For decades, medical claims and prescription databases have been the primary real-world data (RWD) sources tapped for life science and clinical research. This information detailing the visit date, demographics, procedure codes, and drugs dispensed is valuable given its ready availability and easy-to-analyze, structured format. However, these data sources lack clinical nuance, key variables and certain outcomes data that can be very beneficial for research. We recently hosted a webinar on this topic in collaboration with Xtalks. We invite you to watch it here: How to Generate Patient Journey Insights by Leveraging Real-World Data from Electronic Health Records.
Electronic Health Records (EHRs) are revolutionizing the RWD landscape
Now that nearly 90% of US hospitals and office-based physicians have adopted electronic health records (EHRs), these systems are becoming treasure troves of clinically rich information for research on de-identified, aggregated data. EHR data includes clinical nuance (including patient sentiment), and this information is often near real-time — entered and available at the time of the encounter. In addition to the demographic and procedure information available in claims data, EHR data provides additional clinical insights related to symptoms, diagnosis, disease progression and severity, treatment decisions, outcomes, and more. This detailed and timely information is improving the RWD landscape and the real-world evidence (RWE) it generates.
… nearly 80% of the valuable clinical data in EHRs is semi-structured or unstructured — contained in clinical notes, images, PDFs, and more.
The added value provided by EHR data also presents a series of new challenges. Unlike claims and prescription data, which is completely structured, nearly 80% of the valuable clinical data in EHRs is semi-structured or unstructured — contained in clinical notes, images, PDFs, and more. Bringing meaning to this data at scale requires it to be standardized for proper analysis, which often involves deep data engineering and data science coupled with clinical expertise.
Artificial Intelligence enables unstructured data extraction
Physician-informed algorithms can recognize patterns, contextual relationships, and relevant assertions in clinical notes to identify indications of disease, including details such as tumor stage, prostate-specific antigen (PSA) levels and Gleason scores for prostate cancer, and visual acuity and intraocular pressure (IOP) for glaucoma.
Today’s advanced artificial intelligence (AI) capabilities—namely machine learning (ML), and natural language processing (NLP)—are effective at helping to unlock insights from the unstructured data within EHRs. Physician-informed algorithms can recognize patterns, contextual relationships, and relevant assertions in clinical notes to identify indications of disease, including details such as tumor stage, prostate-specific antigen (PSA) levels and Gleason scores for prostate cancer, and visual acuity and intraocular pressure (IOP) for glaucoma.
Given the medical importance of the data being extracted, it is vital that any AI is executed with proper clinical oversight. This includes clinician guidance for algorithm development or feasibility, clinician labeling of data sets for ML training and model validation, and human abstraction of notes to build meaningful variables, data sets, and cohorts.
Specialty medical registries provide in-depth EHR RWD
Verana Health is focused on helping life sciences organizations derive quality insights from EHR data. The company has formed strategic data and analytics partnerships with the American Academy of Ophthalmology, the American Academy of Neurology , and the American Urological Association to curate and analyze their respective qualified clinical data registries (QCDRs), IRIS® Registry (Intelligent Research In Sight), which is the largest specialty society clinical data registry in all of medicine, as well as the Axon Registry®, and AQUA Registry. Collectively, the registries contain EHR data on about 90 million de-identified patients from more than 20,000 contributing clinicians.
VeraQ®, Verana Health’s population health data engine and AI-powered platform integrates with the more than 70 EHR systems contributing to these registries. Data from more than a half-billion healthcare encounters is ingested, de-identified, tokenized, normalized, and curated into high quality RWD, Qdata®, disease-specific, fit-for-purpose data modules designed to confidently drive business insights and inform research outcomes. Verana Health has more than a dozen Qdata modules that have launched or are in development, including Qdata Prostate Cancer in Urology/Oncology, Qdata Glaucoma in Ophthalmology, and Qdata SMA (spinal muscular atrophy) in Neurology. These data modules include both structured and unstructured inputs, often include linked data sources such as claims, and are guided with physician expertise to help ensure clinical context and nuance is reflected.
EHR RWD driving real-world life science research
Qdata is being leveraged to inform research across the drug lifecycle management spectrum – from clinical trial study design and pre-approval to commercialization and post-approval studies.
… Apellis Pharmaceuticals recently worked with Verana Health to conduct the largest retrospective study on Geographic Atrophy (GA) …
For example, Apellis Pharmaceuticals recently worked with Verana Health to conduct the largest retrospective study on Geographic Atrophy (GA), a chronic progressive degeneration of the macula. Currently, there are no approved therapies to treat, prevent, or limit the progression of GA. Apellis Pharmaceuticals wanted to analyze changes in visual acuity among 69,000 patients with GA over two years and determine the rate of progression to GA in patients with subfoveal involvement or exudative age-related macular degeneration (AMD). The research found that patients are three times more likely to develop a new onset of wet AMD in an eye with GA.
Ocular Therapeutix also recently collaborated with Verana Health for a postmarketing study to track endophthalmitis …
Ocular Therapeutix also recently collaborated with Verana Health for a postmarketing study to track endophthalmitis, a rare infection after cataract surgery. The condition has been historically difficult to study given its extremely low occurrence, so the FDA required Ocular Therapeutix to collect post-approval data relative to the incidence of endophthalmitis (within 30 days) for cataract surgery patients who received surgery at practices with access to its ReSure® Sealant product. The objective of the study was to determine the rate of endophthalmitis among the cohort of practices with and without access to ReSure. The RWE effort helped enable Ocular Therapeutix to forgo an expensive postmarketing clinical trial.
EHR RWD gaining acceptance as a complement or substitute source of clinical evidence to clinical trials
EHRs are infusing RWD with an unprecedented level of clinical information and insight. This growing level of detail is enabling RWD to not only supplement, but in some cases even bypass, traditional clinical trial control data. In fact, a recent FDA approval for Novartis Vijoice®, a treatment for select patients with PIK3CA-Related Overgrowth Spectrum (PROS), was based solely on an RWE retrospective chart review study of 57 patients. Another recent utilization of EHR data was in a replication of a phase 3 study in which Verana Health and the American Academy of Ophthalmology set out to explore whether real-world outcomes could be replicated in a retrospective simulation of the VIEW 1 and VIEW 2 (VIEW 1/2) studies. The acceptance of RWD and RWE as control data will likely continue to increase as extraction and use of EHR data becomes more sophisticated.
RWE is greater than the sum of its parts
Through our exclusive partnerships with leading medical societies and leveraging our AI-powered platform, VeraQ, Verana Health is able to harness comprehensive EHR data from qualified clinical data registries to generate quality RWD. Linking this detailed, timely, and clinically nuanced data with data from claims, prescription repositories, imaging and more provides a more robust view into patient journeys to inform life sciences research. This combined RWD that informs RWE holds immense potential to transform healthcare and life sciences research and improve outcomes for patients.
To learn more about the evolution of EHR RWD, watch the Xtalks webinar, How to Generate Patient Journey Insights by Leveraging Real-World Data from Electronic Health Records.