The Rise of Systems of Intelligence: From Data to Action in Healthcare - Part 1

In the world of healthcare, we're on the brink of a revolution. Systems of Intelligence (SoI) are poised to redefine the very fabric of our industry, transforming how we collect, analyze, and act on data. This isn't just an upgrade; it's a transformation that will change lives, save lives, and push the boundaries of what’s possible.

Across the globe, leading companies like Microsoft, Apple, and NVIDIA are recognizing the immense potential of Systems of Intelligence. These tech giants are not only investing heavily in AI but also envisioning a future where SoI drives innovation across various sectors, including healthcare. 

At Apple's recent WWDC24 event, the emphasis on Apple Intelligence showcased how deeply integrated AI and machine learning are becoming in everyday technology. This is a huge testament to the transformative power of Systems of Intelligence and signals what the future holds.

Beginnings are delicate times, laying the foundation for everything that follows. The healthcare industry had its own beginnings.

Picture this: the early 2000s. Healthcare data had transitioned to electronic formats, yet it was akin to a jigsaw puzzle with pieces scattered across various systems. Electronic Health Records (EHRs) were the cornerstone of Systems of Record (SoR) – digital repositories that stored vast amounts of patient data. While EHRs brought significant improvements over paper records, they were largely passive, documenting patient histories without offering deeper insights. But just like everything, recurring action has compounded impact accumulates, and so does data. 

If a physician would diagnose healthcare, they would diagnose it with paralysis. 

Analysis Paralysis.

No civilization in the history of humanity has documented so well. Some could say documented so well so nothing, but that’s debatable. In the context of healthcare, fast forward to today. Healthcare generates a staggering 2.3 billion gigabytes of data every year. From EHRs to wearable devices and advanced imaging techniques, the data deluge is real. Yet, 80% of this data remains unstructured, a vast, untapped potential reservoir.

Systems of Intelligence are not just a distant dream; they are becoming a tangible reality reshaping the healthcare landscape.

SoI doesn't just store information but learns from it. Imagine a healthcare system that doesn't just store information but learns from it. A system that can predict a heart attack before it happens, optimize treatment plans in real-time, and even suggest preventive measures tailored to each individual.

Current approaches are effective, but they often work in isolation. This is where the opportunity for a more unified and impactful solution lies.

Take, for example, the story of Sepsis. Sepsis, a life-threatening response to infection, claims hundreds of thousands of lives each year. Early detection is crucial, but the signs are often missed until it's too late. With SoI, hospitals like Kaiser Permanente have implemented predictive algorithms that analyze patient data in real-time, flagging early signs of sepsis hours before clinical symptoms appear. This proactive approach has saved countless lives and revolutionized emergency care.

Actionable insights at the right time to the right staff have to be the way forward. And it is another possibility with SoIs.

Actionable insights delivered to the right staff at the right time are the cornerstone of SoIs. Imagine an emergency room where data flows seamlessly, ensuring that doctors and nurses receive critical information precisely when they need it.

Doctors deserve better than endless, repetitive documentation that adds no value to patient care. EHRs often bear the blame for this frustration. Too much pajama time spent documenting means less time determining the best treatment options for patients. This isn't good for anyone. 

But beyond immediate care, SoIs can also address the persistent issue of coding gaps. Real-time analytics can identify discrepancies in patient records, ensuring that every diagnosis and procedure is accurately documented. This not only improves patient care but also ensures that healthcare providers are reimbursed correctly for their services.

The story of Franciscan health is fascinating here. Franciscan Health achieved a 2.8% improvement in coding gap closure rate across populations. The automated coding processes helped Franciscan realize a total of $312,000 in value. Seamlessly integrated into the EHR’s clinical workflow for partners and affiliates, Innovaccer’s physician engagement solution helped simplify the process of identifying dropped HCC codes at the point of care through risk adjustment analytics by using historical claims and clinical data. It became easy for the Franciscan team to quickly maintain updated patient records with little to no coding gaps. This also helped care teams gain visibility into responses regarding any identified coding gap, such as “accepted,” “rejected,” or “pending.” 

It’s also about personalization at scale. 

By analyzing patient-specific data, SoIs can tailor treatments to individual patients' needs. This isn't just about better care; it's about revolutionizing the patient experience. For example, in managing chronic diseases, the need is to create personalized care protocols. These protocols can include tailored needs of the patient cohort which are adjusted in real-time based on the patient's current condition and historical data.

CHI Health’s care teams implemented automated and data-driven protocols for managing patient transitions and follow-ups. These new systems allowed the teams to create timely and specific care plans for patients leaving the hospital or emergency department. The protocols triggered automatic alerts for the designated care team members with patients' discharge information, eliminating the need for manual tracking. Between January 2022 and October 2022, CHI Health saved $2.75 million by using these automated systems to improve patient care and follow-up across various insurance groups including Medicare and Medicaid.

The road to fully realizing SoIs isn't without bumps

Data security is paramount, and ensuring the quality and accuracy of AI-generated data is critical. Integrating AI with existing systems requires robust data management and governance.

Data Security: Health systems must implement advanced cybersecurity measures, including encryption, access controls, and regular security audits. For instance, Mayo Clinic has invested heavily in cybersecurity infrastructure to protect patient data from breaches and unauthorized access.

Data Quality: Ensuring the accuracy of AI-generated data involves rigorous validation processes. At Geisinger Health System, AI algorithms are continually tested against real-world outcomes to ensure their recommendations are reliable and evidence-based.

Integration: Integrating AI with existing systems can be complex. Kaiser Permanente has developed a comprehensive data integration strategy that involves using standardized data formats and interoperable systems to ensure seamless data flow across different platforms.

However, the potential rewards far outweigh the challenges. The interpretability of AI insights is crucial, as healthcare providers must trust and understand these recommendations. Continuous collaboration between data scientists and healthcare professionals is essential.

At Partners HealthCare, data scientists work closely with clinicians to develop AI models. This collaboration ensures that the insights generated are clinically relevant and easily interpretable, fostering greater trust and adoption among healthcare providers.

The road ahead 

As leaders in healthcare, you are not just the custodians of technology within your organizations. You are the architects of the new healthcare landscape. The rise of Systems of Intelligence (SoIs) isn't merely an incremental improvement; it’s a revolutionary leap forward. It’s about reimagining the very essence of healthcare, from how we diagnose and treat illnesses to how we manage patient care and operational efficiency.

Imagine a world where every decision is informed by data-driven insights. Where the power of advanced analytics and artificial intelligence transforms mountains of raw data into actionable intelligence. This is a world where healthcare is not just reactive but proactive. The rise of Systems of Intelligence isn't just about technology; it's about reimagining what's possible in healthcare. It's about leveraging data to create a seamless, efficient, and joyful experience in care.

This is our moment. Let's seize it. Let’s bring back the joy in care.

If you'd like to discuss how fragmented approaches to Systems of Intelligence can be streamlined, I invite you to meet with Mike Sutten, Anil Jain, or me. You can also meet at HealthImpact Forum on 12th June 2024 where I’m speaking on how AI can be a tool to be the agent of joy in clinicians’ lives. Together, we can explore the possibilities and chart a path forward.