Largest-ever study of palliative care demonstrates scalable strategy to increase support for seriously ill patients in the hospital

Ordering a palliative care consultation by “default” – via an automatic order programmed into the electronic medical record that doctors may cancel if they choose – is an effective strategy to give more hospitalized patients the opportunity to benefit from palliative care, and sooner, according to a new study led by researchers from the Perelman School of Medicine at the University of Pennsylvania. Palliative care is specialized medical care focused on relieving the symptoms and stress of a serious illness and improving quality of life, in alignment with a patient’s individual goals, values, and priorities. By changing the process for ordering palliative care consultation from doctors opting in and actively placing an order to having the opportunity to opt out and cancel an automatic order, the investigators increased palliative care consultation rates from 16.6 percent to 43.9 percent and decreased the time to consultation by 1.2 days in the largest-ever study of inpatient palliative care, published today in JAMA.

“While early palliative care consultation could help many patients with chronic serious illnesses better understand their diagnosis and align their treatment choices to their individual care goals, there’s never been an established approach to realizing that goal at scale. We found that a simple, pre-programmed order within the electronic medical record can get more palliative care to more people more quickly,” said lead author Kate Courtright, MD, MS, an assistant professor of Critical Care and Palliative Medicine. “This strategy was low-cost and easily implemented in community hospitals, which is where most Americans receive their health care.”

The study included more than 34,000 patients with chronic obstructive pulmonary disease (COPD), dementia, or kidney failure at 11 hospitals in eight states, all part of one of the nation’s largest non-profit health systems. Although palliative care consultation is recommended for millions of Americans with serious illnesses, many patients aren’t referred to palliative care or only receive a consultation at the end of life. Patients with COPD, dementia, and kidney failure have been underrepresented in past studies of palliative care delivery, which have largely focused on patients with cancer or heart failure in academic health centers. But these patients often experience challenges with coping, as well as breathlessness, anxiety, pain, and other symptoms that palliative care can help address through medications, other treatments, and/or referrals to other specialists. Palliative care can be given at any stage of treatment, including along with therapy intended to cure the condition.

Each of the 11 hospitals in the study had an established palliative care program, but to keep the study as realistic as possible, hospitals were neither encouraged to nor prohibited from increasing or decreasing palliative care staffing during the study. Over the course of the study, which took place between March 2016 and November 2018, the hospitals began enrolling patients to the study under usual care (data collected, but no intervention) and transitioned to the intervention phase (default orders via the electronic health record) over time, in a randomly determined order. During the intervention phase, clinicians canceled the default for fewer than 10 percent of patients for whom it was generated. Clinicians could also place their own order for a palliative care consultation at any point during the study.

“Our results suggest that the default order strategy was generally acceptable to clinicians, which is important because if we want to reach as many eligible patients as possible, we need to design approaches that are feasible for ‘real-world’ practice and not just in the research setting,” Courtright said.

Implementing default orders did not impact how long patients stayed in the hospital compared to usual care, perhaps because, even in the intervention group, less than half of patients actually received a consultation from a palliative care specialist. This may be due to limited staff resources or other factors. Anticipating that default orders would not necessarily guarantee a consultation, the study authors planned for a secondary outcomes analysis and found that for patients who only received palliative care consultation thanks to the default order, such care reduced the median length of stay by 9.6 percent. Additionally, the default orders led to more patients being discharged from the hospital to hospice care without affecting mortality, suggesting that such orders improved the quality and patient-centeredness of end-of-life care.

To build on the findings, these researchers at Penn’s Palliative and Advanced Illness Research (PAIR) Center have designed another randomized clinical trial. The planned study, which is supported by a $27.5 million funding award from the Patient-Centered Outcomes Research Institute (PCORI), will test a strategy that includes training and prompting generalists who make up a patient’s primary hospital team—including nurses, advanced practice providers, physicians, and social workers—to provide palliative care themselves.

“We know that health care systems have limited resources and need more evidence to guide future scaling and delivery of inpatient palliative care in an equitable and cost-effective manner,” said senior author Scott D. Halpern, MD, PhD, the John M. Eisenberg Professor of Medicine, Epidemiology, and Medical Ethics and Health Policy. “As we build on this work, our goal is to continuously improve inpatient palliative care so that all patients and families facing a serious illness have access to the support they need to carry on with their daily lives throughout their treatment journey.”

The study was supported by the National Institutes of Aging (UH3AG050311) and the Otto Haas Charitable Trust.