Administrative database study describing the prevalence of CVD in COPD patients, its impact on exacerbations and health care utilization.

Trial description: Background: Chronic Obstructive Pulmonary Disease (COPD) is a progressive chronic disease, including emphysema and chronic bronchitis, that affects over 770,000 Canadians, or 4% of those over the age of 35 years, and 8-10% over 40 years old. The disease is the leading cause of all hospitalizations in Canada with direct health care costs over $1.5 billion. In turn, cardiovascular disease (CVD) remains a leading cause of mortality and morbidity in the Western world despite the availability of effective treatments. A recent study by Gershon based on health insurance administrative databases from Ontario has shown that when compared to patients without COPD, those with the disease are at increased risk for CVD, as ascertained from health care utilization data (HCU). Epidemiological studies have shown that an important cause of mortality in COPD patients is related to CVD. Smoking and increased age are common risk factors for both conditions, with more recent evidence suggesting that a systemic inflammatory process may be the link between COPD and CVD.Objective: In the current investigation, the incremental morbidity, mortality and HCU that can be attributed to CVD over and above that of COPD in these patients will be investigated. Specifically, utilizing the databases of the Régie de l’assurance maladie du Québec (RAMQ), the current study will assess the incremental impact of cardiovascular disease on COPD-related HCU, rates of exacerbation, and mortality in Quebec.Study Design: This will be a retrospective cohort study of incident COPD conducted with administrative databases from Quebec, Canada. Patients with a diagnostic code and treatment for COPD between January 01, 2001 and March 31, 2011 will be used for this investigation. Patients will be followed to the last known claim, death, withdrawal from the RAMQ insurance coverage, or March 31, 2011.Data Analysis Methods: The difference in HCU between CVD+ and CVD- patients will be assessed using the following methods. The Relative Risk (RR) will be used to compare the proportion of patients with one or more HCU events in the CVD+ group to the CVD- group. Multivariate logistic regression will be used to adjust these differences for patient demographics, duration of COPD, COPD treatment, Charlson comorbidity index, and presence of individual comorbidities. The Incidence Density Rate Ratio (IDRR) will be used to compare the CVD+ and CVD- groups. Multivariate Poisson regression will be used to adjust these differences for patient demographics, duration of COPD, COPD treatment, Charlson comorbidity index and presence of individual comorbidities. Ninety five percent confidence intervals will be produced for the unadjusted and adjusted estimates of the IDRR. The CVD+ and CVD- groups will be compared with respect to the mean duration of hospital stay and ICU stay with the Student’s t-test. Multivariate mixed effects general linear models will be used to adjust these differences for the effect of patient demographics, duration of COPD, COPD treatment, Charlson comorbidity index and presence of individual comorbidities. The prevalence of CVD in COPD patients will be estimated by the proportion of patients that are classified in the CVD+ group. The incidence of CV events will be estimated from the number of new events in the cohort. The IDRR will be used to describe the incidence of CVD in COPD. The demographics, disease parameters and treatment of COPD profiles of CVD+ and CVD- patients will be compared using appropriate bivariate analyses and multivariate statistics. Logistic regression will be used to estimate the Odds Ratio for having a COPD exacerbation in the CVD+ group in comparison to the CVD- patients. Multivariate logistic regression model will be used to determine the Odds Ratio for COPD exacerbations in the CVD+ compared to the CVD-, adjusting for age, gender, duration of COPD, and presence of other comorbidities. Limitations: A number of limitations exist. Those generally related to the use of administrative databases, including but not limited to case ascertainment, the lack of laboratory and clinical assessments, and variable follow-up. The cohort is representative of incident COPD patients covered by the public plan and not the broader whole. Limited to an incident cohort, the study may not capture any long-term effects of COPD on the burden of illness attributed to CVD. The current study will supplement the paucity of research investigating the incremental burden of comorbidities in COPD.