Regional outcomes of heart failure in Canada

Background — Heart failure is a condition associated with significant mortality and morbidity. However, demographic features and outcomes following hospitalization for heart failure, and associated regional comparisons have not been performed in Canada.

Methods — Anonymously rendered records of patients hospitalized for incident heart failure in Canada were selected from the Canadian Institute for Health Information discharge abstract and hospital morbidity databases from fiscal years 1997/1998 to 1999/2000. The demographics, in-hospital mortality rate and heart failure readmission rates were compared among provinces and health regions.

Results — A total of 83,406 patients were hospitalized for heart failure across Canada during the study period. The number of cases increased dramatically with each decade after age 50 years, with 85% of hospitalized patients being age 65 years and over. On average, in-hospital mortality per index admission in Canada was 9.5 deaths per 100 hospitalized cases. While the greatest burden of readmissions was among those 65 years of age and over, heart failure readmission rates were similar across age groups. Among all patients surviving the index admission, heart failure readmission rates were 8.7%, 14.1% and 23.6% at 30 days, 90 days and one year, respectively. The highest age- and sex-adjusted in-hospital mortality rates were 11.9% (95% CI 10.6 to 13.2) in Newfoundland/Labrador and 11.6% (95% CI 10.6 to 12.7) in Nova Scotia. The highest readmission rates at one year were 26.9% (95% CI 24.9 to 28.9) in Newfoundland/Labrador, 26.3% (95% CI 25.0 to 27.7) in Saskatchewan and 25.2% (95% CI 24.3 to 26.1) in British Columbia. There were significant regional variations in heart failure readmission rates and mortality.

Conclusions — There is a great burden of heart failure in Canada, increasing significantly with age. The mortality and readmission rates for this condition are high and exhibit variation among health regions and provinces. Factors contributing to regional variations in these outcomes merit further study.