During training, Amy Ho, MD, MPH, was perplexed by why the hospital's door-to-needle time for stroke patients was consistently higher than the targeted 60 minutes, when it seemed clear that the ED staff was providing care as fast as possible.
Upon further investigation, she discovered that physicians were missing the time goal due to unavoidable delays, such as a wait for consent from family, but those justifying circumstances weren't being applied to their statistics.
“The problem was that physicians were documenting in the EHR in a free-text way,” said Dr. Ho, an emergency medicine physician and senior vice president and chief of clinical informatics at Integrative Emergency Services in Dallas. “I created a dropdown menu with a data element behind each option so physicians could easily select an approved exemption for why they didn't meet the goal. That allowed the hospital to exclude qualifying cases easily, and our compliance shot up to 100% as a result.”
For Dr. Ho, the project opened the door to a career in clinical informatics. As the stroke example shows, overlapping expertise in providing care and crunching data can be key to addressing vexing quality and workflow issues. Hospitalists are in an ideal position to get involved in informatics because they're used to working across multiple specialties and departments, said Dr. Ho and other experts.
Opportunities to participate exist at every level, from simply joining a committee or quality improvement initiative to taking courses to pursuing a master's degree or board certification.
“The easiest point of entry for hospitalists is joining a multidisciplinary team that already has informatics and analytics support,” said Sunil Kripalani, MD, MSc, FACP, a member of the clinical informatics center at Vanderbilt University Medical Center in Nashville, Tenn. “Start by finding a project that interests you and aligns with hospital priorities—such as patient safety—where you can offer clinical insight and guidance while learning from other team members.”
Why to do it
Hospitalists already have a first-hand perspective on the clinical workflows analyzed by informatics, said Nicholas Perkins, DO, FACP, a hospitalist and associate chief medical informatics officer at Prisma Health in Greenville, S.C.
“We live and work in an environment where everything is monitored through data analytics,” he said. “Physicians are either in the driver's seat of helping procure, validate, and gain insights into data, or we're handing it off to someone else.”
Clinicians understand how workflow changes can affect common quality metrics, such as length of stay (LOS) or readmission rates, and they can often predict whether a project will capture the right data at the outset and measure appropriate outcomes, according to Dr. Perkins.
“Many scorecards and metrics don't translate to clinicians' actual day-to-day tasks because they're based on processes that were historically defined by nonclinicians,” he said. “Hospitalists can be the voice up front saying, ‘How do we bring clinically relevant metrics into this work?’”
For example, a hospitalist may have better insight into potential causes of readmissions than a data analyst or administrator and should be involved in the design of projects to reduce them, said Shireen Jindani, MD, FACP, assistant clinical professor of biomedical informatics at the University of Arizona College of Medicine-Phoenix.
Hospitalists should get used to thinking of informatics as an extension of their clinical practice, she said. Issues or trends noted during the course of patient care can be turned into quality improvement projects. For example, if a hospitalist notices high body mass index coupled with chronic electrolyte changes, indicating risk for obstructive sleep apnea, she could work with a data analyst to flag those patients in the EHR and automatically connect them with outpatient scheduling to discuss setting up a sleep study.
Similarly, informatics teams can help hospitalists address their frustrations in workflow, said Parag Mehta, MBBS, MACP, a hospitalist and chief medical information officer at New York-Presbyterian Brooklyn Methodist Hospital in Brooklyn, N.Y., and Vice-chair of ACP's Medical Informatics Committee. He recalled one project that was triggered by physicians' frustrations over delays in receiving imaging results promptly.
“Everyone was ordering CAT scans as ‘urgent’ because they believed it really was urgent or they were trying to discharge a patient, which led to long backups and delays,” he said. “Working with data analytics, we created a new category called ‘life-threatening emergency’ that immediately got the highest priority.”
Informatics can also help ensure that every needed step in hospitalization is completed. Issues can easily get lost to follow-up in a high-acuity, fast-paced environment where hospitalists are focused on stabilizing patients and getting them ready for discharge, said Dr. Jindani.
“One of the most exciting things about informatics,” she said, “is that it allows us to help create workflows that are automatic, rather than committing things to memory.”
How to get started
Even hospitalists with little or no prior experience working with data can provide important contributions to informatics projects, said Dr. Kripalani. Data analysts usually welcome clinicians' on-the-ground perspective.
For example, a hospitalist looking at a project aimed at reducing LOS would know to separate patients discharged to home from those going to a skilled nursing facility—but that distinction might be missed when setting up a basic dashboard, he said. That initial input is vital, considering that the interventions needed to shorten LOS would differ between the two groups.
“By working hand in hand with data analysts, hospitalists can provide a clinical lens to focus on the patient populations or subgroups of greatest interest,” said Dr. Kripalani. “In the process, hospitalists can learn a lot about data sources and what goes into preparing the tools for visualizing and analyzing those data.”
To get started, Dr. Mehta recommends joining an informatics committee, which exist in almost every hospital and typically include a variety of subcommittees. Pick a subcommittee that matches your area of interest, he said, such as data governance, EHR template creation, or clinical decision support.
At the same time, it's a good idea to set up a one-on-one meeting with an applications analyst or report writer, said Dr. Ho. The hospital's EHR educators are another great resource.
Dr. Perkins encourages hospitalists at Prisma to participate in its EHR optimization group, which is currently working to reduce the volume of clinical alerts. The team welcomes input on how to make metrics clinically relevant.
“There is always hesitation when a project is brought in by an outside IT group or administrators who are removed from care,” he said. “Having a hospitalist who lives that scenario helps with the credibility of the entire project, and it opens up a line for hospitalists to bring in new ideas.”
Engaging with the hospital's chief quality officer, who will be familiar with the informatics aspects of improvement initiatives, is another pathway, suggested Dr. Kripalani. “The chief quality officer can speak to the hospital's current priorities and composition of various teams,” he said. “They can help you find a team that fits with your interests and expertise.”
Joining a learning collaborative is an option for those working at smaller hospitals that may not have stand-alone IT departments or large-scale quality initiatives, said Dr. Kripalani. One example is the Hospital Medicine Reengineering Network (HOMERuN). Founded by researchers at 13 academic medical centers, the network conducts national, multicenter studies aimed at improving patient care.
“More than 50 hospitals participate in HOMERuN,” he said. “Joining a multicenter project can be a good entry point because you're pooling data with other hospitals and collaborating on quality projects aimed at improving care delivery across the country.”
No prior coursework or advanced skills are required to start getting involved in informatics initiatives, noted Dr. Ho.
“As a clinician, you already understand the knowledge of health care and clinical workflows,” she said. “That deep clinical knowledge already sets you up for informatics work, and any residual technical skills you can acquire as you move along.”