Race and Sex Differences in Mortality in Individuals with Chronic Obstructive Pulmonary Disease

Abstract

Rationale: Despite differences in chronic obstructive pulmonary disease (COPD) comorbidities, race- and sex-based differences in all-cause mortality and cause-specific mortality are not well described. Objectives: To examine mortality differences in COPD by race-sex and underlying mechanisms. Methods: Medicare claims were used to identify COPD among REGARDS (Reasons for Geographic and Racial Differences in Stroke) cohort participants. Mortality rates were calculated using adjudicated causes of death. Hazard ratios (HRs) for mortality comparing race-sex groups were modeled with Cox proportional hazards regression. Results: In the 2,148-member COPD subcohort, 49% were women, and 34% were Black individuals; 1,326 deaths occurred over a median 7.5 years (interquartile range, 3.9-10.5 yr) follow-up. All-cause mortality per 1,000 person-years comparing Black versus White men was 101.1 (95% confidence interval [CI], 88.3-115.8) versus 93.9 (95% CI, 86.3-102.3; P = 0.99); comparing Black versus White women, all-cause mortality per 1,000 person-years was 74.2 (95% CI, 65.0-84.8) versus 70.6 (95% CI, 63.5-78.5; P = 0.99). Cardiovascular disease (CVD) was the leading cause-specific mortality among all race-sex groups. HR for CVD and chronic lung disease mortality were nonsignificant comparing Black versus White men. HR for CVD death was higher in Black compared with White women (HR, 1.44; 95% CI, 1.06-1.95), whereas chronic lung disease death was lower (HR, 0.44; 95% CI, 0.25-0.77). These differences were attributable to higher CVD risk factor burden among Black women. Conclusions: In the REGARDS COPD cohort, there were no race-sex differences in all-cause mortality. CVD was the most common cause of death for all race-sex groups with COPD. Black women with COPD had a higher risk of CVD-related mortality than White women. CVD comorbidity management, especially among Black individuals, may improve mortality outcomes.

Keywords: cardiovascular disease; chronic obstructive pulmonary disease; mortality; race.

Figure 1.

Identification of a cohort of participants in the REGARDS (Reasons for Geographic and Racial Differences in Stroke) study with evidence of chronic obstructive pulmonary disease (COPD). All available Medicare claims data that were filed before the participant’s in-home visit were examined for the following COPD-specific International Classification of Diseases, Ninth Revision codes: 491.X chronic bronchitis, 492.X emphysema, or 496 chronic airway obstruction. Evidence of COPD was defined as having any one inpatient hospitalization with a COPD-specific diagnosis in any position in the discharge diagnoses or two outpatient encounters from different service dates with COPD diagnosis in any position. REGARDS participants with evidence of COPD were then followed as part of the REGARDS study.

Figure 2.

Cohort flow diagram. This diagram outlines the identification of the final analytic cohort consisting of 2,148 REGARDS (Reasons for Geographic and Racial Differences in Stroke) participants with evidence of chronic obstructive pulmonary disease (COPD) at their baseline visit.

Figure 3.

Cause-specific age-adjusted mortality rates for men and women by race. The leading causes of death for all race-sex groups were cardiovascular disease, cancer, infection, and chronic lung disease. The age-adjusted mortality rates per 1,000 person-years are presented for each cause of death, comparing Black versus White individuals in each sex group.