Mauston, Wisconsin-based Mile Bluff Medical Center clinicians faced numerous challenges due to the overwhelming amount of data they encountered daily. This data, found in various formats, including structured, unstructured and scanned documents, often made it difficult for clinicians to make timely and well-informed decisions regarding treatment plans.

THE PROBLEM

Critical information might be buried deep within a scanned document, leading to delays in accessing necessary details. The task of sifting through extensive electronic health records to find pertinent information contributed significantly to clinician burnout and cognitive fatigue.

The constant demand on clinicians' time and energy was exacerbated by the need to see more patients and manage their schedules efficiently, all while dealing with an aging population with increasingly complex medical conditions.

In small rural hospitals like Mile Bluff Medical Center, this issue was even more pronounced. Patients frequently needed referrals to specialists outside the organization, and their transfer records, discharge summaries and specialty consult notes often arrived in disparate formats. Extracting the necessary information from these large documents was time-consuming, further adding to the workload and cognitive burden on providers.

"The heavy provider workload and the complexity of navigating and searching through medical records created significant stress and was a contributing factor to burnout," said Randy Brandt, PA-C, a primary care physician assistant and Expanse Search and Summarization project team leader at Mile Bluff Medical Center. "Establishing care with new patients, who often brought large volumes of transfer records, required extensive time and effort for adequate review.

"This process could take hours or even days, causing delays in patient care, including delays in scheduling or starting appointments, and follow-up on medical conditions," he continued. "Additionally, patients were not always reliable sources of accurate medical history, medication details or treatment dates, further complicating the clinician's task of gathering comprehensive information."

Fragmented and disordered data, scattered across various sections of a medical record, demanded considerable effort to piece together for informed medical decision-making. In busy clinics, emergency departments or hospital settings, this process consumed time and energy that clinicians at Mile Bluff Medical Center simply could not afford.

The need for an efficient technology to manage and summarize this vast amount of data became imperative, paving the way for the adoption of the Meditech EHR Expanse AI-powered search and summarization technology, powered by Google Health, which ultimately aimed to alleviate these burdens and improve the overall efficiency and quality of patient care.

"We are now seeing our primary care physicians report an average savings of 7.5 minutes per patient on pre-visit prep," Brandt reported. "The same average time savings has been reported by the clinicians in our pre-operative clinics, and by those in our sleep medicine department."

PROPOSAL

At Mile Bluff Medical Center, addressing the inefficiencies and challenges in managing patient health records is crucial. The Expanse search and summarization technology offers a transformative solution to streamline and enhance the way clinicians interact with electronic health records, Brandt said.

"This technology aims to resolve the problem of fragmented data spread across various sources, including current EHR systems, legacy data, and scanned or faxed documents," he continued. "The core proposal of Expanse search and summarization is to harmonize disparate data, which comes in various structured, unstructured and even handwritten forms, to create a comprehensive longitudinal health record for each patient.

"This harmonized data is accessible through an advanced search function, allowing clinicians to quickly find relevant information," he added. "The technology offers a summarization feature that consolidates patient information into a comprehensive problem list. This list includes active medical issues, historical conditions, procedures, and frequently occurring signs or symptoms, all presented within seconds."

By integrating data from various sources into a unified view, Expanse search and summarization reduces the time clinicians spend clicking through charts and scrolling through records. The summarized data provides clinicians with substantiating evidence of where the information was found within the EHR, allowing for immediate review and additional context.

This feature not only enhances efficiency but also reduces the cognitive load on clinicians, helping to alleviate burnout and improve decision-making, Brandt noted.

"Moreover, the AI-powered technology significantly impacts both clinician and patient experiences," he continued. "For clinicians, it offers a more efficient workflow by surfacing critical information promptly, allowing them to spend more time engaging with patients rather than navigating the EHR.

"For patients, this translates to more attentive care, as providers can focus more on their needs rather than administrative tasks," he added. "The creation of a longitudinal health record ensures a more informed and engaged provider, leading to improved patient satisfaction and outcomes."

Expanse search and summarization also is designed to address care gaps that often occur during transitions, such as when a patient establishes care with a new provider, is discharged from the hospital or receives specialist care.

"By providing a comprehensive view of the patient's health history, the technology helps clinicians deliver high-quality care consistently, improving overall patient outcomes," Brandt said. "Importantly, this solution integrates seamlessly into existing workflows, enhancing the foundational functionality of the EHR without disrupting routine operations.

"Expanse search and summarization offers a robust solution to harmonize and streamline patient data management," he continued. "By improving the efficiency and accuracy of information retrieval, it supports clinicians in delivering better, more timely care, ultimately benefiting both providers and patients."

MEETING THE CHALLENGE

The Expanse search and summarization pilot project deployed in collaboration with Google Health began with an introduction to 14 clinical champions in December 2023. This group of clinical champions included individuals with strong clinical background, high-level use of the EHR, willingness to work with the team, and the ability to give constructive feedback during testing of the technology.

Additionally, extensive user experience and user acceptance testing took place prior to go-live.

In January 2024, Mile Bluff Medical Center launched Expanse search and summarization, delivering it to 125 users in 19 different departments. The users included physicians, advanced practice providers, RNs and patient care staff. The core group of departments included ambulatory clinics, emergency department, acute care, OB and surgery.

During the three-day go-live period, 86% of users were trained in person and there were zero login or outstanding tech issues.

"This was one of the most successful product launches ever in the experience of the teams involved," Brandt reported. "Clinicians began to quickly use the product and saw an immediate impact on workflows. The intuitive nature of the solution has led to overwhelmingly positive feedback.

"Word regarding the powerful functionality of the product spread quickly throughout the organization," he continued. "Upon launch of the solution, we began to receive multiple requests for access and continue to field requests for delivery of the solution. Since go-live, we have delivered access to about 180 users and have expanded to 23 departments."

Examples of use cases include the emergency department using the conditions section to quickly familiarize themselves with the patient or using search to find DNAR scanned documents. Ambulatory clinics use the technology to find information from transfer records like pathology for a colonoscopy or evaluate trends in HgbA1C for a diabetic patient.

The surgery department uses the technology to find preoperative tests and evaluations. With the help of Expanse search and summarization, infection control can rapidly surface information for sepsis or joint infection reporting. The health information management services department uses the technology for chart abstraction and updating the problem list.

RESULTS

The goals and impact of Expanse search and summarization were measured on five criteria:

ease of access, are users easily able to find and access search and conditions explorer integration; helpfulness, do users find value and joy in using search and conditions explorer; engagement, do users interact frequently and deeply with search and conditions explorer; adoption, do users discover and try search and conditions explorer; retention, do existing users keep returning to search and conditions explorer.

"Regarding ease of access, the post-launch survey revealed 86% of users reported being somewhat or very satisfied with search and summarization," Brandt said. "Regarding helpfulness, the survey showed 76% of users reported search and summarization is very or extremely helpful in accomplishing specific tasks.

"Specifically, our HIMS department reported a complete transformation in their workflow," he continued. "By using the solution, the department could save up to 40% of time when abstracting a large volume chart. With approximately 50 new patients each week and a time-saving average of 20 minutes per patient, this equates to approximately 16.7 hours per week."

Regarding engagement, the survey revealed 91% of users reported search and summarization feels somewhat or much faster than before. Regarding adoption and retention, there is a theme throughout the organization of saving 5-10 minutes with tasks such as chart prep or managing workload items.

Depending on the situation, this can be seen as a savings per patient or a savings while completing certain clinical or administrative tasks.

"I used to spend about 15 minutes per patient cleaning up problem lists, especially for new patients or patients transferred to me from other providers," explained Chief Medical Officer Dr. Angela Gatzke-Plamann. "The search and summarization functionality will help me cut this time down significantly."

ADVICE FOR OTHERS

There is an enormous amount of power in AI to revolutionize healthcare – if used correctly, AI will improve the patient-provider relationship, Brandt said.

"These tools can enable more active engagement with the patient," he noted. "Because of the power of this technology, it is important to have close provider engagement while developing and during the use of the technology. At a minimum, healthcare organizations should maintain this as an agenda item during their medical staff meetings, EHR governance meetings, and, of course, IT meetings.

"When considering use cases, carefully explore the use case and have a good understanding of the current workflow as well as the problem that organizations are trying to solve," he advised. "Make sure current tools or solutions are being used to their best capabilities and provide for additional training on those tools, if needed."

Despite how providers feel about AI, whether that be positive or negative, clinicians need to be engaged in the technology conversation and their ideas and/or concerns need to be heard, he added.

"Listening to those concerns and working to identify each department's pain points allows our team to identify and demonstrate solutions that might mitigate those pains," he concluded.

Follow Bill's HIT coverage on LinkedIn: Bill Siwicki
Email him: bsiwicki@himss.org
Healthcare IT News is a HIMSS Media publication.

The HIMSS AI in Healthcare Forum is scheduled to take place September 5-6 in Boston. Learn more and register.