Journal of the Association for Information Systems (2025)
Toward Triadic Delegation: How Agentic IS Artifacts Affect the Patient-Doctor Relationship in Healthcare
Pascal Fechner, Luis Lämmermann, Jannik Lockl, Maximilian Röglinger, Nils Urbach
This study investigates how autonomous information systems (agentic IS artifacts) are transforming the traditional two-way relationship between patients and doctors into a three-way, or triadic, relationship. Using an in-depth case study of an AI-powered health companion for managing neurogenic lower urinary tract dysfunction, the paper analyzes the new dynamics, roles, and interactions that emerge when an intelligent technology becomes an active participant in healthcare delivery.
Problem
With the rise of artificial intelligence in medicine, autonomous systems are no longer just passive tools but active agents in patient care. This shift challenges the conventional patient-doctor dynamic, yet existing theories are ill-equipped to explain the complexities of this new three-part relationship. This research addresses the gap in understanding how these AI agents redefine roles, interactions, and potential conflicts in patient-centric healthcare.
Outcome
- The introduction of an AI agent transforms the dyadic patient-doctor relationship into a triadic one, often with the AI acting as a central intermediary. - The AI's capabilities create 'attribute interference,' where responsibilities and knowledge overlap between the patient, doctor, and AI, introducing new complexities. - New 'triadic delegation choices' emerge, allowing tasks to be delegated to the doctor, the AI, or both, based on factors like task complexity and emotional context. - The study identifies novel conflicts arising from this triad, including human concerns over losing control (autonomy conflicts), new information imbalances, and the blurring of traditional medical roles.
Host: Welcome to A.I.S. Insights, the podcast at the intersection of business and technology, powered by Living Knowledge. I’m your host, Anna Ivy Summers. Host: Today, we're diving into a fascinating new study titled, "Toward Triadic Delegation: How Agentic IS Artifacts Affect the Patient-Doctor Relationship in Healthcare." Host: With me is our expert analyst, Alex Ian Sutherland. Alex, welcome. Expert: Great to be here, Anna. Host: So, this study sounds quite specific, but it has broad implications. In a nutshell, what is it about? Expert: It’s about how smart, autonomous AI systems are fundamentally changing the traditional two-way relationship between a professional and their client—in this case, a doctor and a patient—by turning it into a three-way relationship. Host: A three-way relationship? You mean Patient, Doctor, and... AI? Expert: Exactly. The AI is no longer just a passive tool; it’s an active participant, an agent, in the process. This study looks at the new dynamics, roles, and interactions that emerge from this triad. Host: That brings us to the big problem this research is tackling. Why is this shift from a two-way to a three-way relationship such a big deal? Expert: Well, the classic patient-doctor dynamic is built on direct communication and trust. But as AI becomes more capable, it starts taking on tasks, making suggestions, and even acting on its own. Host: It's doing more than just showing data on a screen. Expert: Precisely. It's becoming an agent. The problem is, our existing models for how we work and interact don't account for this third, non-human agent in the room. This creates a gap in understanding how roles are redefined and where new conflicts might arise. Host: How did the researchers actually study this? What was their approach? Expert: They conducted a very detailed, in-depth case study. They focused on a specific piece of technology: an AI-powered health companion designed to help patients manage a complex bladder condition. Host: So, a real-world application. Expert: Yes. It involved a wearable sensor and a smartphone app that monitors the patient's condition and provides real-time guidance. The researchers closely observed the interactions between patients, their doctors, and this new AI agent to see how the relationship changed over time. Host: Let’s get into those changes. What were the key findings from the study? Expert: The first major finding is that the AI almost always becomes a central intermediary. Communication that was once directly between the patient and doctor now often flows through the AI. Host: So the AI is like a new go-between? Expert: In many ways, yes. The second finding, which is really interesting, is something they call 'attribute interference'. Host: That sounds a bit technical. What does it mean for us? Expert: It just means that the responsibilities and even the knowledge start to overlap. For instance, both the doctor and the AI can analyze patient data to spot a potential infection. This creates confusion: Who is responsible? Who should the patient listen to? Host: I can see how that would get complicated. What else did they find? Expert: They found that new 'triadic delegation choices' emerge. Patients and doctors now have to decide which tasks to give to the human and which to the AI. Host: Can you give an example? Expert: Absolutely. A routine task, like logging data 24/7, is perfect for the AI. But delivering a difficult diagnosis—a task with a high emotional context—is still delegated to the doctor. The choice depends on the task's complexity and emotional weight. Host: And I imagine this new setup isn't without its challenges. Did the study identify any new conflicts? Expert: It did. The most common were 'autonomy conflicts'—basically, a fear from both patients and doctors of losing control to the AI. There were also new information imbalances and a blurring of the lines around traditional medical roles. Host: This is the crucial part for our listeners, Alex. Why does this matter for business leaders, even those outside of healthcare? Expert: Because this isn't just a healthcare phenomenon. Anywhere you introduce an advanced AI to mediate between your employees and your customers, or even between different teams, you are creating this same triadic relationship. Host: So a customer service chatbot that works with both a customer and a human agent would be an example. Expert: A perfect example. The key business takeaway is that you can't design these systems as simple tools. You have to design them as teammates. This means clearly defining the AI's role, its responsibilities, and its boundaries. Host: It's about proactive management of that new relationship. Expert: Exactly. Businesses need to anticipate 'attribute interference'. If an AI sales assistant can draft proposals, you need to clarify how that affects the role of your human sales team. Who has the final say? How do they collaborate? Host: So clarity is key. Expert: Clarity and trust. The study showed that conflicts arise from ambiguity. For businesses, this means being transparent about what the AI does and how it makes decisions. You have to build trust not just between the human and the AI, but between all three agents in the new triad. Host: Fascinating stuff. So, to summarize, as AI becomes more autonomous, it’s not just a tool, but a third agent in professional relationships. Expert: That's the big idea. It turns a simple line into a triangle, creating new pathways for communication and delegation, but also new potential points of conflict. Host: And for businesses, the challenge is to manage that triangle by designing for collaboration, clarifying roles, and intentionally building trust between all parties—human and machine. Host: Alex, thank you so much for breaking this down for us. This gives us a lot to think about. Expert: My pleasure, Anna. Host: And thank you to our listeners for tuning into A.I.S. Insights. Join us next time as we continue to explore the future of business and technology.
Agentic IS Artifacts, Delegation, Patient-Doctor Relationship, Personalized Healthcare, Triadic Delegation, Healthcare AI