Guideline-Referral Criteria and Risk Profiles of Outpatients Referred to a Specialised Heart Failure Clinic

Abstract

Background

Specialised heart failure (HF) care improves outcomes for patients with HF. To understand the risk profiles of HF outpatients referred to a specialised clinic, we evaluated referral reasons, predicted risk, and the presence of guideline-recommended referral criteria at a large specialised HF clinic.

Methods

We conducted a cross-sectional study including outpatients with HF (≥ 18 years old) referred from November 2021 to November 2022. We calculated 1-year predicted mortality with the use of the Seattle Heart Failure Model (SHFM) and the I-NEED-HELP referral criteria. We compared median SHFM-predicted mortality with referral reasons and the I-NEED-HELP criteria by means of Kruskal-Wallis, Wilcoxon rank-sum, chi-square, and Fisher exact tests.

Results

Among 245 consecutive HF outpatients included, median SHFM-predicted 1-year mortality was 4% (interquartile range [IQR] 2%-8%). Reasons for referral included evaluation for advanced therapies (29%), medication optimisation (23%), diagnostic evaluation (19%), post-hospitalisation/emergency department visit (14%), ongoing HF management (12%), patient request (2%), and transition to adult care (1%). The median SHFM-predicted 1-year mortality did not differ significantly by referral reason (P = 0.11) but differed significantly among patients meeting any (5%, IQR 3%-9%) vs no (3%, IQR 2%-5%) I-NEED-HELP criteria (P < 0.001). Across referral reasons, the presence of any I-NEED-HELP criteria differed significantly (P < 0.001); most patients referred for advanced therapies evaluation (96%) and diagnostic evaluation (94%) met at least 1 criterion.

Conclusions

Patients referred to a specialised HF clinic have a wide risk range. The difference in predicted mortality among patients meeting any vs no I-NEED-HELP criteria appears clinically insignificant. Incorporating model-predicted risk at the time of referral can guide triage and patient prioritisation.