Additional data, feedback on hospital care did not improve heart failure outcomes

A program designed to improve hospital care for patients with heart failure, the leading cause of hospitalization among adults over age 65, did not bring additional benefits beyond existing hospital quality improvement programs in a randomized controlled trial presented at the American College of Cardiology's 70th Annual Scientific Session.

Heart failure is a condition in which the heart becomes too weak or too stiff to pump blood effectively to the rest of the body. It causes symptoms such as swelling and fluid retention, shortness of breath and coughing.

In the CONNECT-HF study, one group of hospitals received additional auditing and feedback on performance and their quality of care for heart failure patients from an external group—in addition to their own existing quality improvement program. However, the group of hospitals showed no significant differences compared with hospitals that did not receive the intervention in terms of heart failure rehospitalization or death and a composite score for heart failure care quality.

"We were disappointed to find no difference," said Adam DeVore, MD, a cardiologist at Duke University Medical Center and the study's lead author. "These principles of audit and feedback don't seem to improve upon what already exists in terms of quality improvement for heart failure. This strategy doesn't work above what we are already doing; we need to find other ones that do, and we have a lot of work ahead of us."

Over the course of three years, the study enrolled 5,647 patients treated for heart failure with reduced ejection fraction at 161 participating hospitals. All hospitals already had an on-site quality improvement program designed to ensure compliance with guideline-based practices for heart failure treatment and reduce negative outcomes such as rehospitalization and death. For those randomized to receive the quality improvement intervention, researchers worked with each hospital's in-house quality improvement team to review and improve existing programs to reinforce evidence-based practices. For example, if a hospital found that many heart failure patients were not taking prescribed medicines reliably, the quality improvement team might work with in-house pharmacists to ensure patients start taking their medications before being discharged. The researchers independently tracked hospital discharges and post-discharge follow-up care and provided feedback to hospital care teams on performance metrics, a strategy of audits and feedback that had worked for other quality improvement initiatives in previous studies.

The first co-primary endpoint, heart failure rehospitalization or death from any cause, occurred in about 39% of patients in both study groups. There was no significant difference between groups in terms of this endpoint or the second co-primary endpoint, a composite score for heart failure care quality at the time of last follow-up.

DeVore said that there remains a need to improve the quality of care for people with heart failure and suggested that future interventions could employ digital and patient-facing tools or focus on care delivery outside of the hospital, such as at heart failure clinics.

"Patients with heart failure aren't getting the medical care they need to stay out of the hospital and live a good quality life without symptoms of heart failure, but it's not clear that the hospital should be the place to fix this problem," DeVore said. "Based on our findings, I don't think we're going to move the needle if we continue focusing our attention on the hospital if there is already a quality improvement program there."

The trial is unique in its use of a randomized study design for a quality improvement initiative. DeVore said that a patient panel helped inform the trial design, which researchers saw as a strength of the study. The trial was stopped early as a result of the COVID-19 pandemic and did not enroll as many patients as planned but was still adequately powered to reveal any differences between study groups, DeVore said.