Beyond ICD Codes: How Real-World EHR Data Is Advancing Ophthalmic Research
Author:
Verana Health
For basic disease categorization and for billing and reimbursement purposes, the International Classification of Diseases (ICD) codes are an effective tool for the job. In ophthalmology, ICD-10 codes (H00-H59) do yeoman’s work, providing a standardized framework for classifying and coding diagnoses, symptoms, and procedures.
But when it comes to novel research and capturing the full spectrum of patient experiences and clinical nuances of complex ophthalmologic diagnoses, ICD codes simply aren’t up to the task.
To gain a more comprehensive understanding of ophthalmic conditions, it’s necessary to dig deeper and glean insight from unstructured clinical notes and details contained in electronic health records (EHR).
It’s here, in the realm of unstructured data, that researchers find the insights and evidence needed to drive discovery, advance the drug and development process, and deliver important improvements to healthcare.
Long considered inaccessible – at the least, impractical to access – unstructured data can now be efficiently wrangled and curated by skilled teams with the help of artificial intelligence (AI) tools. And the potential benefits to clinical researchers and patients are profound.
The Risks of Going (with ICD Codes) Alone
The shortcomings of ICD codes at capturing the details of complex ophthalmic conditions pose real challenges to researchers. Relying solely on ICD codes to gather real-world evidence for drug development can lead to a number of problems, including:
- Missing the differentiation of stages or types of conditions: ICD codes often group diseases broadly; limited patient stratification makes it difficult to identify subgroups, leading to less effective treatments for certain individuals.
- Underrepresenting comorbidities: Many eye conditions are linked to systemic diseases, such as diabetes and hypertension. ICD codes don’t always capture the complexity of comorbidities, which can obscure how a new drug interacts with other health issues.
- Overlooking adverse events: While ICD codes indicate the presence of certain conditions, they often lack important details about the severity or nature of adverse effects. This may contribute to underreporting and a limited understanding of a drug’s safety profile.
- Skewing of drug development priorities: A focus on ICD codes may steer research toward more common conditions, discouraging innovation and treatment for rare or complex ophthalmic diseases and patients most in need.
- Difficulty gaining regulatory approval and reimbursement: Since ICD codes are closely tied to billing and reimbursement, a new drug that doesn’t align well with existing codes may face significant hurdles in approval and reimbursement.
Power Up: Pairing ICD Codes and EHR Data
To avoid the limitations of ICD Codes and structured data alone, it’s important for researchers to leverage the detailed information contained in unstructured EHR data, specifically in clinical notes.
With detailed views of symptoms, disease severity, diagnostic tests, and treatment outcomes, clinical notes add important clarity to insights from ICD codes, detail that is essential for accurately diagnosing patients in complex fields like ophthalmology.
Combined, ICD codes and clinical notes comprise a more comprehensive approach to research and patient care and allow researchers to address more complex inquiries.
But how exactly can researchers tap the insights of difficult-to-parse unstructured EHR data combined with structured data such as ICD codes?
Working with experienced, multidisciplinary teams of specialists, Verana Health applies advanced AI-tools, including natural language processing (NLP) and machine learning (ML), to extract critical, de-identified information from unstructured clinical notes, as well as ophthalmic images and exam documentation.
These efforts can uncover patterns and clinical insights, such as disease progression markers, and identify specific conditions and patient characteristics that ICD codes might miss or for which a single ICD code does not exist.
This vital information, combining structured and unstructured data, is compiled in Verana Health Qdata® modules. Ophthalmology Qdata is exclusive curated data from the American Academy of Ophthalmology IRIS® Registry (Intelligent Research in Sight), which aggregates de-identified data from 80 million patients.
Enriched with key variables, such as diagnosis confirmation, lesion location, visual acuity (VA), central subfield thickness (CST), and intraocular pressure (IOP), these datasets facilitate both clinical and commercial research endeavors.
Big Time Benefits: Combined Structured and Unstructured RWD Reveal a Path Forward
What is the value of research-ready, fit-for-purpose datasets?
Leveraging unstructured EHR data along with structured ICD codes can unlock insights across the entire drug and device development cycle. This combined approach addresses the inherent limitations of traditional coding systems, and help researchers realize key benefits, including:
- Optimize trial design with a comprehensive view of patient populations
- Identify and stratifying patients more accurately for clinical trials
- Improve site selection by finding locations with high concentrations of relevant patients
- Understand real-world treatment patterns and outcomes for a deeper understanding of therapy performance
- Capture clinical characteristics of real-world patients, allowing tailored R&D efforts for actual patient needs
- Track real-world therapy usage, to the brand level
The expansive datasets and advanced analytical capabilities of Verana Health Qdata are revolutionizing ophthalmic research and RWE generation – and helping life sciences companies find a more comprehensive, nuanced understanding of ophthalmic conditions.
To learn more about how to accelerate clinical trials and gain insights with real-world evidence, download our white paper.
Let's Accelerate Research Together
To learn more about Verana Health, please fill out the information below and our team will follow up with you as soon as possible.