Post Interoperability July 17, 2024 Real-world data in life sciences: A starter’s guide to new FDA guidances A recent McKinsey analysis reveals that over the next three to five years, top pharmaceutical companies could unlock more than $300 million a year by adopting RWE across its value chain. Yet, as noted by FDA Commissioner Robert M. Califf, MD, realizing this potential is contingent on critical factors, such as ensuring fit-for-purpose and maintaining data quality with robust data infrastructure. As life sciences companies navigate the challenges and opportunities presented by RWD and RWE, alignment with the rigorous standards set forth by the FDA is critical. The FDA’s view on RWD and RWE The FDA has released several final and draft guidance documents about the use of RWD and RWE in regulatory decision-making, some of which we list below: Real-World Evidence: Considerations Regarding Non-Interventional Studies for Drug and Biological Products (Draft: March 2024) Real-World Data: Assessing Registries To Support Regulatory Decision-Making for Drug and Biological Products (Final: December 2023) Use of Real-World Evidence to Support Regulatory Decision-Making for Medical Devices (Draft: December 2023) Considerations for the Use of Real-World Data and Real-World Evidence To Support Regulatory Decision-Making for Drug and Biological Products: Guidance for Industry (Final: August 2023) These FDA guidance on RWE underscore the importance of relevance and reliability and provides a framework for guiding regulatory submissions. The practical summary below highlights how life sciences companies can design RWE programs that meet regulatory standards while advancing actionable insights. 2 Pillars of robust RWD programs amidst recent FDA guidance Relevance: Fit-for-purpose Start with a clear research question. The success of an RWD program largely depends on the precision and feasibility of its research questions. These questions should guide the data collection strategy, ensuring that every piece of data serves a defined purpose within the research framework. For example, “assessing a diabetes medication’s effectiveness in reducing Hba1c in older patients” or “the impact of wearables on hospital readmission rates in heart failure patients” are targeted inquiries that ensure data collected is relevant and the findings actionable. These well-defined questions streamline data collection and ensure that insights are directly applicable to the healthcare problems being addressed. Select the right data sources. For effective RWD programs, align data collected with the specific needs of healthcare questions or decisions to ensure impact. Data sources may include: electronic health records (EHRs) claims data patient registries wearables and sensors patient-generated data from surveys to track real time patient experiences biosamples For example, the PROGRESS study aimed to evaluate the impact of dietary intake on glucose response using data collected from surveys (health, lifestyle, and nutrition intake), digital endpoints, electronic health records (EHR), and biosamples. Reliability: Ensuring data integrity Data reliability ensures that the data consistently and accurately reflects the real-world scenarios it aims to represent. The following can enhance the reliability of RWD: Standardizing data collection: It’s crucial to standardize data collection across different sources like electronic health records (EHRs), claims data, patient registries, and patient-generated health data (PGHD). This includes: setting common data formats and protocols determining the frequency of data capture establishing standards for accuracy and completeness ensuring interoperability between devices, apps, and other healthcare systems These are essential for seamless data integration and utility in health research and decision-making. Assessing data quality: Maintaining high data integrity involves: Auditing data completeness, accuracy, and timeliness. Implementing robust audit trails and metadata management to track all data modifications and ensure long-term reliability. Automating quality checks on data from wearables and other sources to identify and correct anomalies or inconsistencies. Educating patients on the correct use of wearables and data reporting to ensure there is effective data collection process, and enhanced data quality and reliability. Summary of the 2 Pillars of robust RWD programs Relevance Ensuring data is applicable and useful for specific healthcare questions Data source selection Electronic Health Records (EHR) Claims data Patient-generated health data (PGHD) Wearables, sensors, and digital health technologies Representative patient populations Diversity in sociodemographics Variety in health conditions, medical histories Alignment with research objectives Specific healthcare questions Tailored data analysis Reliability Ensuring data consistently and accurately reflects real-world scenarios Data accuracy Correct data entries Error detection Data completeness Comprehensive data capture Gap analysis Data provenance Traceable data sources Clear data origin Data traceability Easily retrievable and verifiable 4 Practical steps for implementing robust RWD programs Develop a framework: Establish a clear strategy that sets the study goals and key research questions, and details the data sources and methods. Draw on active projects as examples to guide workflows and implementation strategies, like those seen in studies such as PROGRESS, which uses multi-modal data collection to build predictive models. Leverage technology: Employ advanced analytics, AI, and machine learning to handle and analyze large datasets effectively. Widen expertise: Collaborate with experts proficient in the technical and regulatory aspects of RWD to navigate clinical data challenges, such as data gaps and variability. Ensure compliance: Adhere to legal and regulatory standards, including GDPR and HIPAA, and follow guidelines from authorities like the FDA. Ensure ethical practices by obtaining informed consent, maintaining transparency in data use, protecting patient anonymity with effective data de-identification, and using RWD only as consented to by data subjects. This is crucial for preserving the integrity of your research and safeguarding participant rights. Maximizing impact with RWD Prioritizing relevance and reliability is a key first step for life sciences companies planning out RWD/RWE programs. Meticulously selecting data sources, maintaining data integrity, and aligning with regulatory expectations further unlock RWD’s potential to drive meaningful insights.
