Individuals with end-stage heart failure were less likely to opt for left ventricular assist device (LVAD) as destination therapy when they and their physicians were part of a shared decision-making intervention, the DECIDE-LVAD trial suggested.
Patient knowledge of LVADs improved to a greater extent as well, scoring 59.1% on a test before and 70.0% on the same test after the changes, while smaller gains were made among controls at hospitals providing standard clinician and patient education (from 59.5% to 64.9%, P=0.03).
One month after hospitals started the intervention, their patients valued aggressive care to maximize survival less than controls (P=0.03) and favored LVAD therapy less as well (P<0.001), Larry Allen, MD, MHS, of University of Colorado School of Medicine in Aurora, and colleagues, reported online in JAMA Internal Medicine.
The intervention group was not more likely to get a treatment that aligned with their stated preferences. Nonetheless, they had a lower adjusted rate of LVAD implantation by 6 months (53.9% versus 79.9% among controls, P=0.008), with significant variation by site.
"These results suggest that institutional culture and processes can influence medical decisions in life-threatening illness," Allen's group concluded.
"We found that a significant number of patients facing high morbidity and mortality from heart failure decline destination therapy LVAD in favor of avoiding aggressive therapies. The declination rates reported here are among patients who have agreed to undergo formal evaluation; thus, we suspect that destination therapy LVAD declination may be more prevalent in the broader community," they noted.
"This reduction in choosing LVAD in the shared decision-making arm raises the alarming possibility that many patients undergoing usual care are consenting to an LVAD without being fully aware of the risks and benefits, and may have chosen otherwise with more information," an suggested.
In it, Albert Liu, MD, of University of California, San Francisco, and journal editor Gregory Curfman, MD, of Brigham and Women's Hospital in Boston wrote: "It is important for clinicians to be comfortable discussing risks of complex medical decisions, even potentially lifesaving ones such as destination therapy LVAD, to be sure that patients are making choices consistent with their wishes."
They noted that LVADs continue to be associated with considerable symptoms including pain, major depression, and organic mental syndromes, while device-related adverse events include stroke, infection, bleeding, pump thrombosis, ventricular arrhythmias, and right ventricular heart failure.
The randomized trial had a phased roll-out design in which each hospital started with a control period and then transitioned to an intervention period. The intervention consisted of clinician education and use of a pamphlet and video for patients. The six LVAD centers enrolled 248 patients in total.
DECIDE-LVAD suffered from missing data in many cases and a study population made up of mostly white men.
Disclosures
The study was supported through a Patient-Centered Outcomes Research Institute (PCORI) Program award, the National Heart, Lung and Blood Institute, the Heart Failure Society of America, the National Institute on Aging; and REDCap database hosting through University of Colorado supported by NIH/NCRR Colorado CTSI.
Allen reported consulting for Novartis, Boston Scientific, Janssen, Amgen, Duke Clinical Research Institute, and Grants from the Patient-Centered Outcomes Research Institute, National Institutes of Health, National Heart, Lung, and Blood Institute, and the American Heart Association.
Primary Source
JAMA Internal Medicine
Allen LA, et al "Effectiveness of an intervention supporting shared decision making for destination therapy left ventricular assist device: the DECIDE-LVAD randomized clinical trial" JAMA Intern Med 2018; DOI: 10.1001/jamainternmed.2017.8713.
Secondary Source
JAMA Internal Medicine
Liu A, Curfman G "The trade-offs when focusing on the mortality benefit" JAMA Intern Med 2018; DOI: 10.1001/jamainternmed.2018.0067.