IPEG 2023: Setting a Foundation for Healthcare Transformation

I recently had the pleasure of attending the IPEG annual meeting in Sorrento, Italy. The collegiality of the surgeons attending the conference was fantastic, and the views of Sorrento and the Island of Capri were stunning. 

With smaller accessible markets (due to smaller patient volumes and higher incidence of rare disease), specialty areas like pediatrics have been overlooked and under-addressed by industry and health tech. Whereas a normal adult hospital network sees tens of thousands of patients annually, a normal children’s hospital cares for fewer patients, over longer periods of time. Most industry partners can’t justify R&D costs without huge revenue forecast. However, we are witnessing the AI revolution. I have been fortunate to work with some companies like Caresyntax, who view these smaller markets as important development areas for context-specific, data-informed algorithms. The apps can be developed and scaled at a fraction of the R&D cost of normal industry. As AI and machine learning come to healthcare, Caresyntax and its industry partners aim to bring practice-shaping capabilities to smaller, underserved markets. And, after a week with the IPEG team, it seems pediatrics might be the perfect starting point because of the open, collaborative nature of the specialty. 

Collegiality hasn’t always been our strength 

As a young surgeon in training in the late 1980s and early 1990s, I witnessed too many examples of poor surgeon behavior, especially toward laparoscopic surgeon pioneers who were teaching and promoting the new minimally invasive surgical approaches. I witnessed anger, accusations, and belittlement. Years later, one of those pioneers confided to me that he received death threats from another surgeon.  

Another common theme of the time was that surgeons felt that they owned a patient, and if one of “their” patients was seen in the clinic of another “competitor” surgeon, they felt that that surgeon was trying to “steal” their patient. This kind of thinking and behavior should never be a part of healthcare, the one industry that is supposed to be about learning how to best care for each other as human beings. 

IPEG reflects a new standard in collaboration 

At the IPEG meeting, it was clear that culture and language barriers did not prevent meaningful joy and collaboration. I was especially pleased that a surgeon from Russia attended the meeting and gave a presentation. Having done my first live laparoscopic international procedure in Moscow in 1994, I know that the healthcare providers in all other countries, regardless of the political environment, including participating in a war with another country, do not support the divisiveness and harm to other human beings perpetrated by a small minority of political leaders.  

“The collegiality and collaboration I witnessed at the IPEG meeting ensures an environment of psychological safety, where speaking up with opinions that might challenge the status quo is met with curiosity and open-mindedness rather than animosity.” 

The collegiality and collaboration I witnessed at the IPEG meeting ensures an environment of psychological safety, where speaking up with opinions that might challenge the status quo is met with curiosity and open-mindedness rather than animosity. Amy Edmonson, the Novartis Professor of Leadership and Management at Harvard Business School, coined the term “psychological safety” in a 1999 journal article exploring its relationship to team learning and performance. I had the opportunity to attend a two-day seminar led by Dr. Edmonson, sponsored by Athenahealth. It was eye-opening how much research has been done to reveal the critical need for an environment of psychological safety for organizations to thrive. It is still extremely rare in organizations in the 21st century, especially in healthcare, where shame and blame are common.   

Despite the wonderful psychologically safe and collaborative environment of IPEG, what if that is not enough to move the needle to evolve to a sustainable healthcare system?  

There is no such thing as an average patient 

The program included traditional presentations of treatments and outcomes from traditional clinical studies intended to produce generalizable results for all patients, representing the average patient. But patients are not average, and one size does not fit all. 

This concept hit home for me when I read a story about the early days of the US (United States) Air Force in a book called The End of Average by Todd Rose, Director of Mind, Brain, and Education at Harvard’s Graduate School of Education. In the early days of the US Air Force, planes crashed at an alarming rate. At first, no one could figure out why. Lt. Gilbert S. Daniels, who had recently graduated from Harvard, where he studied Anthropology, evaluated human hand shapes and sizes from hundreds of male Harvard students. With that experience, he proceeded to measure over 140 physical dimensions of all 4,063 pilots in 1950.  

What he found was surprising; none of the pilots were in the average range. He concluded that there were no average-sized pilots and no pilot fit well into the cockpit. Some pilots fit so poorly into the cockpit that they couldn’t fly the plane safely, leading to most plane crashes in the early decades of the US Air Force. This discovery resulted in adjustable size accommodations inside cockpits and limitations on the size of pilots. 

Rethinking the one-size-fits-all approach to clinical research 

But this one-size-fits-all strategy is what we still have in healthcare today. We design a study with controls and inclusion/exclusion criteria and attempt to prove that one thing is generalizable to all patients everywhere. It doesn’t work, and it can’t. That is because we are complex biologic organisms, constantly changing, with uncontrollable biologic variability. This reality will not change.  

“This reality was highlighted in an excellent IPEG presentation by Thomas Inge, Surgeon-in-Chief and Chair of the Department of Surgery at the Northwestern University Feinberg School of Medicine on a nearly decade-long study on bariatric surgery results in teenagers … patients are not average…. In a deeper analysis, [Inge’s team] found four unique subpopulations.”  

This reality was highlighted in an excellent IPEG presentation by Thomas Inge, Surgeon-in-Chief and Chair of the Department of Surgery at the Northwestern University Feinberg School of Medicine. He presented a nearly decade-long study on bariatric surgery results in teenagers. The primary procedures were laparoscopic gastric bypass and sleeve gastrectomy. Both procedures produced similar outcomes, and nearly a decade after their weight loss surgery, the average patient had a significant weight loss of about 50% of their excess weight. But the interesting thing is that patients are not average, like the US Air Force pilots. In a deeper analysis, they found four unique subpopulations: one lost most of their excess weight, another lost just under half their excess weight, one lost a small percentage of their excess weight, and another group actually weighed more than they did at the time of the initial operation.  

Why reductionist scientific method is ill-equipped for complex disease such as obesity 

I first learned about the reality that the same treatment would lead to different outcomes for different patient subpopulations when I began to study the interaction between hernia mesh and the human body. One science historian, Henry Cowles, has recently published a book and an article that explains why the reductionist scientific method is ill-equipped to deal with a complex disease such as obesity.  

The way we currently approach research and data in healthcare, based on principles of reductionist science, can never identify the patient subpopulations identified in the bariatric study presented by Dr. Inge. Only the principles of systems and data science can generate high-quality decentralized data and predictive algorithms that can match the best value treatments with appropriate patient subpopulations. Netflix does this for us, using decentralized data and networked algorithms to promote a set of movies and shows to clusters of people based on “taste clusters.” In healthcare, we must do the same to match the best value treatments with the appropriate patient clusters. 

The collaborative environment of pediatric medicine is a case study for US healthcare  

This is where we come back to the collaborative environment at IPEG as the foundation for a solution for our global healthcare system. Multiple tech prize competitions (like Kaggle and the Netflix Prize) have shown that combined algorithms have better accuracy and predictive potential over single, isolated algorithms. Intuitively, this makes sense, but when it comes to healthcare, this becomes even more important given the complexities of patient demographics, training, environmental factors, etc.  The winners of a data science competition called Kaggle are teams that regularly use ensemble learning by combining their algorithms. For example, one contest was called the Netflix Prize. The winner had to improve a Netflix algorithm by 10%. It took three years, but the winners figured out that combining their algorithms would likely be more successful than attempting to continue to improve their single algorithm. In September 2008, the team from AT&T Research (Bellkor) reached out to a small start-up team from Austria (Big Chaos) and formed an alliance, Bellkor in Big Chaos. Although this collaboration led to winning the annual progress prize of $50,000, they didn’t achieve the Grand Prize. So, this collaborative team reached out to another team, Pragmatic Theory, and became a team of three teams, Bellkor’s Pragmatic Chaos. On June 26, 2009, they broke through the 10% barrier and won the $1 million Grand Prize. The second-place team achieved the 10% barrier only ten minutes later. They were also a team of teams called “The Ensemble.” 

If we do not collaborate to solve complex problems that we see in healthcare, the goal of a sustainable healthcare system will continue to be elusive. 

There are significant benefits of this model for learning. By combining the learning from many small teams and networking that knowledge (ensemble learning), the complexity is reduced. And because the raw data remains in each local environment, the privacy and security of the data is maintained.  

“For healthcare to be sustainable, we will need to learn and apply the principles of systems and data science. We will also need to change our thinking and break free from our lower-brain mindset that craves certainty and control.” 

For healthcare to be sustainable, we will need to learn and apply the principles of systems and data science. We will also need to change our thinking and break free from our lower-brain mindset that craves certainty and control. The concepts of certainty and control are harmful illusions. The higher brain can overcome this reductionist thinking and allow our human potential for empathy, creativity, discovery, and innovation, foundations of psychological safety, to be unleashed. The concept of one static “best practice” is an oxymoron. We are all in the global healthcare system together, and we need each other, learning together and networking those learnings and algorithms if we want a sustainable system to emerge. Who doesn’t want this? Why wouldn’t they? 

“IPEG is known for not being uncomfortable challenging the status quo, so it might just be the right surgical society to embrace a new way of thinking.”   

I think the collegiality and the psychologically safe environment that IPEG has created is a perfect foundation for implementing systems and data science principles and the Ensemble Model for learning. They will only need to challenge the traditional thinking that reductionist science principles and the prospective clinical trial are the way to improve patient outcomes. But IPEG has always been comfortable challenging the status quo, so it might just be the right surgical society to embrace a new way of thinking.  They could set an example in applying principles of systems and data science for a sustainable healthcare system. 

Pediatrics is next to leverage AI and healthcare technology 

In the American healthcare research and development (R&D) ecosystem, due to smaller patient volumes and higher incidence of rare disease, specialty areas such as pediatrics have been overlooked and under-addressed by industry and health tech due to smaller patient volumes and higher incidence of rare disease. Whereas a normal adult hospital network sees tens of thousands of patients annually, a normal children’s hospital cares for fewer patients, over longer periods of time. Most industry partners can’t justify R&D costs without a strong financial incentive.

However, the AI revolution is coming to pediatrics. I have been fortunate to work with some companies such as Caresyntax, that offer surgical intelligence and services, who view these smaller markets as important development areas for context-specific, data-informed algorithms. The apps can be developed and scaled at a fraction of the R&D cost of normal industry. As AI and machine learning come to healthcare, Caresyntax and its industry partners aim to bring practice-shaping capabilities to smaller, underserved markets. And, after a week with the IPEG team, it seems pediatrics might be the perfect starting point because of the open, collaborative nature of the specialty.