The IRIS Registry and its enduring impact on innovation

Recently, I had the honor of speaking at the Ophthalmic Business Council’s September meeting about the evolution of the American Academy of Ophthalmology (Academy) IRIS^® Registry (Intelligent Research in Sight) and its impact on medical innovation.

The IRIS Registry, established in 2014, is the largest specialty society clinical data registry in all of medicine. Verana Health is proud to be the Academy’s exclusive end-to-end data integration, curation and analytics partner for the IRIS Registry.

The amount of data contained in the IRIS Registry truly is astonishing: A longitudinal database covering more than nine years includes patient demographics, patient medical and ocular history, clinical examination findings, diagnoses, procedures, medications, and more. This de-identified data is amassed from more than 450 million patient visits, 75 million unique patients, 60 electronic health record (EHR) systems and 16,000 participating ophthalmologists and allied eye care providers across the U.S.

Verana Health’s role in enabling researchers to help turn this data into actionable insights that help support medical breakthroughs is critical.

Verana Health’s role in enabling researchers to help turn this data into actionable insights that help support medical breakthroughs is critical. Verana Health’s VeraQ^® population health data engine sits at the center. VeraQ ingests, analyzes, and curates structured and unstructured EHR data, including diagnostic images, contributed to the IRIS Registry and other specialty medical societies. This real-world data (RWD) is then often linked with other meaningful data sources, such as pharmacy claims, and medical claims.

This clinician-led process—enhanced and scaled by artificial intelligence (AI)—produces quality RWD that is used to help build disease-specific, fit-for-purpose data modules called Qdata^®. Ophthalmology Qdata spans diseases such as dry age-related macular degeneration, wet age-related macular degeneration (with unique Anti-VEGF variables), geographic atrophy, glaucoma (with unique MIGS variables), cataracts, and more.

IRIS Registry and Verana Health in practice

Verana Health’s VeraQ and Qdata are helping to drive business insights and inform research throughout the drug and medical device development lifecycles, from clinical trial site and subject identification to post-market evidence generation and market share tracking.

In one case, a sponsor was running a 30-site clinical trial for a new indication that is rare, complex, and had not yet been studied in a trial setting. The sponsor worked with Verana Health to help identify high-volume study sites to recruit this unique population for a phase II study for a novel ophthalmic indication, which had a complex patient journey and no existing ICD code. This meant that most patient characteristics were in unstructured EHR clinical notes, requiring clinically-led AI to be employed to meet the sponsor’s objectives.

Verana Health used machine learning (ML) and natural language processing (NLP) to help extract valuable and relevant information from text-based clinical notes. This innovative approach helped the sponsor to identify, qualify and onboard four high-volume study sites to recruit participants and start on time.

An ongoing evolution

The value of the IRIS Registry is continuing to expand beyond garnering insights from EHR data alone. This important clinical information is now being linked with claims and ophthalmic imaging data to help give a more comprehensive view of the patient journey. In 2021, for example, Verana Health and Komodo Health announced a partnership to integrate medical and prescription claims data from Komodo’s Healthcare Map™ into VeraQ.

Now, VeraQ is also integrating diagnostic images and linking them with existing EHR data that includes important demographic and clinical information from the IRIS Registry. This advancement requires novel ML and computer vision (CV) algorithms to bring meaning to the data at scale.

Recent Verana Health-led research demonstrated the ability of ML algorithms to automatically confirm geographic atrophy (GA) – an advanced form of dry age-related macular degeneration (AMD) – using real-world fundus autofluorescence (FAF) and infrared reflectance (IR) images linked to other EHR data. ML models were developed to assess FAF and IR image quality based on signal, contrast, noise and sharpness. The algorithms also were trained to differentiate eyes with and without GA.

The model demonstrated satisfactory performance confirming GA in AMD eyes using images collected in routine practice. This indicates the algorithm potentially could be useful in screening patients for GA trials and future GA treatments.

Quality RWD is key to transforming eye care

I have been part of Verana Health since its formation in 2018, as a co-founder, as its executive chairman of the Board, and as the immediate-past chief executive officer of the Academy. We are committed to transforming healthcare by helping to enable researchers and clinicians to inform quality insights and care through quality real-world data. With EHR data and diagnostic images contributed to the IRIS Registry as the foundation, enhanced with pharmacy and medical claims information, helping to improve the quality of life for patients through a data-driven lens is no longer a dream; it is a closer reality.

Dr. David W. Parke II, MD, is Verana Health’s Executive Chairman and served as CEO of the American Academy of Ophthalmology from 2009-2022.