A new palliative care intervention for U.S. patients with advanced heart failure did not improve quality of life or mood after 16 weeks of participation in a randomized trial.
“Future analyses and studies will examine both the patient factors and intervention components to find the right palliative care dose, for the right patient, at the right time,” wrote, of the University of Alabama at Birmingham, and coauthors. The study was published in .
“My first reaction is disappointment,”
Improvement of palliative care in heart failure patients might rest on who needs it most
“One thing to note,” Dr. Allen added in an interview, “is that, in this population of patients, some of the measures they were trying to improve were already relatively mild to start with. It may not be that the intervention didn’t help but that they picked a patient population that wasn’t particularly in need. If you treat someone who doesn’t have a problem, it’s hard to make them better.”
In a separate interview, Dr. Bakitas acknowledged a similar sentiment. “We were a little surprised until we looked at our sample,” she said. “We realized that we had recruited all these very high-functioning, good quality-of-life patients. What we then did was look at a subsample of patients who had low quality of life at baseline. Low and behold, the intervention had an effect. The patients who started with a poor quality of life had a statistically and clinically significant benefit. Their KCCQ score increased by over 5 points.”
As for next steps. Dr. Bakitas noted that they’re twofold: “One is refining the patient population who can benefit, and the second is working on the intervention and figuring out which pieces are the ones that provide the most benefit.
“Because of logistics and practical issues, not everyone in the study got all the intervention that they should have. Think of it like a drug trial; if someone misses a pill, they don’t get the full dose that we thought would work. We need to make sure our interventions have the right pieces in place. We don’t want to develop a great intervention that’s not practical for patients.”
Study design and outcomes
To determine the benefits of early palliative care for patients with heart failure, the researchers developed the ENABLE CHF-PC (Educate, Nurture, Advise, Before Life Ends Comprehensive Heartcare for Patients and Caregivers) intervention. This nurse-led program includes an in-person consultant followed by six telehealth nurse coaching sessions lasting 30-40 minutes and then monthly follow-up calls through either 48 weeks or the patient’s death.
To test the effectiveness of their intervention after 16 weeks, the researchers launched a two-site, single-blind randomized clinical trial made up of 415 patients who were 50 years or older with advanced heart failure. Among the patients, 53% were men and the mean age was 64 years; 55% were African American, 26% lived in a rural area, and 46% had a high school education or less. The average length of time since heart failure diagnosis was 5.1 years.
Patients were randomized evenly to receive either the ENABLE CHF-PC intervention (208) or usual care. The primary outcomes were quality of life (QOL), which was measured by the heart failure–specific 23-item Kansas City Cardiomyopathy Questionnaire (KCCQ) and the 14-item Functional Assessment of Chronic Illness Therapy–Palliative-14 (FACIT Pal-14), and mood, which was measured by the 14-item Hospital Anxiety and Depression Scale (HADS). Pain was measured via 3-item pain intensity and 2-item pain interference scales.
Effect size was measured as Cohen d or d-equivalent, where a small effect is 0.2, medium is 0.5, and large is about 0.854.
At baseline, the mean KCCQ score of 52.6 at baseline indicated a “fairly good” QOL across all patients. After 16 weeks, the mean KCCQ score improved 3.9 points in the intervention group, compared with 2.3 points in the usual care group (d = 0.07; [95% confidence interval, –0.09-0.24]). In addition, the mean FACIT-Pal-14 score improved 1.4 points in the intervention group compared to 0.2 points in the usual care group (d = 0.12 [95% CI, –0.03-0.28]). Only small differences were observed between groups regarding anxiety and depression, but pain intensity (difference, –2.8; SE, 0.9; d = –0.26 [95% CI, –0.43-0.09]) and pain interference (difference, –2.3; SE, 1; d = –0.21 [95% CI, –0.40 to –0.02]) demonstrated a statistically significant and clinically important decrease.