Rheumatologic diseases like systemic lupus erythematosus (SLE), ankylosing spondylitis, and rheumatoid arthritis can present significant challenges for biopharma. These conditions are chronic, complex, and multisystemic. While traditional clinical trials provide essential insights, they often fail to capture the full spectrum of patient experiences or reflect the diversity seen in real-world populations.
Real-world data (RWD), combined with advanced AI technologies, offers a transformative way to bridge this gap and map patient journeys in rheumatology from pre-diagnosis through disease management and into health outcomes.
Case Study: Using RWD and AI to Facilitate a Deeper Understanding of Lupus Across the Patient Journey
Prediagnosis Stage: Facilitating earlier disease detection
Lupus is a complex disease that is often difficult to diagnose. In early disease stages, patients often present with variable clinical manifestations or nonspecific symptoms. On average, it takes almost six years from symptom onset for lupus patients to receive a diagnosis, and patients often face delayed referrals to rheumatologists, which can lead to negative long-term outcomes.
AI-powered digital phenotyping solutions, such as OM1 Orion, can help enable earlier detection of disease. Digital phenotypes are built from complex signals and interactions shared by patients with similar conditions, characteristics, or outcomes that can help distinguish these patients from others. OM1 Orion is trained using OM1’s repository of linked EMR, claims and other data covering more than 300 million patients. The richness of the source data lets OM1 Orion capture many complex facets of the rheumatology patients’ journeys and isolate patients of interest.
Diagnosis Stage: RWD for rapid exploration of SLE phenotypes
Rheumatologic conditions are rarely confined to a single part of the body, and patients typically have a wide range of comorbidities. For instance, lupus patients are at an increased risk for cardiovascular disease, and rare manifestations like lupus nephritis can lead to serious complications. However, clinical trials for lupus often have extensive inclusion and exclusion criteria, which may prevent patients from being well represented or represented at all in clinical trials — limiting the generalizability of findings to broader patient populations.
RWD and AI can provide biopharma companies with the tools to better identify and understand patient subtypes in greater depth. This capability is especially valuable for conditions with high variability, like lupus, where subtypes can inform more targeted and effective treatment strategies. By allowing AI algorithms to analyze patterns in clinical data, biopharma companies can gain actionable insights into disease trajectories and patient responses over time.
Beyond patient stratification for quick insights, RWD and AI can be used to improve the understanding of heterogeneity in lupus and explore questions such as:
- Widespread variation in clinical presentation, such as differences between men and women (women make up about 9 out of 10 lupus patients)¹
- Treatment patterns for patients with specific comorbidities
- Specialties of providers caring for SLE patients as they move through pre-diagnosis to diagnosis and treatment
Post-diagnosis and Ongoing Care: Capturing the Complexity of SLE
In rheumatology and chronic conditions – where patient journeys often switch between highly active disease states to relapses and remissions – it’s critical to understand the patient journey over time fully. This requires adequate breadth and depth of real-world datasets. Large real-world datasets provide key context into the early patient journey, such as pre-referral into rheumatology and the management and consequences of comorbidities like increased cardiovascular disease risk in lupus populations. Robust RWD that includes relevant specialist documentation and multimodal data can provide better insights into key outcomes and treatment effectiveness.
OM1 can achieve this breadth and depth on rheumatology conditions like SLE with our highly curated datasets of networks, linking specialty EMR data from rheumatologists with multimodal data including:
- Unstructured data, including clinician-reported outcomes, extracted with AI and machine learning: Provides insights into key outcomes, history of present illness, treatment plans, and reasons for switching and discontinuation to understand how patients are being managed and how decisions are being made in real-time.
- Open medical and pharmacy claims: Ensures longitudinal follow-up of patients across different care settings.
- Closed claims: Provides insights into healthcare costs and captures the totality of the healthcare encounters for a given patient over a shorter amount of time.
- Augmented death data: Analyzes differences in mortality after patients access different treatment modalities.
- Social Determinants of Health (SDoH): Evaluates potential effects on effectiveness, safety, treatment choices, and discontinuation.
Additionally, rheumatologic diseases like lupus require precise, consistent tracking of disease activity, but obtaining this data can be challenging. In lupus, measures such as SLEDAI scores are vital for monitoring disease progression, yet they are often absent from routine clinical documentation.
OM1 applies advanced AI algorithms to estimate vital disease activity scores in rheumatology, such as CDAI, BASDAI, and SLEDAI, from unstructured clinical notes. This capability significantly increases the availability of critical data points, enabling robust longitudinal analysis and allowing researchers to monitor disease activity over the full patient journey.
Machine learning models can be successfully used to estimate disease activity scores. OM1 generated 30x more SLEDAI scores for longitudinal analysis. Read the research study »
By leveraging real-world data and AI, biopharma companies can gain a more comprehensive understanding of rheumatologic diseases like lupus, enabling earlier detection, precise patient stratification, and ongoing monitoring of disease activity. Comprehensive and integrated real-world data can empower researchers to capture the full patient journey in ways that traditional clinical trials cannot, supporting more personalized and effective treatment strategies for complex, chronic conditions.
About OM1 Specialized Rheumatology Real-World Data
OM1’s rheumatology RWD network includes over 700 rheumatologists, following patients longitudinally in over 1,200 specialty practices in all 50 states. Our specialized datasets in rheumatology include rheumatoid arthritis, systemic lupus erythematosus, psoriatic arthritis, ankylosing spondylitis, and more.
Contact us to discover how OM1’s AI-driven evidence and real-world data in rheumatology can supercharge your research goals.
¹https://www.lupusresearch.org/understanding-lupus/what-is-lupus/about-lupus/