Lalan Wilfong, MD, a medical oncologist/hematologist with Texas Oncology and physician liaison for value-based care for McKesson Specialty Health, discussed how artificial intelligence (AI) prediction models could improve end-of-life care for patients with cancer, per his presentation at the Inaugural MiBA Community Summit.
Historically, patients with cancer do not experience good end-of-life care, Wilfong stated. He noted that when asked, many patients reveal their end-of-life experience was not what they desired, often resulting in aggressive therapies like chemotherapy late in life, hospitalization, or admission to the intensive care unit (ICU). These patients would often prefer to spend their final days at home with family, focused on symptom management and comfort care rather than aggressive therapies.
Wilfong emphasized that continuous improvement in palliative care requires leveraging AI as a predictive model. The goal of such a model is to predict which patients are likely to face poor outcomes, thereby allowing clinicians to intervene much earlier. Although Wilfong acknowledged that most clinicians instinctively know when a patient’s prognosis is poor, the realities of a busy clinical practice can cause physicians to “lose sight” of this crucial need for intervention.
The AI prediction tool serves as a mechanism to prompt the clinician, urging them to pause and “think about this patient a little bit differently”, Wilfong stated. Having a system that more objectively indicates that a patient would no longer benefit fromtreatment facilitates the necessary but difficult conversations with the patient, he explained, adding that this process is considered vital to improving outcomes by ensuring that clinicians and patients are aligned on reality and the path forward. Wilfong stressed that the foundational principle for improving outcomes is aligning treatment with patients goals and values, something that cannot be achieved without a deep understanding of what those goals truly are.