0.635 0.223

3.8 (3.1) 2.8 (2.5)

4.0 (3.2) 2.9 (2.8)

3.7 (3.1) 2.7 (2.3)

0.145 0.986

2.3 (2.4) 4.1 (3.0)

2.1 (2.5) 4.0 (3.2)

2.3 (2.3) 4.1 (2.9)

0.128 0.528

Definition of abbreviations: BOD = body mass index, obstruction, dyspnea; CAT = COPD assessment test; CHAMPS = Community Healthy Activities Model Program for Seniors; COPD = chronic obstructive pulmonary disease; CVD = cardiovascular disease; GOLD = Global Initiative for Chronic Obstructive Lung Disease; HADS = Hospital Anxiety and Depression Scale; ICS = inhaled corticosteroids; LABA = long-acting b-agonist; LAMA = long-acting muscarinic antagonist; MET = metabolic equivalent used to express exercise intensity; MMRC = Modified Medical Research Council; SABD = short-acting bronchodilator; SF = Short Form; SGRQ = St. George’s Respiratory Questionnaire. *P values were obtained by performing Student’s t test (normal continuous variables) or Wilcoxon signed rank test (not normal continuous variables) or chi-square test (category variables). † Based on post-bronchodilator spirometry. GOLD1: post-FEV1/FVC , 0.70 and FEV1%pred > 80; GOLD21: post-FEV1/FVC , 0.70 and 80 . FEV1% pred . 0. Undiagnosed COPD based on no self-report of previous physician diagnosis of COPD before study entry, but with spirometric evidence of obstruction. Diagnosed COPD based on a self-reported previous physician diagnosis of COPD at study entry.

288


Demographic characteristics Age, yr, mean (SD) Male sex, n (%) Ever-smokers, n (%) Cigarette smoker pack-years, mean (SD) Lung function and disease status Post-bronchodilator spirometry, mean (SD) FEV1, L FEV1, % predicted FEV1/FVC, % GOLD classification (fixed FEV1/FVC), n (%)† GOLD 1 GOLD 21 Clinical characteristics Self-reported comorbidities, n (%) CVD Angina Myocardial infarction Diabetes Musculoskeletal Reflux/heartburn Allergy Generalized anxiety disorder Major depression Physician-diagnosed asthma Respiratory medications reported in the past 12 mo, n (%) SABD LABA or LAMA ICS alone ICS combined with LABA/LAMA Any above medications Patient-reported outcomes, mean (SD) BOD score MMRC dyspnea scale, 1–5 CAT score SGRQ total score COPD-specific module total score SF-36V2 Physical component scale Mental component scale

Labonte, ´ Tan, Li, et al.: Health Care Burden of Undiagnosed COPD (0.6) (0.6) (4.8) (11.0) (6.1)

52.3 (7.7) 51.0 (8.4)

0.3 1.4 5.7 10.7 7.2

80 (22.5)

(5.6) (0.0) (6.2) (10.7)

20 0 22 38

(11.3) (6.7) (10.7) (37.3)

(1.1) (0.8) (7.4) (17.3) (9.9) 46.3 (9.9) 49.0 (10.4)

0.9 1.9 11.8 24.8 14.7

99 (66.0)

17 10 16 56

45 13 10 18 66 42 79 12 16 72

(25.9) (3.1) (3.7) (9.6) (30.4) (22.3) (36.9) (2.5) (6.2) (25.4)

92 11 13 34 108 79 131 9 22 90

(30.0) (8.7) (6.7) (12.0) (44.0) (28.0) (52.7) (8.0) (10.7) (48.0)

53 (35.3) 97 (64.7)

(9.6) (48.7) (84.0) (26.8)

226 (63.7) 129 (36.3)

66.9 73 126 32.9

2.0 (0.8) 71.4 (19.3) 56.9 (10.2)

(10.3) (63.1) (65.4) (21.6)

2.5 (0.7) 86.6 (17.9) 62.6 (6.6)

67.7 224 232 17.3

Total (n = 505) Undiagnosed Diagnosed COPD COPD (n = 355 [70%]) (n = 150 [30%])

50.7 (7.9) 49.1 (8.8)

,0.001* 0.058

(0.8) (0.7) (5.3) (13.2) (7.6)

0.5 1.6 7.6 16.1 9.8

,0.001* ,0.001* ,0.001* ,0.001* ,0.001*

30 (38.5)

,0.001*

(2.6) (0.0) (11.5) (24.4)

2 0 9 19

(26.9) (1.3) (3.8) (7.7) (35.9) (16.7) (51.3) (6.4) (10.3) (43.6)

0.025* ,0.001* 0.082 ,0.001*

21 1 3 6 28 13 40 5 8 34

42 (53.8) 36 (46.2)

,0.001* ,0.001* 0.346 0.007* 0.139 0.413 0.003* 0.167 0.001* 0.005* 0.082 ,0.001*

2.3 (0.8) 81.6 (18.8) 60.6 (7.7)

,0.001* ,0.001* ,0.001*

(11.0) (55.1) (60.3) (21.5)

67.1 43 47 17.3

0.303 0.003* ,0.001* ,0.001*

P Value

(9.2) (40.0) (85.0) (32.3)

(10.0) (8.3) (11.7) (50.0)

(31.7) (16.7) (10.0) (13.3) (55.0) (35.0) (65.0) (11.7) (16.7) (58.3)

(1.4) (1.0) (8.2) (19.8) (11.5) 44.6 (9.7) 47.3 (10.5)

1.3 2.1 14.2 30.7 18.8

48 (80.0)

6 5 7 30

19 10 6 8 33 21 39 7 10 35

18 (30.0) 42 (70.0)

1.8 (0.8) 67.0 (19.2) 54.9 (11.6)

67.0 24 51 37.3

,0.001* 0.334

,0.001* 0.004* ,0.001* ,0.001* ,0.001*

,0.001*

0.078 0.014* 0.981 0.002*

0.543 0.001* 0.177 0.277 0.025* 0.013* 0.106 0.277 0.268 0.086

0.005* 0.005*

,0.001* ,0.001* 0.003*

0.857 0.078 0.001* ,0.001*

Subjects Reporting Symptom-based Exacerbation-like Respiratory Events (n = 138) Undiagnosed Diagnosed COPD COPD (n = 78 [57%]) (n = 60 [43%]) P Value

Table 3. Characteristics at Study Entry of Subjects with Undiagnosed and Diagnosed COPD

(11.2) (50.0) (58.0) (21.6)

(0.0) (0.0) (16.0) (26.0)

(4.0) (8.0) (40.0) (14.0) (42.0) (6.0) (10.0) (50.0)

(0.9) (0.8) (5.9) (14.1) (8.2) 51.3 (7.9) 49.8 (8.6)

0.5 1.7 7.5 17.9 10.3

21 (42.0)

0 0 8 13

2 4 20 7 21 3 5 25

14 (28.0)

26 (52.0) 24 (48.0)

2.1 (0.8) 80.5 (18.7) 59.7 (8.3)

69.1 25 29 16.9

(9.5) (39.6) (83.3) (34.1)

(10.4) (8.3) (10.4) (56.3)

(29.2) (14.6) (6.3) (14.6) (52.1) (37.5) (64.6) (10.4) (18.8) (60.4)

(1.4) (1.0) (8.3) (19.9) (11.5) 44.2 (10.2) 48.1 (10.7)

1.4 2.2 14.1 31.5 19.0

41 (85.4)

5 4 5 27

14 7 3 7 25 18 31 5 9 29

12 (25.0) 36 (75.0)

1.7 (0.8) 64.3 (19.0) 53.9 (12.2)

66.5 19 40 38.9

(Continued )

,0.001* 0.541

,0.001* 0.016* ,0.001* ,0.001* ,0.001*

,0.001*

0.025* 0.054 0.415 0.002*

0.898 0.005* 0.674 0.302 0.23 0.008* 0.025* 0.482 0.216 0.3

0.006* 0.006*

0.019* ,0.001* 0.019*

0.181 0.3 0.006* ,0.001*

Subjects Reporting Event-based Exacerbation-like Respiratory Events (n = 98) Undiagnosed Diagnosed COPD COPD (n = 50 [51%]) (n = 48 [49%]) P Value

ORIGINAL ARTICLE

289

290 4.5 (3.4) 3.4 (2.8) 1.8 (2.1) 3.5 (2.5)

3.3 (2.8) 2.4 (2.0) 2.6 (2.4) 4.4 (3.0)

Total (n = 505) Undiagnosed Diagnosed COPD COPD (n = 355 [70%]) (n = 150 [30%])

3.6 (2.4)

2.0 (2.1)

,0.001* 0.003*

4.0 (3.1) 2.8 (2.0)

,0.001* ,0.001*

P Value

3.1 (2.0)

1.3 (1.5)

5.4 (3.8) 3.7 (2.8)

0.216

0.083

0.019* 0.092

Subjects Reporting Symptom-based Exacerbation-like Respiratory Events (n = 138) Undiagnosed Diagnosed COPD COPD (n = 78 [57%]) (n = 60 [43%]) P Value

3.6 (2.6)

1.9 (2.3)

3.5 (2.7) 2.7 (1.9)

3.0 (2.1)

1.4 (1.6)

5.2 (3.7) 3.7 (3.0)

0.532

0.471

0.017* 0.15

Subjects Reporting Event-based Exacerbation-like Respiratory Events (n = 98) Undiagnosed Diagnosed COPD COPD (n = 50 [51%]) (n = 48 [49%]) P Value

Definition of abbreviations: BOD = body mass index, obstruction, dyspnea; CAT = COPD assessment test; CHAMPS = Community Healthy Activities Model Program for Seniors; COPD = chronic obstructive pulmonary disease; CVD = cardiovascular disease; GOLD = Global Initiative for Chronic Obstructive Lung Disease; HADS = Hospital Anxiety and Depression Scale; ICS = inhaled corticosteroids; LABA = long-acting b-agonist; LAMA = long-acting muscarinic antagonist; MET = metabolic equivalent used to express exercise intensity; MMRC = Modified Medical Research Council; SABD = short-acting bronchodilator; SF = Short Form; SGRQ = St. George’s Respiratory Questionnaire. *Statistically significant; P values were obtained by performing Student’s t test (normal continuous variables) or Wilcoxon signed rank test (not normal continuous variables) or chi-square test (category variables). † Based on post-bronchodilator spirometry. GOLD1: post-FEV1/FVC , 0.70 and FEV1%pred > 80; GOLD21: post-FEV1/FVC , 0.70 and 80 . FEV1%pred . 0. Undiagnosed COPD based on no self-report of previous physician diagnosis of COPD before study entry, but with spirometric evidence of obstruction. Diagnosed COPD based on a self-reported previous physician diagnosis of COPD at study entry.

HADS Anxiety score Depression score CHAMPS caloric expenditure per week Moderate and greater intensity, >3 MET (kcal) All activities

Table 3. (Continued )

ORIGINAL ARTICLE


Total (n = 505)

Labonte, ´ Tan, Li, et al.: Health Care Burden of Undiagnosed COPD 0.63 60 (40.0)

48 (32.0)

0.30 78 (22.0)

50 (14.1)

26 (11.5)

,0.001* 38 (13.6)

0.24 42 (18.6)

0.28

,0.001* 60 (21.5)

,0.001*

12 (22.6)

18 (34.0)

0.47

0.033*

0.014*

0.005*

60 (26.5)

78 (34.5)

0.53

Definition of abbreviations: COPD = chronic obstructive pulmonary disease; GOLD = Global Initiative for Chronic Obstructive Lung Disease. *Statistically significant; P values were obtained using a chi-squared test.

Exacerbation rate in Year 0.39 1, no./patient-year Subjects with one or more 138 (27.3) symptoms-based exacerbation in Year 1, n (%) Subjects with one or 98 (19.4) more events-based exacerbation in Year 1, n (%)

Exacerbation-like respiratory events

GOLD1

GOLD21

24 (18.6)

36 (27.9)

0.40

36 (37.1)

42 (43.3)

0.71

0.002*

0.016*

0.002*

Undiagnosed Diagnosed Undiagnosed Diagnosed Undiagnosed Diagnosed COPD COPD Total COPD COPD Total COPD COPD (n = 355 [70%]) (n = 150 [30%]) P Value (n = 279) (n = 226 [81%]) (n = 53 [19%]) P Value (n = 226) (n = 129 [57%]) (n = 97 [43%]) P Value

Total

Table 4. Exacerbation Rate of Subjects Who Reported Exacerbation-like Respiratory Events during 1-Year Follow-up

ORIGINAL ARTICLE

291

ORIGINAL ARTICLE GOLD 2+

Exacerbation-like respiratory events (%)

40 40 30

34

* 20

37

* 28

*

22

43

Symptom-based

GOLD 1

Event-based

All COPD

19

10 0 40 30 20 10

32

* 14

*

23

* 19

12

0

Figure 1. Prevalence of exacerbation-like respiratory events by definition reported over 12 months of follow-up in CanCOLD subjects with undiagnosed (light gray bars; n = 355) and diagnosed (dark gray bars; n = 150) chronic obstructive pulmonary disease (COPD). Exacerbation-like events were classified as being symptom-based or event-based, based on established criteria (Table 1). Undiagnosed COPD based on no self-report of previous physician diagnosis of COPD before study entry, but with spirometric evidence of obstruction. Diagnosed COPD based on a self-reported previous physician diagnosis of COPD, chronic bronchitis, or emphysema at study entry. *Statistically significant (P , 0.05) compared with diagnosed subjects. CanCOLD = Canadian Cohort Obstructive Lung Disease; GOLD = Global Initiative for Chronic Obstructive Lung Disease.

reported at least one exacerbation-like respiratory event during 12 months of follow-up among subjects with undiagnosed and diagnosed COPD. Data

are presented for event-based and symptom-based exacerbation-like events. Overall at least one exacerbation-like event was reported by 60 of 150 (40%) subjects in

the diagnosed group and 78 of 355 (22%) subjects in the undiagnosed group (P , 0.001). The incidence rate of exacerbationlike events was significantly lower in subjects with undiagnosed COPD (Table 4). Subjects with diagnosed COPD reported 0.63 exacerbations per person-year compared with 0.23 exacerbations per person-year in those with undiagnosed COPD (incidence rate ratio, 0.37; P , 0.001). Incidence of reported exacerbationlike events increased with spirometry-based Global Initiative for Chronic Obstructive Lung Disease stratification of severity regardless of definition or diagnosis status (Table 4). Results of this analysis were similar whether or not subjects with a previous physician diagnosis of asthma (n = 162) were included in the analysis (see Table E3). Factors Associated with Reporting Exacerbation-like Respiratory Events in Subjects with Undiagnosed and Diagnosed COPD

Table 5 shows factors associated with increased odds of reporting exacerbationlike respiratory events during prospective follow-up, whether using a symptom-based

Table 5. Factors Associated with Reporting Exacerbation-like Respiratory Events (at Least One Exacerbation during the First Year of Follow-up) during Year 1

Factor

Undiagnosed COPD (n = 355) Symptom-based Event-based Adjusted OR Adjusted OR (95% CI) P Value (95% CI) P Value

Diagnosed COPD (n = 150) Symptom-based Event-based Adjusted OR Adjusted OR (95% CI) P Value (95% CI) P Value

1.05 (1.01–1.10) 0.049 FEV1, L per 100-ml decrease SGRQ total score, per 1.21 (1.10–1.32) ,0.001* 1.21 (1.09–1.34) ,0.001* increase of 4 points Angina, yes vs. no 5.52 (1.07–28.57) MMRC dyspnea 4.25 (1.48–12.19) scale, 3, 4, or 5 vs. 1 or 2 History of at least one exacerbation-like respiratory event in the past 12 mo at study entry, yes vs. no† Allergy, yes vs. no 1.89 (1.10–3.23) 0.021* Generalized anxiety 4.44 (1.07–18.42) 0.040* disorder, yes vs. no

1.11 (1.02–1.21)

0.019*

3.42 (1.26–9.30)

0.016*

0.042* 0.007*

Definition of abbreviations: CI = confidence interval; COPD = chronic obstructive pulmonary disease; MMRC = Modified Medical Research Council; OR = odds ratio; SGRQ = St. George’s Respiratory Questionnaire. Undiagnosed COPD based on no self-reported previous physician diagnosis of COPD before study entry, but with spirometric evidence of obstruction. Diagnosed COPD based on self-reported previous physician diagnosis of COPD at study entry. *Statistically significant; P values were obtained using a stepwise multivariable model adjusted for age, sex, body mass index, and smoking history. † Exacerbation-like respiratory events reported in 12 months before study entry based on subject recall.

292


ORIGINAL ARTICLE or event-based definition (see Table E1 for all factors considered). For both subgroups, reporting exacerbation-like respiratory events was associated with lower health status. Factors associated with reporting exacerbation-like respiratory events only in undiagnosed subjects with COPD included decreased lung function, and reporting allergy and anxiety comorbidities. Factors associated with reporting exacerbation-like respiratory events only in diagnosed subjects with COPD included increased dyspnea, a history of exacerbation-like events at study entry, and reflux-heartburn. This analysis was repeated excluding all subjects who reported receiving a previous physician diagnosis of asthma (n = 162); results remained the same (see Table E4). Differences in Medication and Health Service Use between Undiagnosed and Diagnosed Subjects with COPD Reporting Exacerbation-like Respiratory Events during Follow-up

Figure 2 shows that subjects with undiagnosed COPD had lower odds of reporting medication prescribed at study entry, and antibiotic and/or prednisone used to treat exacerbation-like events.

lower lung function than subjects not reporting exacerbation-like events. Although undiagnosed subjects with COPD report fewer exacerbation-like respiratory events and are less likely to be treated with antibiotics and prednisone when an event occurs, they use health services to treat these events in an equivalent manner as subjects who have received a previous diagnosis. As such, because undiagnosed subjects comprise more than two-thirds of the total COPD cohort in Canada, those subjects with undiagnosed COPD contribute to a great extent to the health care burden of COPD in Canada, despite not being recognized by health care professionals as having this disease, and thus not being properly managed and treated, nor being included in official COPD burden and cost estimates. CanCOLD is the first prospective, longitudinal, population-based cohort designed specifically to study COPD. A recent study by Gershon and colleagues (13) using three CanCOLD sites in Ontario linked to the health administrative databases showed that undiagnosed COPD did not seem to place significant burden on the health care system except for physician

However, the proportion of health services used to treat exacerbation-like events, including hospitalizations, emergency department visits, and unscheduled doctor visits, were similar to that of diagnosed subjects (Figure 3, Table 6). This analysis was repeated excluding all subjects who reported receiving a previous physician diagnosis of asthma (n = 162); the results remained similar (see Figure E2).

Discussion In this population-based study of subjects with undiagnosed and diagnosed COPD, we showed that subjects who reported not having received a physician-diagnosis of COPD, despite having spirometric evidence of airflow obstruction, have milder airflow obstruction, are less symptomatic, have better patient-reported outcomes, and are less likely to be treated with respiratory medication than subjects with diagnosed COPD. However, like diagnosed subjects with COPD, undiagnosed subjects who report exacerbation-like respiratory events have lower health status, in addition to also having more allergy, anxiety, and depression, and

Adjusted Odds Ratio (95% CI) Symptom-based ICS combined with LABA / LAMA at study entry

0.37 (0.16–0.84)

Any respiratory medications at study entry

0.15 (0.06–0.36)

Antibiotics and /or prednisone used

0.28 (0.13–0.60)

Hospitalization

0.79 (0.17–3.76)

ED visit

0.45 (0.11–1.83) 0.68 (0.32–1.43)

Doctor visit Event-based ICS combined with LABA / LAMA at study entry

0.22 (0.08–0.61)

Any respiratory medications at study entry

0.10 (0.03–0.31)

Antibiotics and /or prednisone used

0.27 (0.11–0.68)

Hospitalization

1.10 (0.22–5.68)

ED visit

0.50 (0.12–2.09) 0.92 (0.37–2.33)

Doctor visit 0.01

0.1

1

10

Figure 2. Comparison of the log of odds (695% CI) of medication or health service use in subjects with undiagnosed chronic obstructive pulmonary disease (COPD) (n = 355) versus subjects with diagnosed COPD (n = 150, the reference group) who report exacerbation-like respiratory events over at least 12 months in the CanCOLD cohort. Undiagnosed COPD based on no self-reported previous physician diagnosis of COPD, chronic bronchitis, or emphysema before study entry, but with spirometric evidence of obstruction. Reported COPD based on a self-reported previous physician diagnosis of COPD at study entry. Odds ratios adjusted for age, sex, body mass index, and smoking history. CanCOLD = Canadian Cohort Obstructive Lung Disease; CI = confidence interval; ED = emergency department; ICS = inhaled corticosteroids; LABA = long-acting b-agonist; LAMA = long-acting muscarinic antagonist.


293

ORIGINAL ARTICLE Hospitalization

ED visit

Doctor visit

60

Symptom-based

52% 41%

20

15% 5%

7%

5%

0 64%

65%

60 40 19%

20 8%

8%

Event-based

Health Service Utilization (%)

40

8%

0

Figure 3. Comparison of the proportion of health services used between undiagnosed (light gray bars; n = 355) and diagnosed subjects (dark gray bars; n = 150) who experienced exacerbation-like respiratory events, including both symptom- and event-based definitions, over 12 months of followup in CanCOLD. Exacerbation-like events were classified as being symptom-based or event-based, based on established criteria (Table 1). Undiagnosed chronic obstructive pulmonary disease (COPD) based on no self-report of previous physician diagnosis of COPD before study entry, but with spirometric evidence of obstruction. Diagnosed COPD based on a self-reported previous physician diagnosis of COPD, chronic bronchitis, or emphysema at study entry. CanCOLD = Canadian Cohort Obstructive Lung Disease; ED = emergency department.

visits. Our study, which is a prospective cohort from the nine sites in Canada, offers a unique opportunity to determine the incidence of reported exacerbations in subjects with COPD who remain undiagnosed in the population, to assess the consequences of COPD exacerbations in the population at large, and to address the gap in the clinical practice of undiagnosed COPD. Exacerbations are one of the most important factors in COPD determining functional impairment and health status (2, 3). Furthermore, it is the number one cause of hospital admission in Canada (23). Predictably, the most likely cause for hospital admission is delayed recognition and management of acute exacerbations of COPD (24). In our analysis, we thought it was important to look at event-based exacerbations, the most common way of doing things, but also at symptom-based exacerbations because they too are clinically relevant as shown by their association with poor quality of life in patients with COPD (20). We found the incidence of exacerbation-like respiratory events to be lower among subjects with undiagnosed COPD than among those with a physician diagnosis. COPD exacerbations may be a trigger for patients to come to the attention of the health care system and for a physician to consider the diagnosis of 294

COPD. Exacerbation incidence is well known to increase with disease severity (1, 2, 8, 9). In our analysis, at study entry, undiagnosed subjects included more men, and subjects who were less symptomatic and less impaired in terms of lung function, comorbidities, health status, and other psychosocial attributes. This could explain lower incidence of patient-reported events. Other studies, such as the recent analyses of National Health and Nutrition Examination Survey data (25), have found that subjects with undiagnosed COPD have fewer symptoms, better health status, milder airflow obstruction, and fewer comorbidities than diagnosed patients (26–29). In a pooled analysis from several international population-based cohorts (30), it was recently found that undiagnosed COPD was associated with male sex, younger age, smoking status, education level, and milder airflow obstruction. Most individuals seeking health care that could lead to an initial COPD diagnosis tend to do so because they are experiencing debilitating dyspnea (1), a major risk factor for exacerbations (31). This may explain why milder less symptomatic subjects are more likely to remain undiagnosed. Poorer self-reported health status was associated with increased odds of reporting exacerbation-like events for both subject

subgroups. Frequent exacerbators (experiencing treated and untreated events) have been shown in general to have worse health status (3, 32, 33). Even individuals with unreported exacerbations tend to have lower health status (34). Thus our findings further support a relationship between exacerbations and worsened health status. Our findings also showed that reporting exacerbation-like events was associated with lower lung function, allergy, anxiety, and depression in undiagnosed subjects with COPD; and increased dyspnea, a history of exacerbations, and reflux/heartburn in diagnosed subjects with COPD. All of these factors except allergy, anxiety, and depression were among those identified in the ECLIPSE study (10) as being associated with increased risk of exacerbations. This difference with regards to undiagnosed subjects may relate to their milder and less symptomatic disease state. It is likely that anxiety and depression result from experiencing exacerbation-like events and unmanaged symptoms. In diagnosed subjects, having reflux/heartburn may be related to a more advanced, and comorbidassociated disease status. Overall, our data lend further support that there are intrinsic risk factors associated with exacerbation susceptibility among subjects with more severe airflow obstruction. Yet it remains to be determined how these change over time in milder subjects. Although it is not surprising that subjects with undiagnosed COPD report fewer respiratory medication prescribed at study entry and were less likely to be actively treated during exacerbation-like events, it is quite surprising that the odds of using health services during exacerbationlike events was similar to those of subjects with diagnosed disease. Intriguingly this suggests that these subjects with COPD could be admitted to hospital and visit the emergency department or their doctors for treatment of exacerbation-like events regardless of diagnosis status. Moreover, even individuals with “milder” COPD, who are less symptomatic and less functionally impaired, contribute significantly to the burden of exacerbations. This highlights an important consequence of COPD, which was also shown in ECLIPSE, that exacerbations occur across all COPD severity groups. It also suggests that there is a challenge for the management of exacerbation-like events and COPD by health care providers, resulting in an


Labonte, ´ Tan, Li, et al.: Health Care Burden of Undiagnosed COPD 4 (5.1) 4 (5.1) 32 (41.0) 5 (6.4) 4 (5.1) 3 (3.8) 21 (26.9)

8 (5.8) 13 (9.4) 63 (45.7) 19 (13.8) 8 (5.8) 12 (8.7) 56 (40.6)

35 (58.3)

14 (23.3) 4 (6.7) 9 (15.0)

4 (6.7) 9 (15.0) 31 (51.7) 19 (19.4) 8 (8.2) 12 (12.2) 56 (57.1)

,0.001*

8 (8.2) 13 (13.3) 63 (64.3)

21 (42.0)

5 (10.0) 4 (8.0) 3 (6.0)

4 (8.0) 4 (8.0) 32 (64.0)

35 (72.9)

14 (29.2) 4 (8.3) 9 (18.8)

4 (8.3) 9 (18.8) 31 (64.6)

0.002*

0.016* 1.00 0.054

1.00 0.12 0.95

Event-based Exacerbation-like Respiratory Events (n = 98) Undiagnosed Diagnosed Total COPD COPD (n = 98) (n = 50 [51%]) (n = 48 [49%]) P Value

0.004* 0.73 0.021*

0.73 0.050 0.21

Definition of abbreviations: COPD = chronic obstructive pulmonary disease; ED = emergency department. *Statistical significance. P values were obtained using a chi-square test.

Health services Hospitalization ED visit Doctor visit Increased respiratory medication use Bronchodilators Inhaled corticosteroids Both bronchodilators and inhaled corticosteroids Antibiotics and/or prednisone used

Symptom-based Exacerbation-like Respiratory Events (n = 138) Undiagnosed Diagnosed Total COPD COPD (n = 138) (n = 78 [57%]) (n = 60 [43%]) P Value

Table 6. Impacts of Exacerbation-like Respiratory Events by Diagnosed Status on Health Service and Medicine Use

ORIGINAL ARTICLE

295

ORIGINAL ARTICLE increased health care burden with potentially serious economic consequences. Current concerns about the rising health care burden of COPD reflect only those patients with physician-diagnosed disease; our data suggest that the true impact of COPD may be largely underestimated. Globally, approximately 66% of subjects with COPD remain undiagnosed (16), and COPD prevalence is expected to continue to rise (1). In Canada, the COPD prevalence study CanCOLD showed that COPD prevalence was approximately four times greater than previously thought (16.4%) (35). Correct COPD diagnosis would ensure prompt delivery of care with appropriate medication, patient education, and disease management plans to anticipate and handle exacerbations, significantly improving health status and decreasing hospitalizations and doctor visits (36). Our study includes several strengths and limitations. The former include that CanCOLD is a population-based study reflecting events occurring in the general population that impact the health and well-being of people, and the health care system. A unique and novel aspect of the CanCOLD study is that it provides information on subjects from all Global Initiative for Chronic Obstructive Lung Disease classifications, particularly those with mild disease, and those potentially at risk of disease development. These groups are often underrepresented in other cohorts, but represent important populations, because they are most likely to benefit from early intervention. Our analysis also included men and women, represented in approximately equal proportion, a challenge in many cohorts. Another strength is the evaluation of symptom- and event-based exacerbationlike events. This allowed for the inclusion of milder events that may not warrant medication or health service use, but may still affect quality of life (34), and more severe events requiring treatment. Finally a follow-up of at least 12 months provides stronger and robust data. Limitations include the presence of subjects with asthma in our COPD-based cohort. There is still no definitive way of distinguishing between COPD and asthma when post-bronchodilator fixed airflow obstruction is demonstrated in subjects. Even in the absence of smoking, it remains 296

difficult to distinguish between these conditions, because there is evidence that COPD in never-smokers is more prevalent than previously thought (37). It is also possible that certain individuals were misclassified as having undiagnosed COPD at CanCOLD study entry, despite having previously received a physician diagnosis. Similarly, subjects who previously received a physician diagnosis of asthma may have been incorrectly identified as such, and met spirometric criteria for COPD at CanCOLD study entry. To examine the impact of these subjects, our analyses were repeated after removing all subjects who reported receiving a previous physician diagnosis of asthma (see online supplement), and overall, the results remained the same. Consequently, the inclusion of these subjects in our study likely did not significantly impact the results. Other acute events that may present with exacerbationlike symptoms or manifestation (e.g., coronary events) may have been included in our analysis. Although the administration of the exacerbation questionnaire every 3 months was undertaken to ensure improved subjectrecall, there remains potential for inaccuracy. Finally, we note that COPD is also overdiagnosed in the community, particularly in primary practice where spirometry is underused (38). Our study did not include patients with falsely diagnosed disease, a group whose exacerbation event potential remains unknown. Our study highlights the substantial impact on health service use resulting from exacerbation-like respiratory events experienced by subjects with undiagnosed COPD. This is especially important when considering that these are treated as isolated acute events without awareness of the need for future management of underlying COPD. If these patients were recognized as having COPD, providing proper preventive therapy would have the potential to reduce or prevent complications, such as emergency department visits and hospital admissions. Continued follow-up and analysis of this cohort may provide the information required to identify susceptibility phenotypes in at-risk subjects or those in the early stages of disease. n Author disclosures are available with the text of this article at www.atsjournals.org.

Acknowledgement: The authors thank the men and women who participated in the study and individuals in the CanCOLD Collaborative Research Group. Members of the CanCOLD Collaborative Research Group are as follows: Executive Committee: Jean Bourbeau (McGill University, Montreal, Canada); Wan C. Tan, J. Mark FitzGerald; Don Sin (UBC, Vancouver, Canada); Darcy Marciniuk (University of Saskatoon, Saskatoon, Canada) Dennis E. O’Donnell (Queen’s University, Kingston, Canada); Paul Hernandez (Dalhousie University, Halifax, Canada); Kenneth R. Chapman (University of Toronto, Toronto, Canada); Robert Cowie (University of Calgary, Calgary, Canada); Shawn Aaron (University of Ottawa, Ottawa, Canada); F. Maltais (University of Laval, Quebec City, Canada); International Advisory Board: Jonathon Samet (Keck School of Medicine of USC, Los Angeles, CA); Milo Puhan (John Hopkins School of Public Health, Baltimore, MD); Qutayba Hamid (McGill University, Montreal, Canada); James C. Hogg (UBC James Hogg Research Center, Vancouver, Canada). Operations Center: Jean Bourbeau (PI), Carole Baglole, Carole Jabet, Palmina Mancino, Yvan Fortier (University of McGill, Montreal, Canada); Wan C. Tan (co-PI), Don Sin, Sheena Tam, Jeremy Road, Joe Comeau, Adrian Png, Harvey Coxson, Miranda Kirby, Jonathon Leipsic, Cameron Hague (University of British Columbia James Hogg Research Center, Vancouver, Canada). Economic Core: Mohsen Sadatsafavi (University of British Columbia, Vancouver, Canada). Public Health Core: Teresa To, Andrea Gershon (University of Toronto, Toronto, Canada). Data Management and Quality Control: Wan C. Tan, Harvey Coxson (UBC, Vancouver, Canada); Jean Bourbeau, Pei-Zhi Li, Jean-Francois Duquette, Yvan Fortier, Andrea Benedetti, Denis Jensen (McGill University, Montreal, Canada); Denis O’Donnell (Queen’s University, Kingston, Canada). Field Centers: Wan C. Tan (PI), Christine Lo, Sarah Cheng, Cindy Fung, Nancy Haynes, Junior Chuang, Licong Li, Selva Bayat, Amanda Wong, Zoe Alavi, Catherine Peng, Bin Zhao, Nathalie Scott-Hsiung, Tasha Nadirshaw (UBC James Hogg Research Center, Vancouver, Canada); Jean Bourbeau (PI), Palmina Mancino, David Latreille, Jacinthe Baril, Laura Labonte (McGill University, Montreal, Canada); Kenneth Chapman (PI), Patricia McClean, Nadeen Audisho (University of Toronto, Toronto, Canada); Robert Cowie (PI), Ann Cowie, Curtis Dumonceaux, Lisette Machado (University of Calgary, Calgary, Canada); Paul Hernandez (PI), Scott Fulton, Kristen Osterling (Dalhousie University, Halifax, Canada); Shawn Aaron (PI), Kathy Vandemheen, Gay Pratt, Amanda Bergeron (University of Ottawa, Ottawa, Canada); Denis O’Donnell (PI), Matthew McNeil, Kate Whelan (Queen’s University, Kingston, Canada); Francois Maltais (PI), Cynthia Brouillard (University of Laval, Quebec City, Canada); Darcy Marciniuk (PI), Ron Clemens, Janet Baran (University of Saskatoon, Saskatoon, Canada).


ORIGINAL ARTICLE References 1. Vestbo J, Hurd SS, Agust´ı AG, Jones PW, Vogelmeier C, Anzueto A, Barnes PJ, Fabbri LM, Martinez FJ, Nishimura M, et al. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med 2013;187:347–365. 2. Donaldson GC, Seemungal TA, Bhowmik A, Wedzicha JA. Relationship between exacerbation frequency and lung function decline in chronic obstructive pulmonary disease. Thorax 2002;57: 847–852. 3. Seemungal TA, Donaldson GC, Paul EA, Bestall JC, Jeffries DJ, Wedzicha JA. Effect of exacerbation on quality of life in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1998;157:1418–1422. 4. Spencer S, Calverley PM, Burge PS, Jones PW. Impact of preventing exacerbations on deterioration of health status in COPD. Eur Respir J 2004;23:698–702. 5. Sullivan SD, Ramsey SD, Lee TA. The economic burden of COPD. Chest 2000;117(Suppl 2):5S–9S. 6. Soler-Cataluña JJ, Mart´ınez-Garc´ıa MA, Roman ´ Sanchez ´ P, Salcedo E, Navarro M, Ochando R. Severe acute exacerbations and mortality in patients with chronic obstructive pulmonary disease. Thorax 2005;60: 925–931. 7. Wouters EF. The burden of COPD in The Netherlands: results from the Confronting COPD survey. Respir Med 2003;97: S51–S59. 8. Burge PS, Calverley PM, Jones PW, Spencer S, Anderson JA, Maslen TK. Randomised, double blind, placebo controlled study of fluticasone propionate in patients with moderate to severe chronic obstructive pulmonary disease: the ISOLDE trial. BMJ 2000;320: 1297–1303. 9. Miravitlles M, Guerrero T, Mayordomo C, Sanchez-Agudo ´ L, Nicolau F, Segu´ JL; The EOLO Study Group. Factors associated with increased risk of exacerbation and hospital admission in a cohort of ambulatory COPD patients: a multiple logistic regression analysis. Respiration 2000;67:495–501. 10. Hurst JR, Vestbo J, Anzueto A, Locantore N, Mullerova ¨ H, Tal-Singer R, Miller B, Lomas DA, Agusti A, Macnee W, et al.; Evaluation of COPD Longitudinally to Identify Predictive Surrogate Endpoints (ECLIPSE) Investigators. Susceptibility to exacerbation in chronic obstructive pulmonary disease. N Engl J Med 2010;363: 1128–1138. 11. Hill K, Goldstein RS, Guyatt GH, Blouin M, Tan WC, Davis LL, HeelsAnsdell DM, Erak M, Bragaglia PJ, Tamari IE, et al. Prevalence and underdiagnosis of chronic obstructive pulmonary disease among patients at risk in primary care. CMAJ 2010;182:673–678. 12. Boulet LP, Bourbeau J, Skomro R, Gupta S. Major care gaps in asthma, sleep and chronic obstructive pulmonary disease: a road map for knowledge translation. Can Respir J 2013;20: 265–269. 13. Gershon AS, Hwee J, Chapman KR, Aaron SD, O’Donnell D, Stanbrook MB, Bourbeau J, Tan W, Su J, Victor JC, To T. Factors associated with undiagnosed and overdiagnosed COPD. Eur Respir J (In Press) 14. Labonte L, Mancino P, Tan W, Cowie R, Hernandez P, Chapman KR, Li P, Aaron S, Benedetti A, Fitzgerald M, et al. Measuring exacerbations in subjects with mild to moderate COPD from a population-based cohort: the CanCOLD study. Eur Respir J 2013; 42:4226. 15. Labonte L, Mancino P, Tan W, Cowie R, Hernandez P, Chapman K, Li P, Aaron S, Benedetti A, Fitzgerald M, et al. Characteristics of subjects susceptible to exacerbation-like respiratory events in a population-based cohort: Canadian Cohort Obstructive Lung Disease (CanCOLD) study. Eur Respir J 2014;44:2986. 16. Buist AS, McBurnie MA, Vollmer WM, Gillespie S, Burney P, Mannino DM, Menezes AMB, Sullivan SD, Lee TA, Weiss KB, et al.; BOLD Collaborative Research Group. International variation in the prevalence of COPD (the BOLD study): a population-based prevalence study. Lancet 2007;370: 741–750.

17. Bourbeau J, Tan WC, Benedetti A, Aaron SD, Chapman KR, Coxson HO, Cowie R, Fitzgerald M, Goldstein R, Hernandez P, et al.; CanCOLD Study Group. Canadian Cohort Obstructive Lung Disease (CanCOLD): Fulfilling the need for longitudinal observational studies in COPD. COPD 2014;11:125–132. 18. Bourbeau J, Ford G, Zackon H, Pinsky N, Lee J, Ruberto G. Impact on patients’ health status following early identification of a COPD exacerbation. Eur Respir J 2007;30:907–913. 19. Bischoff EW, Hamd DH, Sedeno M, Benedetti A, Schermer TR, Bernard S, Maltais F, Bourbeau J. Effects of written action plan adherence on COPD exacerbation recovery. Thorax 2011;66:26–31. 20. Xu W, Collet JP, Shapiro S, Lin Y, Yang T, Wang C, Bourbeau J. Negative impacts of unreported COPD exacerbations on health-related quality of life at 1 year. Eur Respir J 2010;35: 1022–1030. 21. Vestbo J, Anderson W, Coxson HO, Crim C, Dawber F, Edwards L, Hagan G, Knobil K, Lomas DA, MacNee W, et al.; ECLIPSE investigators. Evaluation of COPD longitudinally to identify predictive surrogate end-points (ECLIPSE). Eur Respir J 2008;31: 869–873. 22. American Thoracic Society/European Respiratory Society Task Force. Standards for the diagnosis and management of patients with COPD. Version 1.2. New York: American Thoracic Society; 2004 [updated 2005 Sept 8; accessed 2016 Jan 29]. Available from: http://www.thoracic.org/go/copd 23. Health indicators 2008 [accessed 2016 Jun 6]. Available from: https:// secure.cihi.ca/free_products/HealthIndicators2008_ENGweb.pdf 24. Wedzicha JA, Brill SE, Allinson JP, Donaldson GC. Mechanisms and impact of the frequent exacerbator phenotype in chronic obstructive pulmonary disease. BMC Med 2013;11:181. 25. Martinez CH, Mannino DM, Jaimes FA, Curtis JL, Han MK, Hansel NN, Diaz AA. Undiagnosed obstructive lung disease in the U.S.: associated factors and long-term mortality. Ann Am Thorac Soc 2015;12:1788–1795. 26. Balcells E, Gimeno-Santos E, de Batlle J, Ramon MA, Rodr´ıguez E, Benet M, Farrero E, Ferrer A, Guerra S, Ferrer J, et al.; PAC-COPD Study Group. Characterisation and prognosis of undiagnosed chronic obstructive pulmonary disease patients at their first hospitalisation. BMC Pulm Med 2015;15:4. 27. Lindberg A, Bjerg A, Ronmark ¨ E, Larsson LG, Lundback ¨ B. Prevalence and underdiagnosis of COPD by disease severity and the attributable fraction of smoking. Report from the Obstructive Lung Disease in Northern Sweden Studies. Respir Med 2006;100: 264–272. 28. Miravitlles M, Soriano JB, Garc´ıa-R´ıo F, Muñoz L, Duran-Tauleria E, Sanchez G, Sobradillo V, Ancochea J. Prevalence of COPD in Spain: impact of undiagnosed COPD on quality of life and daily life activities. Thorax 2009;64:863–868. 29. Jordan RE, Lam KB, Cheng KK, Miller MR, Marsh JL, Ayres JG, Fitzmaurice D, Adab P. Case finding for chronic obstructive pulmonary disease: a model for optimising a targeted approach. Thorax 2010;65:492–498. 30. Lamprecht B, Soriano JB, Studnicka M, Kaiser B, Vanfleteren LE, Gnatiuc L, Burney P, Miravitlles M, Garc´ıa-Rio F, Akbari K, et al.; BOLD Collaborative Research Group, the EPI-SCAN Team, the PLATINO Team, and the PREPOCOL Study Group; BOLD Collaborative Research Group the EPI-SCAN Team the PLATINO Team and the PREPOCOL Study Group. Determinants of underdiagnosis of COPD in national and international surveys. Chest 2015;148:971–985. 31. Ramsey SD, Hobbs FD. Chronic obstructive pulmonary disease, risk factors, and outcome trials: comparisons with cardiovascular disease. Proc Am Thorac Soc 2006;3:635–640. 32. Miravitlles M, Ferrer M, Pont A, Zalacain R, Alvarez-Sala JL, Masa F, Verea H, Murio C, Ros F, Vidal R; IMPAC Study Group. Effect of exacerbations on quality of life in patients with chronic obstructive pulmonary disease: a 2 year follow up study. Thorax 2004;59: 387–395. 33. Mackay AJ, Donaldson GC, Patel AR, Jones PW, Hurst JR, Wedzicha JA. Usefulness of the chronic obstructive pulmonary disease assessment test to evaluate severity of COPD exacerbations. Am J Respir Crit Care Med 2012;185:1218–1224.


297

ORIGINAL ARTICLE 34. Langsetmo L, Platt RW, Ernst P, Bourbeau J. Underreporting exacerbation of chronic obstructive pulmonary disease in a longitudinal cohort. Am J Respir Crit Care Med 2008;177: 396–401. 35. Tan WC, Bourbeau J, FitzGerald JM, Cowie R, Chapman K, Hernandez P, Buist SA, Sin DD. Can age and sex explain the variation in COPD rates across large urban cities? A population study in Canada. Int J Tuberc Lung Dis 2011;15:1691–1698. 36. O’Donnell DE, Hernandez P, Kaplan A, Aaron S, Bourbeau J, Marciniuk D, Balter M, Ford G, Gervais A, Lacasse Y, et al. Canadian Thoracic Society recommendations for management of chronic obstructive

298

pulmonary disease: 2008 update, highlights for primary care. Can Respir J 2008;15:1A–8A. 37. Lamprecht B, McBurnie MA, Vollmer WM, Gudmundsson G, Welte T, Nizankowska-Mogilnicka E, Studnicka M, Bateman E, Anto JM, Burney P, et al.; BOLD Collaborative Research Group. COPD in never smokers: results from the population-based Burden of Obstructive Lung Disease study. Chest 2011;139: 752–763. 38. Walters JA, Walters EH, Nelson M, Robinson A, Scott J, Turner P, Wood-Baker R. Factors associated with misdiagnosis of COPD in primary care. Prim Care Respir J 2011;20:396–402.
