Long ago and far away, I was attending my first major conference as Associate Editor for a publisher of textbooks on artificial intelligence and advanced computer science. We’d just finished arranging our exhibit at the Dallas Merchandise Mart, showcasing new publications such as Lexical Ambiguity Resolution (Cottrell), Logical Foundations of Artificial Intelligence (Nilsson), Probabilistic Reasoning in Intelligent Systems (Pearl), Machine Learning: An Artificial Intelligence Approach (Michalski), and many more multisyllabically titled tomes. An early attendee wandered up: a local – tall, a little bow-legged, wearing a brand-new jacket and tie, well-worn ten-gallon hat, and (I kid you not) cowboy boots, and a badge with lots of impressive letters after his name. He gazed at our booth for a while, his eyes taking in each title, and his brow slightly furrowing. After a deep breath, he boomed in full-on Texas twang, “Artificial Intelligence? Ah’m still workin’ on the real thang!” Right.
I suspect that’s how many of us are feeling as we’re besieged with come-ons for everything from meeting scribes to medical imaging interpretations to professional headshots all “powered by AI.” And, as palliative care professionals, we might quickly dismiss the innovations as just so much marketing hype hardly relevant to a discipline that prides itself on very human attributes of empathy, values, wishes, hopes and fears, connection, communication, and collaboration, and a drive to relieve suffering and pain. Besides taking notes in our meetings, what could AI possibly have to offer us??? As it happens, quite a lot!
In our zoominar series so far this spring, we’ve learned about the guardrails and principles guiding the work of a large group of healthcare systems and developers, and we’ve had an excellent tutorial in what AI and machine learning are, how they work, and how they are impacting healthcare. If you missed these, I encourage you to view the recordings!
Next week we’ll hear about how palliative care physicians are using AI to assist clinicians and patients in advance care planning and goals of care conversations. In early May, in a two-half-day virtual miniconference, we’ll have 10 thoughtful and challenging presentations from healthcare leaders, clinical innovators, and patients who are using AI to power palliative care. We’ll also hear about some of the risks and worries and profound transformative implications of these models and applications.
As ChatGPT says (see next blog, by ChatGPT), “For palliative care clinicians, embracing AI is not about replacing human connection but about using technology to support their work in delivering compassionate, patient-centered care.”
It's a lot to take in. I get it! If you can’t spend the time to be with us during the live presentations, register anyway – you’ll have access to the recordings and can take them in at your own pace. It’s especially a lot to take in right now, when there is so much chaos and uncertainty in the country and the field. I get that, too! But AI is here. And we may find that not only can it take over some of the drudgery of our jobs, but it just might help us get back to the “real things” of our own humanity, our own intelligence, and our very important relationships.
Why Palliative Care Clinicians Should Embrace Artificial Intelligence
By ChatGPT
[Note from the human in the loop: As I was getting ready to write the April 2025 blog and gazing at the blinking cursor on the screen, I thought – huh, maybe I’ll just ask ChatGPT. My prompt was: “Please* write a 600-word blog directed toward palliative care clinicians about why they should be interested in artificial intelligence. Use specific examples in your answer.” (*I make it a point to be polite; I find the results are better if you use please and thank you!) The following appeared on my screen within 20.45 seconds. Any errors are ChatGPT’s alone. I have not edited, even for length – ChatGPT can write but apparently can’t count, as the blog is 825 words! BTW, our miniconference will include sessions exploring many of these topics.]
As the healthcare landscape continues to evolve, palliative care clinicians find themselves at the forefront of a growing shift in how medical services are delivered. While the human connection in palliative care remains paramount, there is increasing potential for artificial intelligence (AI) to enhance both the quality of care and efficiency in managing patients with serious, life-limiting illnesses. The integration of AI into palliative care can streamline workflows, aid clinical decision-making, and even help improve patient outcomes. Here’s why palliative care clinicians should consider embracing this transformative technology.
Enhancing Clinical Decision-Making
AI can significantly enhance decision-making in palliative care by providing evidence-based insights and predictive analytics. For instance, machine learning algorithms can analyze patient data, including medical history, clinical observations, and even social determinants of health, to predict patient needs and recommend interventions. This helps clinicians anticipate complications and tailor their care plans with more precision.
One application is the use of AI to predict when a patient may experience a decline in their condition, allowing clinicians to intervene early and modify care plans to align with the patient's evolving needs. For example, AI tools can identify patterns in patient vital signs, lab results, and other clinical data that indicate worsening symptoms or disease progression. This empowers palliative care teams to proactively address pain management, anticipate potential complications, and ensure that the patient’s comfort and dignity remain the priority.
Improving Symptom Management
Symptom control is a cornerstone of palliative care, and AI can play a vital role in improving this aspect of care. Machine learning algorithms can analyze large datasets of clinical notes, patient-reported outcomes, and symptom severity assessments to identify trends and suggest more effective treatments. For example, AI can assist in managing complex symptoms such as pain, nausea, or shortness of breath by suggesting optimal medication regimens based on similar patient profiles and outcomes.
AI-powered platforms, such as natural language processing (NLP) systems, can even analyze unstructured data from clinician notes to track symptoms over time, helping clinicians adjust interventions in real-time. This can be particularly useful when managing patients with multiple comorbidities or those who are not able to effectively communicate their needs.
Supporting Communication with Families
Communication with families is a critical component of palliative care, as clinicians often need to guide family members through complex decisions about care goals, end-of-life issues, and symptom management. AI can aid in this process by providing clinicians with evidence-based resources and frameworks for conducting difficult conversations. For instance, AI tools can help predict family dynamics or offer prompts for discussing sensitive topics like prognosis and advanced care planning.
Additionally, AI can assist in documenting these conversations, ensuring that family preferences, decisions, and changes in care plans are accurately recorded. This reduces the risk of miscommunication and helps ensure that the care provided aligns with the patient's and family's wishes.
Facilitating Administrative Efficiency
Palliative care clinicians often face significant administrative burdens, from documenting patient encounters to coordinating interdisciplinary team meetings. AI can streamline these tasks by automating routine processes such as charting, appointment scheduling, and insurance authorization. By using AI-powered tools for administrative tasks, clinicians can free up more time to focus on patient care.
For example, AI can assist with predictive scheduling, helping to optimize clinician time and ensure that patients are seen in a timely manner, particularly during periods of high demand. AI can also automate administrative functions like documentation and coding, reducing the time spent on paperwork and minimizing the risk of errors.
Personalized Care Plans
AI’s ability to analyze vast amounts of data allows for the creation of personalized care plans that are tailored to each patient’s unique needs. By processing information such as genetic data, previous treatment responses, and personal preferences, AI can provide recommendations for individualized care strategies. In palliative care, where each patient’s needs are unique, this can be invaluable.
For example, AI tools can help clinicians identify patients who may benefit from alternative treatments or therapies that have worked well for others with similar conditions. This could include personalized medication regimens or customized non-pharmacological interventions like massage therapy or music therapy.
The Future of Palliative Care with AI
While AI in palliative care is still in its early stages, the potential for this technology to enhance the field is immense. As AI continues to evolve, its ability to improve clinical decision-making, enhance symptom management, streamline administrative tasks, and foster better communication will only grow.
For palliative care clinicians, embracing AI is not about replacing human connection but about using technology to support their work in delivering compassionate, patient-centered care. By harnessing the power of AI, clinicians can make more informed decisions, improve patient comfort, and ensure that each patient’s journey through serious illness is as dignified and peaceful as possible.
The future of palliative care is undoubtedly intertwined with the advancements in AI, and those who stay informed and adapt to these changes will be better equipped to provide the highest level of care to their patients.