A PROPENSITY-MATCHED COHORT STUDY TO ASSESS THE EFFECTIVENESS OF A COMMUNITY HOSPITAL BASED HEART FUNCTION CLINIC

Abstract

BackgroundExisting evidence is conflicting in terms of the effectiveness of Heart Function Clinics (HFC) on heart failure (HF) patients’ clinical outcomes (emergency department visits, hospital readmissions, hospital length of stay and mortality). Our objectives were to compare the healthcare outcomes for HF patients enrolled in a community hospital-based HFC versus HF patients residing in the same catchment area but not enrolled in the same HFC.Methods and ResultsA total of 344 HF patients were referred to an HFC located inside of a community hospital in Ontario between 2014 and 2017, of whom 246 attended their first HFC visit (i.e., index date) between February 1, 2014 and March 1, 2017. We used propensity-scores to match the HFC and control patients diagnosed with HF residing in the Central LHIN in Ontario between 2014-2017 who were not enrolled in the HFC. Primary outcomes were all-cause and HF Emergency Department (ED) visits and all-cause and HF hospitalizations at 6 and 12 months from the index date. Secondary outcome measurement was mortality 6 and 12 months after index date. The HFC patients had reduced all-cause and HF related ED visits 6 months post index date (all-cause ED visit risk difference (RD) of 35.7% versus 16.8% [p=0.0487] and HF ED visit average count reduction of 0.69 versus 0.21 [p=0.0018]) and as well 12 months post index date (all-cause ED visit RD of 18.9% versus 0.3% [p=0.0004] and HF ED visit average count reduction of 0.49 versus 0.15[p=0.1256]). HFC patients also had reduced all-cause and HF hospitalizations 6 months post index date (all-cause hospitalization RD 39.3% versus 24% [p=0.9394] and HF hospitalization RD 42.2% versus 29.3% [p=0.2194]) and as well as 12 months post index date (all-cause hospitalization RD 26.2%% versus 12.2% [p=0.2771] and HF hospitalization RD 35.1% versus 26.6% [p=0.0678]). Within 6- and 12-months follow-up, 12.2% and 22.4% of HFC patients died versus 24.7% and 32.7% of control patients respectively. The better outcomes of the HFC patients relative to the matched controls was despite the HFC patients having a significantly worse Charlson score (p=0.0076).ConclusionView Large Image Figure ViewerDownload Hi-res image Download (PPT) BackgroundExisting evidence is conflicting in terms of the effectiveness of Heart Function Clinics (HFC) on heart failure (HF) patients’ clinical outcomes (emergency department visits, hospital readmissions, hospital length of stay and mortality). Our objectives were to compare the healthcare outcomes for HF patients enrolled in a community hospital-based HFC versus HF patients residing in the same catchment area but not enrolled in the same HFC. Existing evidence is conflicting in terms of the effectiveness of Heart Function Clinics (HFC) on heart failure (HF) patients’ clinical outcomes (emergency department visits, hospital readmissions, hospital length of stay and mortality). Our objectives were to compare the healthcare outcomes for HF patients enrolled in a community hospital-based HFC versus HF patients residing in the same catchment area but not enrolled in the same HFC. Methods and ResultsA total of 344 HF patients were referred to an HFC located inside of a community hospital in Ontario between 2014 and 2017, of whom 246 attended their first HFC visit (i.e., index date) between February 1, 2014 and March 1, 2017. We used propensity-scores to match the HFC and control patients diagnosed with HF residing in the Central LHIN in Ontario between 2014-2017 who were not enrolled in the HFC. Primary outcomes were all-cause and HF Emergency Department (ED) visits and all-cause and HF hospitalizations at 6 and 12 months from the index date. Secondary outcome measurement was mortality 6 and 12 months after index date. The HFC patients had reduced all-cause and HF related ED visits 6 months post index date (all-cause ED visit risk difference (RD) of 35.7% versus 16.8% [p=0.0487] and HF ED visit average count reduction of 0.69 versus 0.21 [p=0.0018]) and as well 12 months post index date (all-cause ED visit RD of 18.9% versus 0.3% [p=0.0004] and HF ED visit average count reduction of 0.49 versus 0.15[p=0.1256]). HFC patients also had reduced all-cause and HF hospitalizations 6 months post index date (all-cause hospitalization RD 39.3% versus 24% [p=0.9394] and HF hospitalization RD 42.2% versus 29.3% [p=0.2194]) and as well as 12 months post index date (all-cause hospitalization RD 26.2%% versus 12.2% [p=0.2771] and HF hospitalization RD 35.1% versus 26.6% [p=0.0678]). Within 6- and 12-months follow-up, 12.2% and 22.4% of HFC patients died versus 24.7% and 32.7% of control patients respectively. The better outcomes of the HFC patients relative to the matched controls was despite the HFC patients having a significantly worse Charlson score (p=0.0076). A total of 344 HF patients were referred to an HFC located inside of a community hospital in Ontario between 2014 and 2017, of whom 246 attended their first HFC visit (i.e., index date) between February 1, 2014 and March 1, 2017. We used propensity-scores to match the HFC and control patients diagnosed with HF residing in the Central LHIN in Ontario between 2014-2017 who were not enrolled in the HFC. Primary outcomes were all-cause and HF Emergency Department (ED) visits and all-cause and HF hospitalizations at 6 and 12 months from the index date. Secondary outcome measurement was mortality 6 and 12 months after index date. The HFC patients had reduced all-cause and HF related ED visits 6 months post index date (all-cause ED visit risk difference (RD) of 35.7% versus 16.8% [p=0.0487] and HF ED visit average count reduction of 0.69 versus 0.21 [p=0.0018]) and as well 12 months post index date (all-cause ED visit RD of 18.9% versus 0.3% [p=0.0004] and HF ED visit average count reduction of 0.49 versus 0.15[p=0.1256]). HFC patients also had reduced all-cause and HF hospitalizations 6 months post index date (all-cause hospitalization RD 39.3% versus 24% [p=0.9394] and HF hospitalization RD 42.2% versus 29.3% [p=0.2194]) and as well as 12 months post index date (all-cause hospitalization RD 26.2%% versus 12.2% [p=0.2771] and HF hospitalization RD 35.1% versus 26.6% [p=0.0678]). Within 6- and 12-months follow-up, 12.2% and 22.4% of HFC patients died versus 24.7% and 32.7% of control patients respectively. The better outcomes of the HFC patients relative to the matched controls was despite the HFC patients having a significantly worse Charlson score (p=0.0076). Conclusion