Six Capabilities that Make Astrata NLP Different

Apr 20, 2021 | Technology

If your healthcare organization is looking at Natural Language Processing (NLP) vendors to help you transform your HEDIS operations, you’ll want to benchmark them against these six NLP capabilities that will make or break your HEDIS transformation.  How well does your NLP vendor measure up?

1. NLP that’s optimized for HEDIS

Most NLP vendors can identify a range of conditions, procedures, and medications in the text, but that doesn’t mean they are able to identify the full range of HEDIS evidence. For example, provider notes can use an astonishing array of synonyms. Let’s use the Colorectal Cancer screening measure COL as an example. A colonoscopy may be called “c-scope”, or a FIT-DNA might be referred to by a brand name like “Cologuard”.  Misspellings such as “Colonscopy” occur frequently. Without the full range of synonyms and common misspellings, you’ll see false negatives that depress your clinical rates. Astrata’s proprietary technology starts with the HEDIS value sets, but greatly expands them to include terms and phrases that do not exist in other terminologies.

Image showing NLP extraction from clinical notes

Using NLP to identify HEDIS evidence means that your NLP vendor will need to extract temporal information, such as the date when a colonoscopy occurred. And NCQA’s requirements are subtle. For example, mentions of procedures like colonoscopy that are referenced by temporal expressions such as “three years ago” are acceptable as evidence of gap closure as long as they are within the required range of ten years. But lab tests such as FIT DNA require different reasoning because the entire date (day, month and year) must be explicitly mentioned. Astrata’s NLP Insights product provides the power to identify when the event occurred, and to evaluate it against the HEDIS specification to know whether the gap can be closed with current documentation.

Image showing NLP extraction from clinical notes

2. Understands the context to accurately code Hits, Exclusions, Leads and Gaps 

Your abstractors make complex judgements to determine whether a specific case has some or all of the evidence to close the gap,  or meets criteria for exclusion.  Your HEDIS NLP system should be able to do the same. For example, not all colonoscopy reports provide evidence of a completed colonoscopy. But based on the current HEDIS specification, only completed colonoscopies can be used to close the COL Gap. Astrata’s NLP Insights understands the context and nuances of HEDIS to deliver highly accurate classifications across your populations. 

Image showing NLP extraction from clinical notes
Image showing NLP extraction from clinical notes

3. Integrated into workflow tools 

The most sophisticated data analysis won’t move the needle unless it’s integrated into your workflow, to drive efficiency and increase HEDIS clinical rates. How does your NLP vendor surface insights to your Quality team? Astrata’s Chart Review product seamlessly integrates our analysis into a workflow tool for gap closure that was designed in collaboration with a HEDIS abstraction team. Surface closeable cases faster and get down to the HEDIS evidence with a single click. Our chart review application makes year-round HEDIS for hybrid measures possible across entire populations.  


4. Customized for your environment and providers 

Natural Language Processing has an important limitation that can throw your organization off kilter. NLP systems typically experience a decrease in accuracy when they are implemented in a new location. There are many reasons for this “NLP portability problem”, but the key for your organization is to make sure your NLP vendor has a built-in process for modifying its system to your environment to boost accuracy in your setting. At Astrata, our customer agreements build in this “tuning” phase as part of our engagement with you. Our ontology-based systems make it easy to adjust to the nuances of your providers.  


5. All about accuracy

When all is said and done, your NLP system’s accuracy will be one of the most critical success factors. When evaluating HEDIS NLP, look for vendors that understand how to measure their NLP performance, and incorporate evaluation across the entire software lifecycle. With an in depth understanding of measuring NLP performance, Astrata incorporates evaluation in everything we do, from measure development to ongoing monitoring on site. We provide the reports and tools to give you line of sight into our accuracy, and empower you to meet your auditing requirements. 


 6. In sync with NCQA

This is an exciting time for NLP, as better technology opens up new use cases and product possibilities. It’s exciting to see NCQA recognize the potential of NLP for its HEDIS program through the development of the NLP Working Group and a pending NLP pilot. Astrata was one of only three companies invited to participate in the NLP Working Group. As experts in this burgeoning field, we will always keep our company in sync with NCQA and its transformational goals in Digital Quality. Can your NLP vendor say the same? 

Read this next…

Astrata News

Astrata NewsAstrata named as recipient of the KLAS Payer/Provider Points of Light Award May 20, 2024 PITTSBURGH, PA —Astrata has been named a recipient of the KLAS Payer/Provider Points of Light Award in collaboration with AmeriHealth Caritas, ELLKAY and Prospect...

Astrata has been named a recipient of the KLAS Payer/Provider Points of Light Award

Astrata has been named a recipient of the KLAS Payer/Provider Points of Light Award in collaboration with AmeriHealth Caritas, ELLKAY and Prospect Medical Holdings, Inc for its work in Closing Care Gaps for Healthcare Effectiveness Data and Information Set (HEDIS)...

The Road to Digital Quality – Astrata’s Maturity Model Approach

In this month’s blog, we’ll unpack our Digital Quality Maturity Model to help you stage your technology transition to Digital Quality Measurement. Whether or not you are using Astrata’s eMeasure Digital Engine, you can use the digital quality implementation maturity...

What one year of ChatGPT has taught me about the future of Quality measurement – Part 2. Moving Beyond the Hype

If you had a chance to read Part 1 of this blog last week you know that I am extremely optimistic about the value of generative AI to healthcare quality measurement. In Part 2 of this blog, I am going to give you a sense of how these technologies work as well as what...

What one year of ChatGPT has taught me about the future of Quality measurement – Part 1. The AI HEDIS Analyst

It’s been over a year since the release of OpenAI’s ChatGPT. And this is my first blog on the topic of how I think this new technology will fundamentally change healthcare Quality measurement and improvement. Why the wait? With the extraordinary flurry of activity,...