What initially drew me to clinical intelligence was the realization that healthcare generates an enormous amount of valuable patient data, yet much of that information has historically been difficult to use effectively at the bedside to optimize clinical decision making. In critical care, clinicians make high-stakes decisions every minute, but when reviewing data, they’re seeing gaps in time between what data is charted and the moment in time when they are assessing the patient’s condition. As a result, they have to manually analyze and integrate data from different systems while managing multiple critically ill patients simultaneously. The result can be an imperfect picture of the patient’s condition, with this imperfect picture driving key decisions concerning the escalation or de-escalation of critical care.
Over the years, my perspective has evolved from viewing clinical intelligence as a way to organize information to seeing it as a foundational layer for improving decision-making. The challenge is transforming continuous, complex physiologic information into meaningful clinical insights that help teams review risk-related trends in a timely manner, understand patient trajectories more clearly and make more informed decisions.
Today, I see clinical intelligence as a means of creating a shared understanding of a patient's condition across caregivers, shifts and disciplines. When everyone is working from the same longitudinal view of patient risk and recovery, it helps reduce variability, improve communication and support more consistent care without replacing clinician expertise.
Connecting Fragmented Patient Data at the Bedside
The challenge isn’t necessarily the existence of data, but the organization and analysis of it. With access to more data than ever, the challenge is deriving actionable insights that can be implemented based on that data. Most ICU data are stored in separate systems, such as bedside monitors, ventilators, infusion pumps, laboratory systems and electronic health records (EHRs). ICU data can also pop up in isolation as static numbers without any fluidity, context or awareness of trend. Traditional monitoring systems are effective at showing what’s happening right now, but they’re less effective at helping clinicians understand how a patient’s condition is evolving over hours or days. Physiologic deterioration often develops as a pattern over time, rather than a single abnormal value, which can be missed without that longitudinal context.
Relying on snapshots, thresholds and alarms that may not fully reflect a patient’s trajectory can lead to late escalation or de-escalation in the case of longitudinal improvements. These challenges can be addressed, however, with continuous insights that show how variables interact over time and contextualize those data according to a hospital’s own guidelines and practices, allowing clinicians to feel better equipped to identify emerging risks, prioritize interventions and make more confident decisions.
Turning Data into Actionable Clinical Insights
At Etiometry, we believe effective clinical intelligence requires more than just aggregating data. We’re the first to use AI to deliver advanced analytics and actionable clinical intelligence at the bedside. These AI-driven analytics, combined with our clinical intelligence, enable trending of patient data against hospital-defined criteria to support clinician review of patient trajectory, including conditions such as cardiogenic shock, ARDS, respiratory deterioration and hemodynamic instability. Optional informational notifications (not alarms) may indicate when configured criteria are met, while all escalation, de-escalation and treatment decisions remain with the clinician. This approach has driven meaningful improvements in patient outcomes and financial performance, helping top hospitals across the U.S. and internationally optimize care delivery and improve patient flow.
What makes this approach valuable is the combination of standardization and personalization. Clinical pathway automation helps ensure that evidence-based practices are applied consistently. At the same time, the platform continually analyzes each patient's unique physiologic response, enabling care teams to tailor decisions based on individual circumstances rather than relying solely on generalized protocols.
This combination helps reduce unwarranted variation while preserving the flexibility clinicians need to manage highly complex patients. The result is a more proactive and data-informed approach to care that supports both quality improvement and individualized decision-making.
We do not view clinical intelligence as a substitute for clinical expertise. Rather, it serves as an additional layer of situational awareness that helps clinicians make better-informed decisions to ultimately improve patient care and recovery.
That balance is absolutely critical in any hospital setting. We do not view clinical intelligence as a substitute for clinical expertise. Rather, it serves as an additional layer of situational awareness that helps clinicians make better-informed decisions.
Healthcare is inherently complex and no algorithm can fully account for every clinical nuance. Clinicians bring invaluable experience, judgment, context and patient-specific knowledge which remain essential to providing excellent care. Tools like the Etiometry platform offer guidance that helps teams deliver the right personalized care at the right time, transforming how clinicians treat life-threatening conditions to improve patient care and outcomes and reduce healthcare costs.
Using Longitudinal Intelligence to Improve Patient Care
One area where we have seen a significant impact is supporting decisions around de-escalation of care in pediatric and cardiac intensive care environments. In the Boston Children’s ICUs, this longitudinal visibility has fundamentally altered how the team thinks about de-escalation of care. Bedside clinicians use continuous risk indicators as secondary measures to cross-check readiness for extubation and for weaning vasoactive medications. Charge nurses and team leaders review unit-level views to assess patient trends and bring attention to patients who may warrant additional clinician review because of the trend in their clinical profiles.
Clinically, these insights do not dictate action. They prompt questions. They elevate discussion. They help teams align around a shared, data-driven understanding of where a patient is along their recovery trajectory, which then helps the team determine if and what actions are needed to improve recovery.
For critically ill pediatric patients, the transition off of mechanical ventilation and vasoactive support represents one of the most fragile phases of recovery. Decisions around when to wean or extubate balance the risks of acting too soon against the harms of unnecessary delay and even small misjudgments can carry significant clinical consequences. Time-aligned physiologic analytics can provide an objective, data-rich and continuous view of respiratory-status trends to support clinician assessment of weaning and extubation readiness.
“We’ve seen firsthand how AI-driven clinical intelligence supports clinical judgment, rather than replacing it,” said Dr. Josh Salvin of Boston Children’s. “Instead of acting as an ‘if-then’ directive or an automated decision-maker, the Etiometry Platform aggregates continuous, high-fidelity physiologic data into interpretable trajectories. It allows clinicians to see not just isolated values, but patterns over time: how oxygen delivery, ventilation, perfusion and metabolic demand interact as a child recovers.”
Addressing Cardiogenic Shock through Earlier Detection
Identifying and supporting cardiogenic shock intervention has been a large focus area for the Etiometry team. Recent estimates suggest 50 percent of cardiogenic shock cases go undocumented, despite being one of the most lethal conditions in cardiology. And for patients suffering from cardiogenic shock, their mortality risk increases by 11 percent for each hour that goes by without intervention.
We recently introduced the first FDA-cleared Cardiogenic Shock Tool in the market, which applies hospital-defined SCAI criteria. The tool applies Society for Cardiovascular Angiography & Interventions (SCAI) guidelines to support clinician review of cardiogenic shock classification and stage progression over time, along with more complete, accurate documentation of the cardiogenic shock condition.
The tool is an innovative support software for overburdened care teams and for patients at risk for cardiogenic shock. With continuous visibility into shock progression alongside intelligent respiratory and hemodynamic trajectory tracking, care teams see a more comprehensive and more timely picture for managing some of the most complex patients. By providing time-aligned views of hospital-defined stage criteria and physiologic trends, the tool supports timely clinician review, team communication and enhances documentation of the patient’s condition to support appropriate coding and billing. Clinicians remain responsible for all diagnosis and treatment decisions.
Looking ahead, our team will continue to enhance the Etiometry Platform to support the identification and severity tracking of a broader range of hemodynamic and respiratory complications. This includes functionality in development to help clinicians identify conditions such as right ventricular dysfunction following cardiogenic shock, while expanding the platform’s ability to support timely recognition and tracking of other complex clinical conditions. These advancements build on Etiometry’s broader goal of improving patient and operational outcomes, including reduced length of stay, fewer ICU readmissions, shorter durations of critical therapies and better documentation of the severity of patients’ conditions. To date, use of the platform has been associated with a 36 percent reduction in ICU length of stay and a 41 percent decrease in ICU readmissions, as well as approximately $2 million in annual savings-more than $20,000 per ICU bed each year.
