Novo Nordisk Strategic Framework for Fit-for-Purpose Real-World Data

At the 2026 SCOPE Summit, Tom Dougherty, Real World Data Strategy Lead at Novo Nordisk, presented a comprehensive roadmap for navigating the rapidly expanding landscape of real-world data (RWD). As the pharmaceutical industry’s investment in RWD reaches an estimated $6 billion to $9 billion in 2024, the focus has shifted from simple data acquisition to the strategic selection of fit-for-purpose assets. The session underscored that while nearly 88% of mid-to-large biopharmaceutical companies now utilize RWD for clinical trials and health economics and outcomes research (HEOR), the primary challenge remains bridging the gap between vast data availability and high-quality, actionable evidence.

The Evolution of the RWD Ecosystem

The clinical research environment is witnessing a fundamental shift in how data is perceived and utilized across the drug development lifecycle. The panel indicated that the market for RWD is no longer confined to traditional safety and pharmacovigilance; it has expanded into comparative effectiveness, market access, and the design of external control arms. This momentum is supported by a significant shift in the regulatory landscape, with bodies such as the FDA, EMA, and various national Health Technology Assessment (HTA) organizations issuing new guidances and pilots to increase the acceptance of RWE.

The discussion highlighted that approximately 45% to 60% of research programs now report measurable business value from RWD integration. Specifically, in the clinical trial space, the use of strategic data assets has led to a 15% to 35% reduction in typical enrollment times. However, the ecosystem remains bifurcated between tried and true sources—such as Electronic Health Records (EHR) and insurance claims—and emerging streams including genomics, wearables, and direct-to-patient data. The challenge for modern data strategists is to serve as a matchmaker, funneling these disparate sources through a rigorous framework to ensure the data aligns with specific research objectives.

Navigating Claims and Clinical Data Architectures

A central theme of the session was the differentiation between various categories of RWD and their specific utility in evidence generation. The discussion categorized insurance claims into closed and open systems, each offering distinct advantages and limitations. Closed claims, sourced directly from payers, provide comprehensive, adjudicated records that include actual approved costs and longitudinal tracking. These are essential for HEOR and medication adherence studies. Conversely, open claims, captured via practice management systems, offer near real-time data on nearly 300 million U.S. lives, making them ideal for monitoring the immediate market uptake of newly launched therapies, despite the lack of adjudication.

Beyond claims, EHR data remains a cornerstone for clinical depth, particularly through the use of Natural Language Processing (NLP) to extract insights from unstructured clinical notes. This methodology allows researchers to understand complex patient behaviors, such as the reasons behind treatment discontinuation or switching, which are rarely captured in coded data. One speaker illustrated that while EHR data is rich, its utility is often hampered by varying degrees of data completeness and limited generalizability. Therefore, the strategic linking of these sources through tokenization—connecting lab results, mortality data, and social media insights—is becoming the industry standard for constructing a holistic view of the patient journey.

Objective Frameworks for Data Fitness

To move beyond subjective preferences in data selection, the presentation introduced a ten-pillar framework designed to assess data fitness. This objective approach evaluates potential data partners across dimensions such as reputation, coverage, accessibility, and quality. A key component of this framework is the distinction between data validity (accuracy of the real-world representation), reliability (consistency over time), and representativeness (applicability to the entire population). Experts suggested that the industry must move toward objective measures to satisfy regulatory inquiries regarding why specific data sources were selected for a study.

The framework emphasizes the following criteria for high-stakes research:

• Regulatory Provenance: Documentation of previous successful regulatory submissions using the data source.
• Granularity and Depth: The fill rate and missingness of key variables, such as Body Mass Index (BMI) or specific biomarkers, which are critical for therapeutic areas like obesity and oncology.
• Technical Interoperability: The ability to ingest raw data into local environments or access it through global cloud platforms while adhering to strict privacy standards like HIPAA and GDPR.
• Quality Pilots: The rising use of no-cost pilots (typically 30 to 60 days) to evaluate data ranges and cleanliness before committing to a long-term subscription.

Future Outlook: AI and Global Standardization

The future of RWD is increasingly tied to the adoption of common data models and the integration of artificial intelligence. While the discussion suggested that the industry is still in the early stages of leveraging Large Language Models (LLMs) for complex data modeling, AI is already proving valuable for low-hanging fruit tasks such as document generation and accelerating data analysis. The panelists emphasized that as the FDA continues to accept more RWE-based submissions, there will be a broader move toward standardizing RWD into common models like OMOP to facilitate cross-border data collection.

Looking ahead, the internationalization of RWD presents both a significant opportunity and a legal challenge. For global organizations, the requirement to navigate varying data privacy laws—where U.S. tokenization may conflict with European double de-identification standards—remains a barrier to a truly unified global data platform. However, the movement toward more transparent, patient-consented data from wearables and social media suggests a shift toward more inclusive and diverse evidence generation. The session concluded that while the sheer volume of available data is daunting, the implementation of rigorous, objective selection frameworks will be the primary driver of transformation in clinical research and patient outcomes.