Chronic obstructive pulmonary disease (COPD) is a progressive disease with self-reported diagnoses in 4% of Canadians aged 35–79 years [
The impact of COPD-related exacerbations on burden of illness has been assessed in several studies [
Existing studies on COPD-related HCU are limited. Given the chronic nature of COPD, patients can be treated by several physicians and be covered by more than one insurance plan. This can lead to loss of information regarding HCU. These issues limit the generalization of the results to the general COPD population. In addition, in observational prospective studies, the total cost of COPD and related exacerbations may be underestimated because of limited follow-up time and recall bias. Larger population-based studies with a long-term follow-up and comprehensive ascertainment of HCU are required to provide a complete assessment of the burden of illness related to COPD exacerbations. This longitudinal study aimed to investigate the impact of COPD exacerbations on disease-related HCU in Québec, Canada.
The primary objective was to describe the burden of illness related to COPD exacerbations in Québec, Canada. Burden of illness was determined based on the incidence of COPD-related exacerbations and HCU related to the management of COPD exacerbations. Secondary objectives were to describe the profile of COPD exacerbations in terms of severity, frequency, characteristics (e.g., medication use, hospitalizations, and emergency room [ER] visits), and time to first exacerbation and to estimate the risk of a COPD exacerbation based on patient characteristics, disease parameters, and treatments.
This was a retrospective observational cohort study of patients with COPD using deidentified patient data derived from provincial health insurance databases (Régie de l’Assurance Maladie du Québec [RAMQ]). Eligible patients classified with COPD were selected from a random sample of 250,000 with a health administrative claim linked with an International Classification of Diseases, Ninth Revision (ICD-9) diagnostic billing for either asthma or COPD. Patients were followed from the date of COPD classification (index date), until March 31, 2011, death, or withdrawal from the health insurance plan. A 2-year look-back or run-in period prior to the index date was included to assess eligibility and detect comorbidities (Figure
Data extraction time frame.
A moderate exacerbation was defined as a physician visit with a diagnosis code for COPD and an oral corticosteroid (OCS) or an antibiotic prescription for a respiratory infection (filled within 2 weeks of the physician visit), whereas a severe exacerbation was defined as an ER visit with a primary diagnosis code for COPD or hospitalization with a primary discharge diagnosis code for COPD. To ascertain the study outcomes, all RAMQ medical and pharmaceutical claims and Med-Écho hospitalization records for the patients included in the study were extracted from the index date to March 31, 2011. Survival status was determined by record linkage by the Institut de la Statistique de Québec Death Registry.
The study was conducted in accordance with the Declaration of Helsinki. Access to the data, for the purpose of this study, was granted by the Commission d’Accès à l’Information du Québec, the data custodian. At the time of the data acquisition, external research ethics board approval was not required.
Eligible patients ≥ 40 years of age at treatment initiation had a new HCU claim with a diagnostic billing for COPD between January 01, 2001, and December 31, 2010, with a respiratory medicine prescription claim within 2 weeks of the diagnostic billing claim for COPD. To assess eligibility and detect comorbidities, patients were also required to have available data in the RAMQ database for the 2-year look-back period preceding the index date. Eligible HCU claims for disease classification included physician visit or ER visits or hospitalization. The COPD diagnostic billing included either chronic bronchitis or emphysema (ICD-9 codes: 490.xx, 491.xx, 492.xx, or 496.xx). Respiratory medicines considered for eligibility included are shown in Box
Formoterol fumarate Salmeterol Salbutamol Fenoterol Terbutaline Fenoterol/ipratropium Salbutamol/ipratropium Ipratropium bromide Tiotropium bromide Salmeterol/fluticasone propionate Budesonide Theophylline Oxtriphylline Fluticasone propionate Beclomethasone
The primary study outcome was the number of exacerbation-related HCU events per year of follow-up. Annual COPD exacerbation rates during the follow-up period and time to first COPD exacerbation since diagnosis of COPD were assessed as secondary study outcomes. Study outcomes were assessed according to whether they experienced no or at least one exacerbation (any moderate or severe event), a moderate exacerbation (but no severe exacerbations) only, or severe exacerbation (patients in the “severe” category may have also had a moderate exacerbation) during the follow-up period. Diary assessment data were not available for the assessment of mild exacerbation events. Utilization of the following health care resources was assessed: prescriptions for rescue medications (e.g., short-acting
Descriptive statistics were produced for all relevant study variables including patient characteristics, disease parameters, and treatments. Unadjusted analyses and multivariable Poisson regression models were used to compare patient groups according to exacerbation classification with respect to the rate of HCU. In unadjusted analyses, between-group differences in HCU were analyzed using incidence density rate ratios (IDRRs). For the multivariable models, the covariables included age, gender, baseline CCI score, baseline comorbidity profile, follow-up duration, and HCU in the look-back period; follow-up duration was used as an offset variable. The time to first exacerbation was described using the Kaplan-Meier estimate of the survival function as measured from the time of diagnosis of COPD.
A total of 53,349 patients with COPD met the study eligibility criteria and were included, of whom 25,259 (47.3%) experienced no (“no”) COPD exacerbations and 28,090 (52.7%) experienced at least one COPD exacerbation during the follow-up period. The majority of patients were female (57.6%) and the mean (SD) age was 68.1 (12.8) years at the index date. The total study population was followed for a mean (standard deviation [SD]) of 4.43 (2.70) years (Table
Patient demographics and baseline characteristics.
Exacerbation classification | Total study population |
||||
---|---|---|---|---|---|
“No” |
Any (moderate |
Moderate |
Severe |
||
|
|||||
Sum | 107,206 | 129,529 | 79,218 | 50,310 | 236,735 |
Mean (SD) | 4.24 (2.66) | 4.61 (2.70) | 4.73 (2.70) | 4.66 (2.73) | 4.43 (2.70) |
|
|||||
Mean (SD) | 65.85 (13.12) | 70.10 (12.21) | 66.96 (12.48) | 75.13 (9.87) | 68.09 (12.83) |
|
|||||
|
60.0 | 55.5 | 61.8 | 45.5 | 57.6 |
|
|||||
Respiratory disease |
14.5 | 26.9 | 34.2 | 45.6 | 21.0 |
Hypertension | 6.2 | 11.2 | 13.5 | 20.0 | 8.9 |
Depression | 4.2 | 7.8 | 9.4 | 13.7 | 6.1 |
Diabetes | 6.7 | 11.9 | 12.7 | 22.2 | 9.4 |
|
|||||
mean (SD) | 1.00 (1.67) | 1.40 (1.87) | 0.92 (1.58) | 2.15 (2.03) | 1.21 (1.79) |
|
|||||
|
|||||
Mean (SD) | 0.77 (2.20) | 0.83 (2.40) | 0.80 (2.17) | 0.88 (2.71) | 0.80 (2.31) |
|
|||||
Mean (SD) | 0.28 (1.94) | 0.62 (4.49) | 0.49 (4.15) | 0.83 (4.98) | 0.46 (3.52) |
|
|||||
Mean (SD) | 0.41 (2.29) | 0.86 (3.74) | 0.41 (2.63) | 1.58 (4.95) | 0.64 (3.15) |
|
|||||
Mean (SD) | 0.07 (1.25) | 0.16 (1.74) | 0.08 (1.30) | 0.30 (2.28) | 0.12 (1.53) |
|
|||||
|
|||||
Mean (SD) | 0.40 (0.94) | 0.58 (1.64) | 0.25 (0.90) | 1.10 (2.29) | 0.50 (1.35) |
|
|||||
Mean (SD) | 0.06 (0.43) | 0.11 (0.68) | 0.03 (0.34) | 0.23 (0.99) | 0.08 (0.57) |
|
|||||
Mean (SD) | 0.10 (0.33) | 0.10 (0.38) | 0.05 (0.25) | 0.19 (0.51) | 0.10 (0.36) |
|
|||||
Mean (SD) | 0.06 (0.25) | 0.18 (0.42) | 0.04 (0.21) | 0.39 (0.55) | 0.12 (0.35) |
The exacerbation rate during the follow-up period was 34.3 events per 100 patient-years in patients experiencing an exacerbation; 22.7 events per 100 patient-years in patients experiencing only moderate exacerbation; and 11.6 events per 100 patient-years in patients experiencing at least one severe exacerbation. Of the patients who experienced an exacerbation, 17,288 (61.5%) experienced moderate events, while 10,802 (38.5%) experienced at least one severe exacerbation during the follow-up period. Compared with patients who experienced “no” or only moderate exacerbations, patients who experienced severe exacerbations were older (mean [SD] age of 75.1 [9.9] years), were mostly male (54.5%), and had a higher use of medications and more medical visits during the follow-up period (Table
Median time to a first COPD exacerbation was 37 months from the index date (Figure
Kaplan-Meier plot showing time to first COPD exacerbation for (a) an exacerbation, (b) moderate exacerbation, and (c) severe exacerbation. COPD, chronic obstructive pulmonary disease.
Patients who experienced a COPD exacerbation had significantly higher HCU rates across all parameters compared with patients who experienced “no” exacerbations during the follow-up period (Figure
Relative risk of health care resource utilization for (a) COPD therapy and (b) health care visits in patients experiencing ≥1 exacerbation (
Relative risk of health care resource utilization for (a) COPD therapy and (b) health care visits in patients experiencing moderate exacerbations (
Relative risk of health care resource utilization for (a) COPD therapy and (b) health care visits in patients experiencing ≥1 severe exacerbation (
Adjustment for demographics, comorbidities, and prior HCU did not directionally alter the results. As with the unadjusted analyses, in patients who experienced an exacerbation or moderate exacerbation, HCU was higher during the 3 months after an exacerbation compared with preexacerbation HCU (Table S1 in Supplementary Material available online at
In this retrospective observational cohort study of patients with COPD using the Québec, Canada, provincial health insurance databases, patients experiencing exacerbations had a significantly higher HCU rate compared with those patients who did not experience exacerbations. The effect was more pronounced in patients who experienced severe exacerbations. These results are consistent with those reported previously [
However, direct comparisons between studies are challenging due to differences in methodology, quality of evidence, and regional standards of care. Recently, in the Salford Lung Study, a randomized clinical trial designed to reflect the reality of clinical practice, comparative groups were based on allocated treatment rather than exacerbation severity as in the current study, which may have resulted in treatment groups with mixed exacerbation frequencies. The Salford Lung Study also had a shorter duration of patient follow-up (1 year) compared with the current study (average 4.44 years), providing a shorter period of time for HCU assessment and also enrolled patients prospectively, potentially biasing the patient inclusion away from the general COPD population [
Limitations of this study include those that are typical of administrative database studies. The most important one of these limitations is the validity of the information included in the database, specifically relating to the diagnosis and treatment of COPD, which were based on ICD codes and claims submissions. Indeed, a previous study suggested that many RAMQ diagnoses of COPD lacked validity [
This retrospective observational cohort study confirmed that patients with COPD and exacerbations contribute significantly to the COPD-related burden of illness. We found that COPD exacerbations were associated with increased HCU in patients with COPD in Québec, Canada. This effect was more pronounced for patients who experienced severe exacerbations compared with no or moderate exacerbations. The study findings support that reducing the risk of exacerbations in patients with COPD could substantially reduce the health care burden of COPD in Québec, Canada.
The current affiliation for Alexander Simidchiev is Department of Functional Diagnostics, Medical Institute MVR, Sofia, Bulgaria.
Tam Dang-Tan and Xiwu Lin were employees of GSK at the time of the study and hold stocks/shares in the company. Alexander Simidchiev was employee of GSK at the time of the study and holds stocks/shares in the company and is now an employee of Medical Institute MVR, Sofia, Bulgaria; Afisi S. Ismaila is an employee of GSK and holds stocks/shares in the company and is also affiliated with McMaster University, Hamilton, ON, Canada. Shiyuan Zhang, Ruben V. Tavares, Alexander Simidchiev, Diane Corriveau, Richard H. Stanford, and Gilbert A. Nadeau are employees of GSK and hold stocks/shares in the company; John S. Sampalis, Melissa Stutz, and Julie Vaillancourt are employees of JSS Medical Research Inc.; Daria Parsons is on contract with JSS; John S. Sampalis is affiliated with JSS Medical Research Inc. and McGill University, Montreal, QC, Canada.
Tam Dang-Tan, Shiyuan Zhang, Ruben V. Tavares, Melissa Stutz, Afisi S. Ismaila, Julie Vaillancourt, Diane Corriveau, and Richard H. Stanford contributed to the study concept and design and were involved in data analysis and interpretation. Xiwu Lin, Gilbert A. Nadeau, and Alexander Simidchiev were involved in data analysis and interpretation. Daria Parsons participated in data acquisition, analysis, and interpretation. John S. Sampalis contributed to the study concept and design, data acquisition, analysis, and interpretation. All authors were involved in the preparation and review of the manuscript and approved the final version to be submitted.
This study was funded by GSK (Study no. 201121 [HO-13-14099]). Editorial support (in the form of writing assistance, assembling tables and figures, collating author comments, grammatical editing, and referencing) was provided by Alex Lowe, Ph.D., Fishawack Indicia Ltd., UK, and was funded by GSK.