Patient admission and the associated manual tasks performed prior to medical examination are key components of the operational efficiency and clinical quality of a health system.
As a result, some hospitals and primary care clinics are experimenting with artificial intelligence and large language models to focus on note-taking and transcription work.
As physicians burn out from spending more and more time on electronic health records (EHRs), can this technology be applied to other departments outside of the clinic?
Dr. Josh Reischer is CEO of Health Note, a platform that automates pre-visit documentation and helps optimize exam room time by limiting EHR time. We spoke with him about the role of AI in patient intake and other workflow efficiency efforts.
Q. You address ways to automate patient admissions and manual tasks performed prior to a medical examination. Why is this area important to the overall performance of a healthcare system?
ONE. Pre-appointment interactions set the tone for the entire healthcare experience. If done right, patients and doctors can communicate better, administrative staff can work more efficiently, and organizations are best prepared to get paid for the care they provide . If you don’t do it well, you will end up with the system you have today.
You wouldn’t walk into an interview unprepared, but for doctors, it’s the norm, and patients can tell when they’re expected to re-share the exact information they filled out at home or in the waiting room. Doctors don’t have much choice in the matter, and the demands on them are overwhelming.
The goal should be to improve efficiency for patients, staff, and physicians. That means people are freed from calling cards, paper surveys, repeat questionnaires and other unnecessary medical details. Furthermore, when physicians have complete information about their patients, it has the potential to improve examination. They can ask prescriptive, non-redundant questions that help put the patient at ease and lead to a better diagnosis.
We are seeing physicians spending more time in the EHR. This is compounded by I used to struggle with endless notes and late-night “pajama time,” as physicians were required to code and summarize patient interactions into the EHR.
Only one in five doctors said they had enough time to take notes.. In the same study, 42% felt they had little control over their workload and more than half spent too much time at home completing paperwork.
Revenue integrity and compliance is another group within a health system that can benefit from a streamlined, automated import and export process. Depending on billing, there are different documentation requirements and patient-reported outcomes need to be collected for each group.
The solution is to collect relevant patient-reported results before the appointment begins. That way, the doctor has a confirmation reminder and doesn’t have to worry about forgetting to add them. Traditionally, updates like this require training and have a delay. However, workflow automation can help integrate these updates for all doctors and their patients at once.
Q. Health systems are increasingly testing AI and large language models for note-taking and note-taking problems. Is this technology applied in places outside the clinic or is it just a single tool?
ONE. Artificial intelligence is an amazing technology because it is so adaptable. I think AI will replace some areas of our industry, but I don’t believe it will replace humans in the loop. Most importantly, AI will make healthcare more humane than it is today.
Healthcare technology feels like it’s in its awkward adolescence – still figuring out its purpose. Healthcare technology today is clumsy and clunky, and doesn’t always do what it’s supposed to. Artificial intelligence feels like the beginning of the next phase of adulthood.
Scribes is a good start for generative AI, but there are a number of use cases throughout the patient and physician journey that will be improved by technology outside of the visit. Yes, it starts with simple manual tasks and labor that are outdated and unnecessary in healthcare today, but I’m excited about how healthcare can be better for everyone.
When AI is applied to the end-to-end healthcare experience, there are some significant benefits. For example, during the intake process, a chatbot or intelligent tool can work with a patient to discern pain levels, previous medical history, social determinants of past flare-ups, or simply recall a previous visit.
Innovative AI can communicate in many different languages, translate medical shorthand and empathize with the patient’s condition. This is a huge win for organizations serving diverse and underserved populations.
Even in my medical career, I have seen the industry move from paper forms and keyboards to buttons and touchscreens. I believe AI will open up new, more intuitive ways for us to communicate with each other and with the technology that supports us.
Q. Physicians spend more time working in the EHR and this contributes to clinician burnout. What are some ways to improve technology to ensure doctors are more efficient and prescribe on time?
ONE. It found that the time clinicians spend on the EHR increased by 28 minutes from 2019 to 2023. That’s unlikely, especially at a time when more and more medical professionals are signaling want to leave the industry. Technology makes the mundane better and allows physicians to practice at the highest level of licensure.
It sounds cliché but it’s true. Today’s tools will encourage me to continue working in patient care. I left internal medicine in 2018 to co-found Health Note. We all want to get rid of paper notes; we are constantly regurgitating every minute of our patient interactions – the good, the bad and the ugly – for accountability and outdated payment models.
Artificial intelligence has been changing thatwith the document companies handling that part. We need to expand this to the admitting process and other areas to enable doctors to spend more time with patients and sleep better at night. Bottom line: Doctors can rely on technology to increase personalization and get to know their patients before interacting in the clinic.
Q. Are there other ways that health IT design fails? If so, how to fix these?
ONE. In this first wave of innovation, doctors, nurses, and clinicians were completely excluded from the design phase. I have seen a lot of incompatibilities and misunderstandings in the use of technology. That’s why I believe that more doctors like me are now developing much more practical tools.
We’re going to have to interact with EHRs for the foreseeable future – but it can evolve. Point solutions are going away for the foreseeable future. Now we’re talking about enterprise platforms that can penetrate multiple care sites and specialties, from urology to musculoskeletal. Healthcare is multilayered, and you can’t just design with one use case in mind.
It must be flexible enough to apply to a variety of contexts and populations, from rural Appalachia to New York City. As we prepare for the silver tsunami of the aging Boomer generation, where 10,000 people turn 65 every day, we should all expect health care utilization to increase. Therefore, it is imperative that we build a sustainable workflow for physicians and create a personalized experience to meet volume pressures.
The HIMSS AI in Healthcare Forum is scheduled to take place September 5-6 in Boston. Learn more and sign up.
Follow Bill’s HIT coverage on LinkedIn: Bill Siwicki
Email him: [email protected]
Healthcare IT News is a publication of HIMSS Media.
