Dr. Greg Latham

How Seattle Children’s Bellevue Surgery Center employed AI-powered technology to eliminate opioids for most pediatric surgeries

May 27, 2020
By Greg Latham, MD

Long before COVID-19 threatened to overwhelm our healthcare system, the United States was dealing with another health crisis—the opioid epidemic. In 2015, something occurred that had not happened in over a century: life expectancy in the U.S. entered a period of sustained decline. According to the World Bank Group, the country’s average life expectancy fell from 78.8 years in 2014 to 78.7 years in 2015, and then to 78.5 years in 2016 and 2017. One of the major reasons for this drop was a surge in drug overdoses and suicides, both linked to opioid misuse. Between 1999 and 2014, drug overdoses nearly tripled, and almost 450,000 people died from an opioid overdose. In 2017, the U.S. Department of Health and Human Services declared the opioid crisis to be a public emergency.

Despite this recognition, opioid use is increasing, and research shows that 7% of adult surgical patients and 5% of adolescent patients develop persistent opioid use following surgery. Given that there are over 50 million surgical procedures performed annually, surgery represents a major gateway to opioid misuse, potentially causing harm to more than 2 million surgical patients each year.

This epidemic, coupled with an opioid shortage, drove anesthesiologists at Seattle Children’s Bellevue Clinic and Surgery Center (BCSC) to find ways to address this issue. We worked with medical leadership to assemble an interdisciplinary project team focused on reducing the use of opioids in our practice. At the same time, we sought to add value to our clinical services by maintaining or improving surgery outcomes while reducing costs.

In January 2018, our quality improvement project began at BCSC. As a first step, our project team focused on reducing the use of opioids in pediatric tonsillectomies, a painful surgery in which pain management includes opioids. Using MDmetrix, an AI-powered analytics platform, our physician group analyzed more than 10,700 surgical cases across many different clinical protocols and approaches, reevaluating medications that had been long been considered “standard practice.” We also reexamined older studies that might be used as a basis to develop opioid-free anesthesia protocols.

Based on our analyses of published research, we devised a series of pain management protocols. In accordance with best improvement practices from the Institute for Healthcare Improvement (IHI), we used standard methodology for the development of clinical improvements using real-world data, namely the Plan-Do-Study-Act (PDSA) cycle.

Typically, it takes years to complete a clinical PDSA cycle. However, by employing the new AI-based technology, our team was able to complete improvement cycles in just 12 weeks. Utilizing real-world data routinely collected in electronic medical records (EMRs), our team monitored and evaluated the comparative effectiveness of current and historical protocols on an ongoing basis. As a result, our team was able to quickly adapt our protocols, incorporating evidence-based best practices to effectively manage surgical pain while reducing patients’ opioid exposure.

Following this initial success, the BCSC’s clinical team expanded our approach to include other surgeries across a range of disciplines, including orthopedics, urology, plastics, dental, ophthalmology and general surgery. The results were groundbreaking. We developed protocols that eliminated the use of opioids during surgery from the vast majority of our outpatient procedures (dropping from 84% to 8% of procedures), while also reducing the need for postoperative morphine administration from 11% to just 6%. To date, over 6,000 patients have now successfully undergone surgery at BCSC without opioids.

With these new opioid-limiting protocols, we found that post-surgery nausea and vomiting, caused by opioids, virtually disappeared. In addition, pain scores, total anesthesia time, and total post-anesthesia care unit (PACU) time remained stable. Overall, costs also decreased (85% reduction in cost for analgesic medications), while patient/parent satisfaction increased dramatically.

This quality improvement initiative powerfully demonstrates how the healthcare industry can leverage digitized data and AI-based technology to improve patient care, patient safety, and operational efficiency.

About the author: Dr. Latham is a pediatric anesthesiologist and director of pediatric cardiac anesthesiology at Seattle Children's Hospital and has authored many publications on advances in anesthesiology and healthcare improvement. He also serves as an associate professor of Anesthesiology and Pain Medicine at University of Washington. Dr. Latham earned his MD at the University of Kansas.