Background: Heart failure is common and confers high mortality and morbidity. Our organization sought to improve heart-failure-patient care, working to reduce mortality index (MORTi) and 30-day readmission. The effort included a combination of education and incentives followed by EHR-based clinical decisions support systems (CDSS) within order sets, notes, and more. We sought to explore associations of outcomes with what we presumed to be the most impactful interventions – the novel, evidence-based admission order set which promoted diuresis and GDMT and a templated note (HFNOTE) – which automatically centralized diuretic and GDMT data and helped to capture complexity.
Methods: Patients were analyzed during our “Post2” period (12/1/24-6/30/25) which included both education and incentive interventions and the EHR CDSS interventions. Outcomes data was obtained from Vizient. EHR data was exported from Qlik. Data was matched and analyzed in Excel. If a match could not be found that data was excluded. Data is reported as “index” (observed/Vizient-model expected). Given interacting variables, results are reported for descriptive purposes only.
Results: A total of 4,865 patients were analyzed across 10 hospitals. The overall Post2 MORTi was 0.83 (0.027/0.032) and 30-day readmission index (30Di) was 1.07 (0.15/0.14). For patients with admission order set use, the MORTi was 0.66 (0.019/0.029) compared to 0.91 (0.031/0.034) for those without use; 30Di was 1.12 (0.16/0.14) and 1.04 (0.15/0.14) for use and non-use, respectively. For patients with HFNOTE use, the MORTi was 0.50 (0.016/0.031) compared to 1.12 (0.037/0.033) for those without use; 30Di was 1.08 (0.15/0.14) and 1.06 (0.15/0.14) for use and non-use, respectively.
Conclusions: Though there are promising trends, we note likely confounders that will need to be explored in a logistic regression. We suspect some of the HFNOTE MORTi association relates to our presumption that patients who die do so early in the ED or ICU and we suspect HFNOTE is more commonly used on the floors. Though HFNOTE does promote GDMT which is associated with mortality benefit, we would anticipate seeing these results downstream. We had hoped to see decreased 30-day readmissions but there is no large effect.The admission order set was designed to promote early and aggressive diuresis which has been shown to reduce in-patient mortality, so we are heartened to see a lower MORTi with use vs non-use. As with HFNOTE, we will need to explore confounders that make some of the order set use or non-use benefit non-casual. Similarly, the higher 30Di may be non-causal. It is also possible, however, that the promotion of GDMT and diuresis leads to unexpected higher readmission due to adverse effects that are not mitigated in a non-clinical-trial environment (in RCTs the recommendations found in the order set demonstrated lower 30-day readmission).Thus, though we note dramatic decrease in MORTi across both interventions (which aligns with an overall 32% decrease of MORTi during this period compared to a seasonally-adjusted comparator), we suspect true effects may be more modest; we will seek to isolate effects. Thereafter we will iteratively improve efficacy and breadth of deployment.