Frank Jackson, Vice President and GM at Wolters Kluwer, Health talks about the benefits that NLP brings to the healthcare table and the significance of clinical data
1. Can you tell us about your role at Wolters Kluwer?
I am the General Manager of a business unit at Wolters Kluwer called Health Language. My responsibility is for the success of solutions that are designed to improve healthcare data quality and interoperability, as well as the delivery of technology that unlocks clinical data to improve quality of care and patient outcomes.
2. Can you tell us about your journey into this market?
My journey origins come from working for payers like Kaiser Permanente and BlueCross BlueShield of North Carolina to better understand and improve the use of healthcare data. My journey closely parallels the overall payer data journey in terms of the adoption of the idea that meaningful use and application of healthcare data is a critical competitive requirement. I have transitioned to business leadership roles with payers that utilize these skills to drive business results and evangelize that all business leaders must have analytic acumen to succeed.
My path has also included time as chief analytic officer, leadership positions in organizations focused on population health, and other COO type roles. During those experiences, I saw first-hand the value of building a strong data/analytic capability. When I was working for a Medicaid company, I saw a very different health system working to manage patients with very different needs driven by factors outside the “four walls” of traditional healthcare settings. These factors and the related data are what we are considering when we refer to the social determinants of health.
The most recent part of my journey has been driven by a desire to join companies like Wolters Kluwer where I can better serve these needs of payers. It is an exciting time in healthcare with the rapid expansion of data and technological advancements that allow us to use it to uncover and leverage important insights, along with the promise of artificial intelligence.
3. Why do health insurers have such a hard time forming a comprehensive view of a member’s (or patient’s) health?
First, the legacy systems in use at most payer organizations are built to be transactional and are not designed for analytics purposes.
Payers have traditionally relied upon the information contained in individual claims, which generally lack the clinical specificity to provide real analytical value.
Second, recent interoperability mandates are intended to increase data sharing across the healthcare ecosystem, but when payers do gain access to that shared data, it is missing semantic interoperability, which is the ability to derive a common shared meaning from the data that they receive. So, they have the data, but they are limited in the amount and depth of meaning they can pull from it.
Lastly, payers have only recently started receiving clinical data and are still in the early stages of learning how to apply the clinical insights to inform a 360-degree view of member.
4. Why is it important to have this holistic view, both in terms of medical outcomes and costs?
Claims data only goes so far. Yes, it gives you information on specific medical events, but it does not capture the detail around an individual’s overall health status.
By contrast, the clinical data that is captured by the clinician at the point of care provides detail about lab results, radiology reports, medication lists, and even things like family medical history. While this information is contained in electronic health records (EHRs), most of the richer clinical insights are actually captured in free-text fields, and therefore considered unstructured data (which represents up to 80% of all healthcare data!).
The ability to leverage both claims data and clinical data, structured and unstructured, is essential to develop a comprehensive view of an individual’s health. This view can help inform care decisions to improve health outcomes, as well as be used to accurately perform risk adjustment, report on quality measures and empower analytics models about patient populations.
5. How can health insurers overcome the challenges they face in making sense of data? How can Natural Language Processing (NLP) technology help?
Payers need to continue to gain expanded access to clinically rich EHR data and accelerate their learning about how to utilize it to build a holistic view. If they rely only on claims and prescription data, the view they can build of a patient is very limited. Adding clinical data to that view expands the ability to understand the patient more holistically and inform actions and decision-making that can lead to better outcomes.
As noted above, since 80% of patient data in an EHR is stored in unstructured or semi-structured text, technology like clinical NLP can help those clinical insights in an efficient manner.
6. How has NLP technology evolved over time?
Machine learning has advanced precision and accuracy across general use cases. NLP has evolved to support specific use cases and increase accuracy for specific purposes. Our clinical NLP technology is built to learn fast and tuned to support high accuracy across wide set of use cases.
7. What are some of the other high value uses cases NLP can support for health insurers?
Payers using NLP will gain significantly more access to clinical data which will serve to fill a major gap related to better understanding their members’ complete health picture. Some of the key use cases we have seen include risk adjustment, quality (NCQA HEDIS Measures/CMS Star Rating Program), analytics, medical necessity/medical review, pre-authorization, care management, and population health.
8. Wolters Kluwer was recently in the news for expanding its cNLP portfolio. Can you lead us on to that?
We have spent 20 years working with hundreds of the clinicians using our clinical interface tool that is integrated into EHRs to build, expand, and continuously refine our clinical lexicon of provider-friendly terminology. We have combined that comprehensive foundation with advanced machine learning to build a highly accurate NLP engine that extracts clinical insights from medical charts.
Our latest release in the news is our next step on the journey: the addition of Chart Review Accelerator, which targets such payer use cases as risk adjustment, quality, analytics, and medical review, further enabling a holistic view of the member. Along with that, we have built clinician workflows to expand the value of cNLP to a wide range of interested stakeholders from payers, life insurance organizations, providers, and HIT vendors who utilize chart data.
9. What advice would you like to give the upcoming healthtech companies?
First, I would say perform your research to really understand the market requirements you intend to address with what you plan to commercialize.
It is a common mistake to let feedback from your latest customer or prospect to overly-influence your roadmap rather than allowing your research to inform and drive product development.Lastly, it is critical to be able anticipate buyers’ evolving needs and the inevitable market shifts, so it is important to learn to fail fast and always try to “skate to where the puck will be.”
10. What movie inspires you the most?
One of my favorites is Lord of the Rings. Out of the many messages, the one I like is: “Even the smallest person can change the course of history.”
11. We have heard that you have a very joyful work culture, we won’t mind having a look at some of the pictures?
Due to COVID-19, the majority of Wolters Kluwer employees have been working remotely for the past year plus. The Health Language team has not been able to gather together since last March, so we’re making the most of our virtual tools to stay connected.
And as you can see from this next picture, in those rare cases where some of us are able to see each other in person, we are thrilled to be together (safely)!
For more such updates and perspectives around Digital Innovation, IoT, Data Infrastructure, AI & Cybsercurity, go to AI-Techpark.com.
Frank Jackson
Vice President and GM at Wolters Kluwer
Frank Jackson is the Vice President and General Manager at Wolters Kluwer, Health. With more than 25 years of healthcare experience, Jackson has a proven track-record of developing partnerships within complex technical and data environments.
