AI Front and Center at 2023 eClinicalWorks Health Centers Summit

Product enhancements based on the application of artificial intelligence (AI) featured prominently at the recent 2023 Health Centers Summit (HCS2023) hosted by eClinicalWorks. CEO Girish Navani showcased new AI powered search capabilities, AI-generated patient summaries, and next-step recommendations. The company also debuted a new macro feature and AI-assisted appointment reminders designed to improve no-shows.

HCS2023 is an annual event specifically for eClinicalWorks’ community health centers and Federally Qualified Health Centers (FQHCs). These healthcare organizations have unique needs and challenges that are different than mainstream ambulatory healthcare providers.

eClinicalWorks has a significant presence in this segment of the market – the company claims upwards of 60% of FQHCs and community health centers in the US operate on their platform. With such a significant share of the market, it makes complete sense for them to host a summit that caters to this special audience.

Girish Navani, CEO at eClinicalWorks opened the Summit by sharing that this was his 24th year at the company and that he is as excited today as he was all those years ago.

It didn’t take long for Navani to announce that the company was jumping into AI with both feet.

Sprinkling AI Everywhere

Over the next hour of his opening keynote, Navani spoke about the importance of AI in healthcare. He cited the emergence of ChatGPT as a watershed moment – one that took AI from the realm of possibility to a tangible reality for non-data scientists.

Navani and his team have invested a lot of time and effort to build AI based on large-language-models, like the one ChatGPT is based upon, into their EHR platform. The company did not restrict themselves to just one or two use-cases. Instead, they sprinkled AI liberally on many aspects of their platform.

One of the most useful AI-based enhancements was the ability for users to search for reports based on the content of the report rather than the title of it. This feature generated a lot of applause from the audience and it wasn’t hard to see why.

With this enhancement, users would no longer have to remember how reports in eClinicalWorks are labeled. Instead they can just type “report that shows the number of patients with diabetes and their average length of stay” into the search bar and up will pop all the reports that show that content. That is much quicker and easier than recalling that “Diabetic Patient ALOS Report – Monthly Average” is the report a user wants.

Navani and the eClinicalWorks team also demonstrated how they have used AI to help generate better work schedules, appointment selections for patients, and self-service options for patients.

AI for Patient Note Summaries

Most interesting was a notes summary feature powered by AI. Through eClinicalWorks’ PRISMA product, users already have the ability to search for patient information that is held outside of their organization. This information comes from a variety of external sources that participate in the Carequality and CommonWell Health alliance networks.

eClinicalWorks showed a new AI feature that can read through the pages of information retrieved via PRISMA and generate a short summary for that patient. There is no doubt this feature is a time saver for users. Instead of wading through the page and pages of health information, with a click of a button, they can get a nice succinct summary.

The question is: are clinicians ready to trust an AI generated patient summary?

I’m not 100% they are…yet. I’m sure the AI tools of today are pretty good, but what if they missed something? What if 90% of the time a 2 day bout of cough means nothing for a patient but this particular one falls in the 10% where it is significant? Are our AI engines smart enough to pick that up?

This use case also raises the question of the completeness of the records that the AI engine was trained on. Are we confident that there are enough indigenous patients in the dataset? Or those that are struggling with food insecurity? Or those that are working 3 jobs?

Personally, I think it will be a while before there is full trust in AI for the purpose of diagnosis and summary. The potential liability alone will likely give clinicians pause.

AI Guardrails

Navani did spend time addressing some of the potential concerns over their use of AI. He highlighted how the company DOES NOT send patient data to public AI engines (like ChatGPT), All AI processing is done on eClinicalWorks’ servers. It is never public.

Navani also stressed how their approach to AI at this point is that is meant to be a co-pilot, never a pilot. End-users, especially clinicians will always have the final say and can override anything a AI suggests.


Other feature announcements

There were two other product-related announcements of note from eClinicalWorks. One was a “playlist feature” which was basically a customizable macro that users can configure to perform system tasks in an automated manner.

With this feature, users would no longer have to click through multiple screens to enter information that they have entered elsewhere (like patient name, address info, latest medication). With the Playlist a macro could be created that automatically takes that information and fills in the appropriate forms.

The other feature was an new appointment reminder module that uses AI to target those patients that are most likely to not show up. The AI looks at past history within the EHR and mixes in other factors to generate a priority call/text lists that case workers or other staff can use to contact these patients.

Overall Positive Vibe with Genuine Excitement & Appreciation

From the reaction of the audience to the sessions and from the 1:1 conversations I had with attendees, there was genuine excitement for the new features that eClinicalWorks debuted. There was also appreciation for eClinicalWorks for providing a forum for FQHCs and Community Health Centers to voice their needs and interact directly with company executives.

Overall the HCS2023 had a positive vibe and was a nice pre-cursor to the eClinicalWorks 2023 National Conference later this year where I’m sure we’ll hear even more announcements as AI related products and features continue to develop and mature.

Learn more about eClinicalWorks at https://www.eclinicalworks.com/

eClinicalWorks is a sponsor of Healthcare Scene.

 

About the author

Colin Hung

Colin Hung is the co-founder of the #hcldr (healthcare leadership) tweetchat one of the most popular and active healthcare social media communities on Twitter. Colin speaks, tweets and blogs regularly about healthcare, technology, marketing and leadership. He is currently an independent marketing consultant working with leading healthIT companies. Colin is a member of #TheWalkingGallery. His Twitter handle is: @Colin_Hung.

   

Categories