Hospice care and palliative care are two beneficial options that all patients have for end-of-life care. When a patient is best suited for end-of-life care, the first step is having a conversation with the dying patient, a sensitive act that can be very difficult for the patient and their families. Though challenging, initiating a conversation with the patient can allow for a more seamless transition to hospice or other end-of-life care options, rather than acute treatment that could negatively impact the patient and family’s well-being and experience, especially during their final days.
Appropriate end-of-life treatment options and health-related costs
Recently, a great deal of importance has been given to end-of-life care as it has come to light that some patients may not necessarily be receiving the medically appropriate care during the last phase of their lives. The number of Americans using end-of-life care services continues to grow and was estimated at 1.7 million for palliative care alone, or about 46% of those who die.
Yet these services are being utilized too late: the average length of stay in hospice care in 2014 was only 17 days. Additionally, 35.5% of hospice patients were discharged or died within 7 days of hospice enrollment. High costs associated with end-of-life care have also triggered an interest in the overall health care system owing to its impact on Medicare and Medicaid. Though care-related costs are a factor, quality of care is most the most crucial element, giving the dying patient and their family a more comfortable and compassionate experience during the end-of-life.
Predictive analytics assists caregivers and health systems facing end-of-life care quality and cost challenges
While a physician is likely aware when a patient’s condition warrants end-of-life care, the variable course keeps the clinical team and the patient focused on the hope of improvement and hesitant to talk about predicting mortality. Studies show that physicians are inaccurate when predicting time to patient death and tend to be overly optimistic about patient survival.
This deficit can result in late referrals to hospice, set false expectations of survival for families and patients, and lead to missed opportunities to focus on quality of life. Emerging predictive analytics solutions allow physicians to move away from the low accuracy in predicting a patients’ end of life and spur the vital conversation where sharing prediction statistics helps make informed decisions.
HELP for a transition to compassionate care with a quality, cost-conscious patient experience in mind
What assists a peaceful transition for patients to end-of-life care? “HELP”: combining technology with candid dialogue. In a care setting, the following four actions can improve quality and experience for patients and caregivers, control costs and lead to early intervention to transition patients to the best options for care.
H – Have the conversation: Doctors and nurses can lead discussions with patients and their loved ones, providing them with options for hospice and palliative care. In one study, 90% of adults in the Medicare population said they would prefer to receive end-of-life care in their home if they were terminally ill, yet only one-third of Medicare beneficiaries die at home. Like talking through advanced care planning, which Medicare now reimburses (3), an open conversation with a clinician about the benefits of end-of-life care could lead to a patient choosing more comfortable care options in the last stages of their life.
E – Elevate quality of care: Prolonged acute care treatments can cause discomfort leading to diminished quality of care during the final phase of life. This period is commonly marked by aggressive medical interventions representing a significant burden to the dying patient and their family, without improvement in outcomes or quality of life. Both palliative and hospice care offer the patient a compassionate option for treatments in the final phase of life. The options should be provided early and often for a patient at the end of life whenever possible.
L – Lower care costs: In 2011, the United States Medicare cost for medical care directed at the last six months of a patient’s life was $170 billion out of the total $554 billion Medicare spend. Improving end-of-life care can control costs for healthcare organizations, who endure the financial burden, and the U.S. health system’s high costs of futile care during end of life. Palliative and hospice care programs increase value-based care by both improving quality and reducing costs of care (5). Fortunately, there has been a cultural shift in medicine towards end-of-life care, and more dying patients are adopting palliative and hospice care.
P – Predictive analytics to stay ahead: End-of-life prediction platforms using artificial intelligence, specifically machine learning, are currently available. These tools ingest and integrate multiple data sources including claims and electronic health records (EHR) data to detect the patterns indicating impending mortality far better than traditional heuristic approaches for forecasting end of life. Predictive analytics tools can help lead data-driven conversations about end of life care to assure patients receive appropriate care, improve their experience and lower futile care costs.
Leveraging technology for quality
Now is the time for doctors, nurses and health systems to evaluate how they can minimize the predictive deficit that prevents accurately forecasting which patients are nearing end of life. Emerging technologies, such as risk prediction, are helping providers begin to engage in sensitive conversations, in the effort to provide patients with better quality of life during their final phase. An informed discussion is helpful, mainly when conducted before times of high emotional stress and provides the patients and their families a more precise understanding, and, ultimately, better overall experience at the end of life.
Whende M. Carroll, MSN, RN-BC, is the Director of Nursing Informatics at KenSci, Inc. In this role, Whende collaborates with clinicians, engineers and data scientists to develop advanced analytics solutions that improve care delivery, quality, outcomes, and patient and provider experience.
