Literature - Lee DS, Straus SE, Farkouh ME, et al. - N Engl J Med. 2023 Jan 5;388(1):22-32. doi: 10.1056/NEJMoa2211680

Introduction and methods

Background

For patients with acute HF (AHF), the first point of medical contact is often the emergency department (ED). There, physicians have traditionally relied on their clinical judgment to decide whether to discharge these patients or admit them to the hospital [1]. However, an obstacle for the selection of the patients is the inability to accurately predict the occurrence of adverse events based on risk estimation by physicians alone [2].

Aim of the study

The authors assessed the effect of a strategy to support clinical decision-making for the discharge or admission of patients who presented to the ED with AHF on clinical outcomes compared with usual care.

Methods

The COACH (Comparison of Outcomes and Access to Care for Heart Failure) trial was a multicenter, cross-sectional, stepped-wedge, cluster-randomized trial conducted in Ontario, Canada. The authors randomly assigned 10 hospitals to staggered start dates for one-way crossover from the control phase (usual care) to the intervention phase. During the trial, 5452 patients were enrolled (2972 during control phase and 2480 during intervention phase). Median follow-up time was 280 days (IQR: 82–520) for the control phase and 144 days (IQR: 64–286) for the intervention phase.

The intervention consisted of the use of a validated point-of-care algorithm to stratify AHF patients according to the risk of death, in addition to clinical judgment. For patients who had a low risk of both death within 7 days and death within 30 days, the recommendation was early discharge (≤3 days) and standardized transitional care in the Rapid Ambulatory Program for Investigation and Diagnosis of Heart Failure clinic, while hospital admission was recommended for high-risk patients. For intermediate-risk patients, clinicians were advised to use their clinical judgment, but the general guidance was to admit patients who had an intermediate-to-high risk and to consider early discharge for patients who had a low-to-intermediate risk. For the control phase, risk scores could be calculated after completion of the trial.

Outcomes

The first coprimary endpoint was a composite outcome of all-cause mortality or hospitalization for CV causes within 30 days after presentation to the ED. The second coprimary endpoint was the composite outcome within 20 months after presentation. Secondary endpoints were hospitalization for CV causes, hospitalization for HF, all-cause mortality, and a composite outcome of first nonelective ED visit, all-cause mortality, or hospitalization for CV causes.

Serious adverse events included death or hospitalization for either CV causes or any cause that occurred in low- or intermediate-risk patients after early discharge and before the first outpatient visit.

Main results

• Early discharge was frequently observed in low-risk patients in the control phase (58.2%) and intervention phase (57.0%) and less often in intermediate-risk patients (44.2% and 50.7%, respectively) and high-risk patients (27.0% and 19.1%, respectively).
• Death from any cause or hospitalization for CV causes within 30 days (i.e., first coprimary endpoint) occurred in 301 patients (12.1%) who were enrolled during the intervention phase and in 430 patients (14.5%) enrolled during the control phase (adjusted HR: 0.88; 95%CI: 0.78–0.99; P=0.04).
• When stratified by risk group, there were no significant differences between the intervention and control phase groups in the number of low-risk, intermediate-risk, or high-risk patients meeting the first coprimary endpoint.
• Within 30 days, the risk of hospitalization for CV causes appeared to be lower during the intervention phase than during the control phase (8.1% vs. 10.6%; adjusted HR: 0.85; 95%CI: 0.74–0.98), as was the risk of HF hospitalization (6.1% vs. 8.0%; adjusted HR: 0.81; 95%CI: 0.69–0.95). The incidence of the other secondary endpoints did not differ between the intervention and control phase groups.
• The cumulative incidence of death from any cause or hospitalization for CV causes within 20 months (i.e., second coprimary endpoint) was 54.4% (95%CI: 48.6–59.9) among patients enrolled during the intervention phase and 56.2% (95%CI: 54.2–58.1) among those enrolled during the control phase (adjusted HR: 0.95; 95%CI: 0.92–0.99).
• In low- and intermediate-risk patients, no deaths or hospitalizations for any cause occurred before the first outpatient visit within 7 days, and fewer than 6 events occurred within 30 days after discharge. In addition, the risk of serious adverse events within 30 days did not appear to be higher in patients enrolled in the intervention phase than that in patients enrolled during the control phase.

Conclusion

References

Find this article online at N Engl J Med.