How Process Mining in Healthcare Can Improve the Employee and Patient Experience

As part of Solutions Review’s Contributed Content Series—a collection of articles written by industry thought leaders in maturing software categories—Grace Nam, the Strategic Solutions Manager at Laserfiche, explains how process mining in healthcare can help organizations improve employee and patient experiences.

Even before the COVID pandemic, healthcare workers were feeling overwhelmed and overworked. In studies conducted before 2020, 38-54 percent of physicians, 36 percent of residents and medical students, and up to 41 percent of nurses reported burnout symptoms. Though some obvious causes—like challenging work/life balance, high-pressure workload, and low staffing levels—are at the root of emotional and physical fatigue, another burnout catalyst often goes undiscussed: information overload.  

Healthcare systems generate enormous volumes of data, from lab results to treatment plans to payment-related paperwork, daily. Performing data-related administrative tasks—especially within aging IT systems—prevents healthcare providers from actually providing healthcare. More than 6 in 10 physicians say they “have little to no time to address their patients’ social determinants of health,” a fact that many feel contributes to burnout.  

One surprising key to giving practitioners more time with patients is through a technique called process mining. Evaluating systems and processes is essential for healthcare organizations looking to transform their operations digitally—but many don’t know where to start.  

Process Mining Unearths Helpful Insights in Clinical Settings 

Every workflow process has an information trail, but that trail is often left unattended. The Institute of Electrical and Electronics Engineers (IEEE) task force on process mining describes how “process mining techniques…extract knowledge from event logs commonly available in today’s information systems.” Mining software analyzes event data and provides a visual working model of that process, so organizations can “discover, monitor, and improve processes in a variety of application domains.” 

In a healthcare setting, the three types of process mining—process discovery, conformance checking, and enhancement—can relieve practitioners of burdensome administrative tasks. During an ER visit, a health information system records an event log for each admission. This contains information such as a case ID, activity, timestamp, whether the interaction has been completed, and who the attending professional is. However, the event log relies on manual input and is subject to human error.  

Process discovery uses algorithms to build a model based on the event log. Often, the model will focus on the order of activities, how those activities relate to one another, and the resources needed for each activity. In this case, the model will help teams visualize the flow of ER admission, triage, testing, treatment, and release or readmission, highlighting activity patterns humans may not have seen.  

Conformance checking compares the information presented in a model with the information in the actual event log. This version of process mining evaluates whether the model accurately describes the activities in question and uncovers deviations between observed and modeled behavior. The conformance-checking phase might show providers, for example, that antibiotic administration to a sepsis patient violated medical guidelines for timely administration in some instances.  

The third type of process mining is enhancement. This iteration of process mining uses actual event data to improve or extend an existing model. Enhancement can apply triage-form data to a model of sepsis-related ER admissions to show where process bottlenecks or data re-entry incidents exist. 

Click and Notification Overload is the Impetus Behind Process Mining Adoption 

As the sepsis example demonstrates, there are myriad drivers of process optimization in healthcare settings, from variable data quality to protocol deviation. Ninety percent of the physicians who wanted more time to address patients’ social determinants of health say “removing low-value work, including EHR clicks and minimizing inbox notifications” would be the most helpful action to improving health and wellbeing.  

Nearly three-quarters of physicians in a different survey are “very or extremely interested in technology that cuts down the time of administrative work,” which includes visit summaries, charting, and data entry. Process mining in healthcare can sift through data patterns from these activity logs to indicate where automation may effectively cut down redundant work. 

Process Mining Solutions Streamline Arduous Processes and Data Handling 

A story from one healthcare provider—the Translational Pulmonary & Immunology Research Center (TPIRC)—exemplifies how healthcare organizations can implement solutions that interest physicians. For example, TPIRC manually managed data for patient lab results and developed treatment plans for food-based allergies. By identifying solutions with process mining, TPIRC realized it could automate lab-results communications and treatment plan creation and even predict the level of allergens that will cause a reaction.  

Now, when patients show up, TPIRC clinicians only need to enter a patient’s identifying number. The electronic records system then displays all the required information for a given visit while enabling the TPIRC team to add information to its growing data set and improve treatment effectiveness.

By automating patient communications, TPIRC reduced the time practitioners spent poring over charts and spreadsheets. At the same time, it increased the number of patients it could see by 400 percent while maintaining the same number of staff and high levels of employee satisfaction. 

Process Mining Can Improve Healthcare Outcomes 

In healthcare, employee burnout—and the often-lagging pace of innovation—is not merely an inconvenience: it’s detrimental to care delivery and saving lives. Every industry has plenty of operational inefficiencies, but optimizing processes to eliminate those inefficiencies is critical to improving outcomes for practitioners and patients. While it is difficult to know where to begin, given the sheer volume of data and systems providers have, process mining in healthcare is the key to identifying those discrepancies and gaps which are diminishing care. Process mining enables organizations to go behind the scenes to gather valuable insights providers can use to inform decision-making, upgrade processes, and spend more time with patients.