Learning Center

Learning Tracks    Reports    Case Studies    Webinars    Members     Charts     My Files     Vendors

 Download Full Report

Similar Reports

 View Members

Finding the Right  Recipe for Documentation
Dec 2019

Finding the Right Recipe for Documentation

Speech Recognition, Scribes & Other Methods—Impact Report

Authored by:  Taylor Davis and Connor Bice, 12/04/2019 | Read Time: 4 minutes

Many healthcare organizations wonder what recipe will lead them to EHR documentation success. Are scribes a cure-all? What factors lead to success with speech recognition? And should organizations still use dictation?

Want to see full details?


Want to see full details?

or Login 

Close X
I am interested in joining the Arch Collaborative.

Here is my information:

loading Sending Email...
After clicking "submit" above, a KLAS Representative will personally contact you within a day or two.

Based on the feedback of 37,718 physicians from 202 organizations, this report examines the impact of various documentation “recipes” on the following satisfaction and efficiency metrics:

Net EHR Experience Score (NEES) (EHR satisfaction rating)

Burnout rate (percent of physicians reporting burnout)

Self-reported ambulatory same-day chart-closure rate

Self-reported inpatient same-day chart-closure rate

Satisfaction with personal documentation quality

Satisfaction with peers’ documentation quality

Documentation Recipes

To determine how an organization’s documentation recipe is likely to impact EHR satisfaction and efficiency, each measured EHR deployment in the Arch Collaborative was sorted into one of eight groups depending on the organization’s average adoption level (as self-reported by physicians) of speech recognition, scribes, and dictation (adoption is considered high if it is above the median). While satisfaction and efficiency vary across these groups, not all variation is statistically significant. Overall, documentation method is a poor predictor of EHR experience and provider burnout (mathematically has only a limited impact). Statistically significant correlations are called out in future pages.

organization ehr satisfaction and efficiency by documentation recipe

Key Takeaways

Speech recognition works best when utilized on top of strong EHR proficiency. Speech recognition adoption must include strong change management and robust education.

While many physicians expect scribes to solve their documentation challenges, organizations that invest in scribes do not usually report significantly better outcomes. The few exceptions are organizations that ensure their scribes are well trained and that expect providers using scribes to be highly proficient with the EHR.

“A provider who does not know how to use the EHR will not know how to use speech recognition—or a scribe—to make the EHR easier to use.” —3M M*Modal Insights, pg. 7

“Done right, new technologies can alleviate administrative burdens for providers; done wrong, they may magnify administrative burdens, with the onus on organizational leadership for deployment, training, and support.” —Dr. Robert Budman, CMIO, Nuance Communications

Shifting the Paradigm

Surprised by the lack of impact that documentation method has on EHR satisfaction and clinician burnout? KLAS encourages organizations to consider that how these technologies and services are implemented can play a significant role in how well they move the needle.

As noted above, the success of scribes is often dependent on the quality of the training and the expectations set for providers. Similarly, use of speech recognition technology today is not always accompanied by sufficient user education and workflow redesign. Though the technology is likely to change drastically in coming years (thanks to the development of ambient intelligence), KLAS asserts that proper training and user assistance will still be important success factors.

Outcomes by Recipe

organization documentation recipe and outcomes

After controlling for the EHR in use, KLAS found the following correlations at the organizational level:

Organizations with higher use of dictation also have lower EHR satisfaction (p<.01).

There is no statistically significant correlation between documentation method and burnout rate.

Physicians at organizations with higher speech recognition adoption report higher same-day chart-closure rates (p<.01).

Physicians at organizations with higher dictation rates view their peers’ documentation as being of lower quality (p < .01).

Overall, the most effective organizations focus on only a couple of documentation methods (including personal entry and excluding dictation), likely because specializing in a couple of documentation methods allows for better support and training. Organizations with dictation or with many means of documentation see lower success.

Don’t Just Check the Box for Speech

Over 2,000 of the surveyed physicians commented on their speech recognition technology, with 45% reporting frustrations. The high number of positive comments indicates that the technology can be utilized well when accompanied by excellent EHR proficiency and speech recognition education. The quotes below are representative of the most common sentiments about speech recognition:

provider perceptions of speech recognition

“I LOVE the new [version of our speech recognition solution]. I feel like after just a few days, it knows me better than the other version did after years of me yelling at it.”

“My biggest complaint is with [our speech recognition solution]. It’s not reliable. It shuts down and crashes Epic on a daily basis. I expected this to be resolved with the latest Epic upgrade, but it was not. I usually stop using [the speech recognition solution] for the day if it crashes. The cycle repeats itself the next day. When it works smoothly, it makes the workflow more efficient.”

“Please provide the opportunity for ongoing, hands-on, at-the-computer training for both [our speech recognition solution] and Epic. I believe giving employed physicians the opportunity to observe Epic-savvy staff use and optimize the system would be a great financial investment for the corporation. Additionally, having an optimization-minded advocate from our organization looking at user problems and resolving them with Epic would really reduce job frustration and improve user efficiency.”

The data shows that the use of speech recognition technology does not guarantee documentation success—an organization can achieve very strong or very weak results depending on how the technology is implemented and how well users are trained. With how speech recognition technology is commonly implemented today, providers often report better efficiency, similar or lower EHR satisfaction, and sometimes lower confidence in their own documentation. In general, physicians report better outcomes when organizations make an all-in bet with speech recognition (i.e., don’t use scribes or dictation).

outcomes by recipe


Among those using scribes, the differences are significant depending on how the data is analyzed.

The physicians who are given scribes are often the most frustrated, so it is not surprising that as a group, they report lower EHR satisfaction and worse chart-closure rates than their peers who use other documentation methods. However, the longitudinal data indicates that these metrics can improve with use of a scribe. Next year, KLAS will be able to show trends in how scribes impact physician burnout rates.


In Summary

It is increasingly clear that organizations looking for a quick fix to their EHR satisfaction challenges, either through scribes or speech recognition, are likely to be disappointed. Those that adopt these technologies and processes in tandem with strong change management and strong EHR proficiency will likely see efficiency gains.

It should be noted that the results achieved today could change as ambient intelligence supplements speech recognition technology, as organizations adopt better implementation and training programs for their technology environment, and as organizations improve how they implement scribe programs. As these technologies and services evolve, the Arch Collaborative will continue to monitor their impact and provide insights into how they can best be implemented and what results organizations can expect.

Report Non-Public HTML Body

Report Public HTML Body


Report Topics

Opioid Abuse Prevention

Upload Full Report

 Download Full Report

This material is copyrighted. Any organization gaining unauthorized access to this report will be liable to compensate KLAS for the full retail price. Please see the KLAS DATA USE POLICY for information regarding use of this report. © 2019 KLAS Research, LLC. All Rights Reserved. NOTE: Performance scores may change significantly when including newly interviewed provider organizations, especially when added to a smaller sample size like in emerging markets with a small number of live clients. The findings presented are not meant to be conclusive data for an entire client base.