From Surgery to Recovery: Dr. Christian Péan on the Future of Orthopedic Care with AI
Preventing post-operative falls and avoidable readmissions is a growing priority in orthopedic care, with new technologies offering promising solutions. Caspar sat down with Dr. Christian Péan, Assistant Professor of Orthopedic Surgery at Duke University School of Medicine, to discuss how continuous, non-intrusive monitoring, Generative AI, and predictive analytics can improve recovery, enhance care coordination, and help hospitals succeed in a value-based care world.
Q: Post-operative falls are a major concern in orthopedic recovery. What do you think are the biggest missed opportunities in preventing these events—and how could continuous, non-intrusive monitoring help?
Post-operative falls are a major concern in the inpatient setting and are something that is tracked as a quality reporting index. Preventing them is critical for both patient safety and the hospital’s bottom line. Non-intrusive monitoring can ease the burden on nursing staff by providing continuous data—not just if a fall has occurred, but if a patient is at risk—so proactive steps can be taken before an event happens.
Q: How significant a role do post-operative issues play in hospital readmissions—and how might continuous, passive insights into patient behaviors (like falls, mobility patterns, or vital signs) help reduce avoidable readmissions?
Avoidable readmissions are a huge issue for hospital and health systems for a few reasons. One, in value-based care arrangements, we’re accountable for the spend and that includes readmissions. Two, they’re taking up bed space that could be used to take care of other patients. And three, no patient wants to get readmitted to the hospital after they have surgery. One of the big ways that we can prevent readmissions is by having really good asynchronous post-operative outreach and monitoring.
Continuous monitoring of a patient’s mobility at home can provide crucial insights. For example, a stalled mobility curve might signal a potential issue. Frequent nighttime movements, especially bathroom trips, can indicate poor sleep quality and may be a leading indicator for a UTI, which is the most common complication after hip fractures – and a frequent contributor to readmission. By identifying these patterns early, we can intervene to prevent avoidable readmissions.
Q: At RevelAi Health, you emphasize conversational AI. What does that look like in practice, especially when it comes to coordinating patient care after surgery?
RevelAI Health is a platform that not just helps coordinate care but orchestrates it as well. Our conversational AI agent provides patient education at different intervals after their surgery or after they’ve been diagnosed with a condition. It’s designed to educate them about their condition and also help prevent issues that can negatively impact recovery.
The other component that conversational AI has made possible is this ability to orchestrate care bidirectionally. We can now have a continuous line of communication with patients by deploying an AI agent to take that communication, contextualize it, and allow for staff members to easily act on it. It’s important for care team members and clinicians to be looped into the patients’ progress and recovery.
Q: What are some ways you see machine learning improving risk stratification before orthopedic procedures—and how might AI-driven long-term trend analysis help improve post-op monitoring, particularly during the transition from hospital to home or long-term care?
Some of my research has actually focused on utilizing machine learning algorithms to risk stratify patients to predict their outcomes. And we know that these tools can be tremendously helpful if they’re implemented into the workflow. Understanding where someone is on their recovery curve from a mobility perspective – as well as how they’re engaging with the care team – can really help us inform the kinds of resources we commit to a patient.
Specifically, long-term trend analysis can be extremely helpful for splitting patients up into buckets of red, yellow, or green. This allows the team to focus their attention on those patients who need it most as opposed to those who are recovering well on their own. In this way, machine learning helps us dedicate resources in the smartest way possible.
Q: With the vast amounts of data available to both researchers such as yourself and companies like Caspar, how do you see Generative AI and large behavioral models reshaping the face of healthcare over the next 5 to 10 years?
The best thing about behavioral models and Generative AI is that these models allow the clinicians and care team members to receive context about data in a way that they understand it. There are many situations where we have dumb data – we get a lot of dashboards, but not a lot of actionable insights.
With these evolving models, we have the opportunity to take that data, contextualize it for clinicians, and then even coach clinicians or care team members on how to act on it. We finally have insights that are paired with the data. And I’ve never seen a technology that has the ability to bridge that gap between data and action until now.
I think that in the next few years, companies like Caspar will start combining advanced tools for tracking a person’s recovery or mobility with the kinds of health data we’ve traditionally collected in other ways. It’s a really innovative time, and we’ve never seen more policy momentum to implement tools like this. A lot of the new mandatory bundles in episode-based and ambulatory specialty care really emphasize continuous patient monitoring. Being able to do that is going to be critical—not just for better patient outcomes, but also for the hospital’s bottom line.
About Dr. Christian Péan
Dr. Christian Péan is an orthopedic trauma surgeon at Duke University School of Medicine, specializing in fracture care, arthroplasty, and complex hip and shoulder procedures. His research focuses on using machine learning and AI to predict surgical risk and improve recovery outcomes. As Co-founder of RevelAi Health, he is advancing equity-conscious, AI-driven approaches to Value-Based Care in orthopedics and musculoskeletal health.