Self-Care Intervention Fails to Significantly Improve Outcomes Among Patients With Heart Failure

Hoping to improve outcomes among patients with acute heart failure discharged from the emergency department, study findings show that 30-day improvements from a self-care intervention were not sustainable through 90 days.

Hoping to improve 90-day outcomes among patients with acute heart failure discharged from the emergency department (ED), investigators instead found that 30-day outcome improvements from a self-care intervention were not sustainable through 90 days, reports JAMA Cardiology.

“Up to 20% of patients who present to the emergency department with acute heart failure are discharged without hospitalization,” the authors noted. “Compared with rates in hospitalized patients, readmission and mortality are worse for ED patients.”

Therefore, they set out to see if their home visit and telephone-based coaching self-intervention program could improve 90-day outcomes among these patients in the unblinded, parallel-group, multicenter GUIDED-HF trial that took place between October 28, 2015, and September 5, 2019. Patient participants were randomized 1:1 to the tailored self-care intervention program (n = 235) or usual care (n = 244), comprising a structured discharge process, including heart failure medication reconciliation, heart failure medication prescriptions, and an appointment for a 7-day follow-up with a heart failure clinician.

With a primary outcome of global rank of cardiovascular (CV) death, heart failure–related events (eg, unscheduled clinic visit due to heart failure, ED revisit, or hospitalization), and changes in the Kansas City Cardiomyopathy Questionnaire-12 (KCCQ-12) summary score (SS) at 90 days, the results show no significant difference between the study arms (HR, 0.89; 95% CI, 0.73-1.10; P = .28).

Most of the study participants were African American (n = 302; 63%) and men (n = 305; 64%). Slightly more than one-third (37%) had a previous ejection fraction above 50%. The median (interquartile range [IQR]) age was 63.0 (54.7-70.2) years. The authors noted that African American patients present more often to the ED instead of being directly admitted to a hospital, as well that minority patients more often use EDs to access primary care.

“These factors underscore the importance of emergency care clinician–targeted transition of care initiatives to avoid unnecessary hospitalization while ensuring patient safety and well-being,” they noted.

For this study, the criteria for a vulnerable population were brief health literacy score below 9, being of a non-White race or ethnicity, and having a national area deprivation index (ADI) score above 85. Additional results show that close to equal numbers of patients in the intervention and control groups had low literacy scores, at 11% vs 14%; 45% of all patients had low socioeconomic status; and the median ADI was 83 in each arm, with close to equal IQRs:

  • Overall: 58 to 96
  • Intervention arm: 58 to 96
  • Control arm: 58 to 95

Eighty-nine percent of patients in the intervention arm had telehealth (11%) or in-person home visits (78%) during the study period, whereas 80% and 74% participated in the coaching calls within the first 30 days and between 30 to 90 days, respectively. More patients who had in-person home visits also saw an improvement in CV death and heart failure events vs those who had telehealth visits: 32% vs 20% (P = .23).

Also, despite the intervention group seeing higher median changes in KCCQ-12 scores at the 30- and 90-day marks, these results actually plateaued after 30 days:

  • Intervention group:
    • 30 days: 9.5 points (median, –7.3; IQR, 5.7-16.5)
    • 90 days: 10.9 (median, –2.6; IQR, 9.4-25.4)
  • Control group:
    • 30 days: 5.7 points (median, –1.6; IQR, 9.5-22.9; P = .048)
    • 90 days: 9.4 (median, –2.6; IQR, 10.9-25.8; P = .75)

Additional analyses of both groups determined the following outcomes:

  • No significant difference in unadjusted 90-day CV and heart failure events: 32% (intervention) vs 36% (usual care)
  • No significant difference in adjusted 90-day CV and heart failure events (HR, 0.78; 95% CI, 0.57-1.06; P = .11)
  • No significant risk reduction in 30-day CV death and heart failure events: 18% vs 14% (P = .18)
  • The intervention group experienced a 20% risk reduction in 30-day global rank outcome (HR, 0.80; 95% CI, 0.64-0.99; P = .04)

Overall, there was no apparent risk reduction by day 90 for both CV death and heart failure events, nor was there a reduction in relative risk for the global rank primary outcome (HR, 0.93; 95% CI, 0.73-1.18; P = .55).

The authors attribute this finding to the small size of their study, which gave them less power to achieve statistical significance.

“These findings suggest that either the current self-care strategy may benefit from additional home visits after the 30-day follow-up period to potentially prevent the apparent diminished effect of our intervention over time or the HF clinician visit in the first 7 days was an important component of our intervention,” they noted.


Collins SP, Dandan L, Jenkins CA, et al. Effect of self-care intervention on 90-day outcomes in patients with acute heart failure discharged from the emergency department: a randomized clinical trial. JAMA Cardiol. Published online November 18, 2020. doi:10.1001/jamacardio.2020.5763