Getty Images/iStockphoto

Artificial Intelligence (AI) Use Cases to Mitigate Clinician Burnout

Artificial intelligence (AI) tools can automate administrative tasks to help address clinician burnout and give providers more time to deliver patient-centered care.

Artificial intelligence (AI) tools may be key to improving physician satisfaction with ongoing staffing shortages and all-time high clinician burnout rates.

The HITECH Act of 2009 encouraged EHR adoption to streamline clinical workflows, reduce paper waste, and support the cost-effective delivery of healthcare. However, physician time on medical record-keeping has doubled since HITECH's passage. Clinicians now spend two minutes at the computer for every one minute spent with patients, and workdays have extended into providers' home lives.

AI tools are helping automate administrative tasks to give providers time to focus on patient-centered care. In a 2021 survey, virtually all healthcare executives (99 percent) said they recognized the opportunity to mitigate clinician burden through AI and automation.

AI can be applied to several healthcare use cases that may each move the needle on clinician burnout.

Health Data Extraction

According to an athenahealth survey conducted by the Harris Poll, 58 percent of physicians said they often feel so overloaded with information in the EHR that it increases their stress.

"I work with physicians a lot and talk to them about data searching," Johnathan Hartmann, clinical informaticist at Georgetown University, told HealthITAnalytics in a January 2022 interview. "Their main complaint is that they can't pull out the precise information they want quickly when they do a search. It takes them too long and too much work to get to the information that they want."

To address information overload, Georgetown University Medical Center adopted a text-mining tool from health IT vendor Linguamatics.

Text-mining EHR integrations that leverage natural language processing (NLP) can help address information overload by allowing physicians to search through large amounts of medical data for clinical decision support.

NLP is the overarching term that describes using computer algorithms to identify key components in everyday language, extract meaning from unstructured or written input, and turn it into usable data. NLP requires the use of artificial intelligence, computational linguistics, and other machine learning methods.

Research has shown that these AI-based tools for health data extraction can significantly cut down on EHR screen time and improve physician satisfaction.

A 2021 JAMA Network Open study of 12 gastroenterology physicians/fellows found that an AI tool that extracted relevant patient health data and displayed it next to the patient record cut EHR use time by 18 percent. Survey results found that 11 of 12 physicians (92 percent) preferred AI-optimized record review to standard chart review.

While the clinicians noted a learning curve, 11 of 12 said they believed the health IT solution would save them time and were interested in using the EHR integration in their clinic.

Legacy Data Conversion

When healthcare organizations move to new EHR platforms, patient records are often incomplete, inconsistent, or missing.

When an EHR sends patient data to another EHR, it often comes in a different language, spurring the need for manual data translation. For instance, one EHR may use the terminology "take by mouth" in a medication sig, while another EHR may say, "The route is oral, not by mouth."

When WellSpan Health was transitioning to a new EHR system from three legacy systems, employees had to reach out to pharmacists and other providers to gather medication data and manually enter it into the EHR.

"It can take a ton of clicks and scrolling and selecting dropdown menus for a human to do that translation manually," Robert Lackey, MD, FAAFP, CMIO of WellSpan Health, told EHRIntelligence in an interview. "If you're expected to do a bunch of work, it decreases your efficiency, creates opportunities for errors, and that's where your patient safety issues can show up."

To streamline the medication transcription process, WellSpan tapped an AI-based tool called MedHx from health IT vendor DrFirst. The solution provides a comprehensive medication database comprised of local and national medication history sources, including HIEs and EHR partners, directly in the native Epic workflow.

Then, to automate the translation process, WellSpan leveraged the vendor's AI solution SmartSig, which takes prescription transaction data and translates it into the native EHR nomenclature.

Lackey explained that the automated system saved clinical time and cognitive effort, allowing employees to focus on providing quality care.

At first, translating each patient record into the new EHR took an average of 20 minutes. The health IT integration shaved off five to seven minutes, which can have good implications for clinician burnout rates.

Medical Scribing

Healthcare organizations have long used medical scribes to assist in clinical documentation. Medical scribes are typically unlicensed paraprofessionals whose job is to document patient encounters in the EHR.

Prompted by advances in speech technology, AI-based EHR scribing services have come to market in recent years.

The solutions often involve providers using their smartphones as voice input devices during patient encounters. Once the recording is complete, the dictation will transfer into the EHR progress note on the clinician's laptop or desktop computer. The tools aim to free up provider time for patient care instead of clinical documentation.

"Relative to the cost, it's a very good return on investment, given the support it provides our clinicians," Daren Wu, MD, chief medical officer at Open Door Family Medical Center, which has utilized AI-based scribes, told EHRIntelligence in an interview. "The feedback, on the whole, has been this tool has been tremendous in cutting down after-hours time, home time, and weekend time on finishing notes."

Human scribes do have their benefits over AI-based scribes. For one, clinicians can ask humans to do specific things when scribing, like populate a certain area of the EHR or to clarify an error. But there are also key pros to using AI-based scribes. Wu noted that the AI-based scribing service allows clinicians to lock their notes faster because they're not waiting for a human scribe to lay out the note.

As healthcare organizations look to retain employees amidst a nationwide staffing shortage, adopting innovative AI tools that improve EHR satisfaction could be key.

A 2022 KLAS report found that clinicians who are very dissatisfied with the EHR have almost three times the proportion reporting they are likely to leave compared to clinicians who are very satisfied with the EHR.

Dig Deeper on Health IT optimization

Cloud Computing
Mobile Computing