News of the tool, currently in clinical pilots, comes a week after reports circulated that Google was working with healthcare system Ascension on a controversial project involving patient data.

Google demos its EHR-like clinical documentation tool

By Laura Lovett
Share

Photo by Carsten Koall/Getty Images

Google Health has offered a peek at what it's planning for its next set of clinical digital tools. It includes a more integrated charting system that aims to make it easier for doctors to search for a variety of metrics and notes. 

“With a single login, doctors can access a unified view of data normally spread across multiple systems. All the types of information clinicians need are assembled together such as the vitals, labs, medications and notes,” Dr. Alvin Rajkomar, product manager and practicing physician, said in a demo video

“Clicking on any value will start a deeper exploration showing recent and historical trends both graphically and with tables," he explained. "Doctors can query the entire chart with their own words and typos. Results are not strict keyword matches. A variety of Google technologies are used to identify related concepts.”

The new tool, which is still in the pilot phase, lets clinicians use shorthand to search and will bring up results that may have typos. Users can also jump to different parts of the chart, such as vitals or notes, by using the search feature. Doctors will also be able to search through scanned documents and faxes, both with handwritten and typed notes, for information.

St. Louis-based Ascension will be the first health system to pilot Google's new technology. This past week it was reported that Google has been working with Ascension since 2018 on a collaboration involving patient data, called Project Nightingale. 

While that news drew concerns among many patients (and some Ascension employees) about patient privacy, the partnership appears to be HIPAA compliant.

“Our work adheres to strict regulations on handling patient data, and our business associate agreement with Ascension ensures their patient data cannot be used for any other purpose than for providing our services — this means it’s never used for advertising,” Dr. David Feinberg, head of Google Health, said in a blog announcing the new technology.

Feinberg went on to specify that there are strict controls for Google employees who handle any health data. First, the tool was tested on fake data and openly available data sets, he said. He went on to write that Google employees who come into contact with data have a HIPAA and medical ethics training. 

However, he said that the company will look to reduce the number of Google staffers who need to access patient data. The system will also undergo a third party audit.

WHY IT MATTERS

Documentation and EHR burden has been shown to contribute to physician burnout. In fact, a Health Affairs study, which centered on administrative and clinical data from Palo Alto Medical Foundation's primary and specialty physicians, found that the number of messages received by these docs had a correlation with their probability of burnout.

A recent Mayo Clinic report also linked burnout with a suboptimal EHR experience.

Now new entrances to healthcare, like Google, are entering the space. However, the public remains wary of Google having access to private data. The tech giant has come into question before.

In July Google, the University of Chicago Medical Center and University of Chicago were named defendants in a class action suit that alleges they failed to properly de-identify sensitive patient medical data.

However, Feinberg used this blog to address privacy concerns about this project.

THE LARGER TREND

Google has been interested in the clinical documentation space for sometime. Last February Google published a patent application on a predictive EHR system first highlighted by the tech company last May. The application was filed by Google in July 2017, and has not yet been granted.

As described in the application, Google’s system can aggregate and store EHRs for a diverse population, while compiling each individual patient’s records into a single chronological document. A computer or computer system running deep learning models would then use this collection of data to guide predictions of future health events, and to better contextualize the collected data from an individual’s record to highlight pertinent past events on an EHR.