Abstract 13529: Outcomes After Emergency Department Visits for Dyspnea: Population-Based Interrupted Time-Series Analysis of Implementation of B-Type Natriuretic Peptide Assays

Abstract

Introduction: The population-level impact of hospital-based natriuretic peptide (NP) implementation in a universal health care system is unclear. We examined temporal associations between introduction of NP testing in-hospital with outcomes after emergency department (ED) visits for dyspnea. Hypothesis: Implementation of NP testing is associated with improved outcomes for ED patients seeking care for dyspnea. Methods: Administrative databases were linked to identify adults ≥40 years of age with a first ED visit for dyspnea between 2014 and 2019 at hospitals introducing onsite NP testing between 2016 and 2018 in Ontario, Canada. We calculated quarterly rates of 1-year age- and sex-standardized mortality and readmission (all-cause, cardiovascular, heart failure [HF]), restricted to 2 years before and after NP introduction to minimize temporal advances in treatment. We conducted an interrupted time series analysis using linear regression and Newey-West autocorrelation adjusted standard errors to quantify rate of change in outcomes before and after NP introduction. Time zero at each hospital was set at 2 years prior to the introduction of NP tests for each hospital (point of interruption). Results: We studied 20,294 patients (median age 69 years, 52% female) before and 21,857 patients (median age 68 years, 53% female) after NP introduction across 16 hospitals. The cohort before NP introduction had a higher prevalence of prior HF and chronic obstructive pulmonary disease (P<0.01). Rates of all outcomes were stable prior to NP introduction. Following NP introduction, there were significant declines in rates of all-cause mortality (-1.5/100 persons per year), all-cause readmission (-3.6/100 persons per year), cardiovascular readmission (-1.4/100 persons per year) and HF readmission (-0.8/100 persons per year; Figure). Conclusions: Introduction of hospital-based NP tests was associated with decreasing rates of adverse outcomes after ED visits for dyspnea.