Reading Time: 3 minutes

While once the stuff of science fiction, there have been significant advances in artificial intelligence (AI) over the last several decades. Today, AI is transforming industrial sectors across the world, including finance, transportation, national security, travel, city management and healthcare.

Artificial intelligence (AI) is defined by the English Oxford Living Dictionary as “the theory and development of computer systems able to perform tasks normally requiring human intelligence, such as visual perception, speech recognition, decision-making, and translation between languages.”

In the field of advance care planning (ACP), there is much anticipation and excitement about the potential of AI. AI applications may be used to target patient populations, facilitate clinical trials, improve end-of-life care conversations, and improve health equity.

 

Better identify patients in need of ACP

AI can analyze large amounts of data – often in real time – far more rapidly, accurately, and efficiently than humans. It is useful for detecting abnormalities and patterns, extracting insights, and making predictions.

In healthcare, AI applications can quickly process and evaluate data from electronic health records (EHRs) to facilitate management of clinical care, quality improvement, regulatory compliance, and research. AI also holds the promise of increasing the utilization of ACP,  palliative care, and hospice care.

AI-based algorithms and predictive models are being tested and used at major academic medical centers – including Penn Medicine, NYU Langone Health, Stanford Medicine, and Duke Health – to increase ACP by identifying patients at high risk for mortality and flagging them in the EHR for an intervention. In some cases, clinicians receive a notification message.

 

Accelerate clinical trials and support decision-making

Natural Language Processing (NLP) is a branch of AI that enables computers to extract keywords and phrases, understand the intent of language, and translate it to another language or generate a response. When it comes to large clinical trials studying decision-making related to ACP needs and documentation, manual chart review is the gold standard; however, it is time-consuming.

A recent study found that the use of NLP to identify ACP documentation was more efficient – saving between 29 to 115 minutes per patient – and just as accurate as manual review. Not only can NLP expedite clinical trial results to identify evidence-based ACP practices, but it also has the potential to support ACP clinical decision-making for individual patients.

 

Prepare clinicians for difficult conversations

For most clinicians, sharing a poor prognosis and initiating a conversation about end-of-life care is tough. Given that physicians receive little to no training in medical school about how to conduct these conversations in a meaningful way, it’s little wonder they find it challenging to broach ACP.

“During difficult conversations about facing the potential of one’s own death, patients are frightened and don’t know how to ask the right questions, and clinicians may oversimplify, omit, or sugar-coat information, or feel too pressed for time to address patients’ emotions,” Ehsan Hoque shared with University of Rochester’s Newscenter. Mr. Hoque, an associate professor of computer science at Rochester’s Hajim School of Engineering and Applied Sciences, is a pioneer in the area of research that shows how people can learn and improve their social and interpersonal skills by interacting with automated systems.

Thanks to AI, doctors may be able to learn how to engage in more effective, empathetic end-of-life conversations by practicing with an online avatar. Mr. Hoque’s Rochester Human-Interaction Computer Lab in collaboration with the University’s Center for Communication Disparities Research, led by palliative care expert Ronald Epstein, MD, developed AI-powered SOPHIE (Standardized Online Patient for Healthcare Interaction) with the goal of improving communication between terminally ill patients and their providers.

 

Rectify racial disparities

The U.S. is experiencing a health disparity crisis. A 2018 study by the W.K. Kellogg Foundation and Altarum estimated that racial inequities contribute to $93 billion in excess medical care costs and $42 billion in lost productivity per year.

The pandemic shined a light on the ongoing disparities in ACP and end-of -life care. While racial and ethnic minority groups have experienced a disproportionate rate of death from COVID-19, they are less likely to participate in ACP and more likely to receive poor end-of-life care. Significant racial disparities have also been found when it comes to the provision of palliative care services among seriously ill patients.

Racial discrimination and implicit bias diminish access to equitable care according to a recent report from the Urban Institute and the Robert Wood Johnson Foundation. Research has shown that many clinicians avoid ACP with certain racial and ethnic groups based on biases and assumptions.

AI has the potential to rectify health disparities in end-of-life care by offering clinicians consistent, nonbiased identification of patients who will benefit from ACP. Pairing AI with access to patient decision aids that are inclusive and structured around discovering a patient’s individual values, concerns, and goals for medical treatment can potentially promote health equity.

 

Empower your patients

By harnessing the power of AI, health systems, clinicians, and researchers have the potential to transform ACP, ensuring that every patient has access to the medical care they wish to receive throughout their healthcare journey.

To learn how ACP Decisions can help your healthcare organization facilitate shared decision making and empower patients to make informed medical decisions, contact us!

Don’t forget to share this post!