Nam Nguyen, Vice President of Medical Informatics and Data Science at Veradigm, recently participated in a panel discussion hosted by Datavant on 1/26/22, concerning one of the most crucial concerns in today’s healthcare landscape: How can we use innovative technology platforms to improve diversity in clinical trials and clinical research, as well as enhance the type and quality of data obtained during patient visits?
The webinar brought together a panel of industry experts, all of whom recognize the significant gaps in the level of diversity in clinical research and are dedicated to overcoming such barriers through enhancing our use of health information technology platforms.
Mining and analyzing real-world data directly from the point of care
Some healthcare technology companies can deliver real-world clinical data, others work to turn real-world data into insights, and others serve providers directly with point-of-care software. In the panel discussion, Nam remarked, “Veradigm does it all in our unique position at the center of healthcare. We drive value through our scalable network called the ‘Veradigm Network,’ which brings advanced capability and data-driven solutions to help transform healthcare.”
He further explained that the Veradigm Network addresses diversity “by linking Veradigm data with our partners’ data sources to improve both data content and geographic representation. If we don’t have the data points for the patient representation that the study needs, we can look to our network partners to link to and fill in those gaps.”
Nam noted that the geographic and demographic diversity within the Veradigm Network of healthcare providers can help promote diversity and inclusion in clinical research. Veradigm, an Allscripts business, provides Veradigm with direct access to data from Allscripts’ EHRs and their partners, such as Allscripts Touchworks EHR, Allscripts Professional EHR, and Practice Fusion. “With this bidirectional data exchange through our network,” Nam explained, “we can provide real-world data for both retrospective and prospective studies directly from the point of care,” providing an end-to-end solution that brings timely data from the point of care directly to research.
Due to its direct access to clinicians, through their EHRs, provides Veradigm access to unstructured data, which include clinical notes, attachments, and images. He stressed that “The real-world data studies using this data have better representation of populations via geography and a more diverse population of patients [from which] to base studies.”
Prospective research powered by the Veradigm Network
Direct access to data from the point of care enables Veradigm to “target and recruit sites and patients for prospective studies at scale” within the Veradigm Network, providing investigators with access to novel sites from which recruitment doesn’t traditionally take place. Through this access, patient information is not disclosed, and HIPAA laws are adhered to. Although clinical trial recruitment is largely done through large academic medical centers and hospitals, “adding Veradigm to your recruitment strategy,” he explained, “gives you access to our network of solo practitioners and other small, medium, and large practices from rural, suburban, and urban settings, (increasing) your access to a more diverse population of patients.”
Research-ready real-world data
Veradigm curates its real-world data “and makes it what we call ‘research ready,’” Nam explained. “For those who have used EHR data before, it can be messy, inconsistent, and difficult to use.” Veradigm’s team of clinicians and informaticists curate and harmonize the real-world data, making it fit-for-purpose for research. The data is longitudinal and timely, with some data just weeks’ old, since it’s coming directly from the point of care. Our EHR connection enables us to “prospectively capture data, including social risk factors, directly from the providers at point of care through our EHRs and other technologies, Nam emphasized, “so our prospective data capture can potentially gather social risk data gaps directly from sites.”
EHR data enrichment to mitigate socioeconomic risk
Veradigm can use Natural Language Processing (NLP) to mine data from free text to integrate and enhance research-ready data sets. Nam explained that useful information is included in unstructured narrative clinical notes. Researchers can only obtain such information from a data partner who has access to the source. He noted that “We have done NLP for a variety of therapeutic areas … Bone mineral density values went from barely 100 data points to over 80,000 values using NLP. … And in terms of social risk factors, we were able to improve data points on intimate partner abuse from 13.5 thousand data points to over 300,000 using NLP.”
Preserving social risk data to fit research goals
While real-world data can help mitigate some of the issues associated with inclusion in clinical research, Nam also stressed that social risk data may be erased through the de-identification process. He gave the example of racial minorities such as Native Americans that are generalized to the “Other” category during the de-identification process. “So even if you have permission to use identified patient data in your study,” he noted, “if you link to a partner that only can provide de-identified data, you may not get the granularity and social risk data like race that you were hoping for.” Yet with Veradigm’s access to source data and their control over the de-identification processes, they can compliantly customize data sets to fit research goals, particularly when including sensitive data, like social risk factors, in studies.
Veradigm case studies
Nam concluded by reviewing several Veradigm studies that provided real-world examples of steps to increase representativeness in clinical research. Veradigm is working with academic partners to understand which social risk factors clinicians are capturing in the EHRs today and how they are doing so.
The first study he described “created a framework for extracting and reporting social risk factors from ambulatory EHR data … (providing) an open-source Python dictionary to facilitate reporting and extraction for social risk factors that anyone can download.” This research, Creation of a Mapped, Machine-Readable Taxonomy to Facilitate Extraction of Social Determinants of Health Data from Electronic Health Records, was published at the American Medical Informatics Association (AMIA)’s 2021 Symposium and received a Distinguished Paper award at AMIA in 2022.
In a second study, Veradigm is using the same framework to describe the social risk factors they identified in structured and semi-structured data from 10 years of EHRs across three different EHR solutions. With the third study, Nam explained that they will assess “physician notes, chief complaints, and other unstructured data to see what additional social risk data we can find to complete our understanding of the current state of social risk factor documentation.”
Want to learn more from Nam and the other panelists on this important topic? Click here to view the webinar recording.
Contact us if you would like to learn more about how Veradigm can help promote diversity in your clinical research.