P209: Signal Management (SM) Activities using Real-World Evidence (RWE)
Poster Presenter
MIDHUN RAJ K
Associate Director, Signal Management
IQVIA India
Objectives
PV uses various data sources (solicited and unsolicited reports) for signal management activities which may lack contextual information, that real-world data can supply. This is an effort to discover different real world data sources and its applications on improving safety monitoring activities.
Method
Various sources of RWE and its application on SM were reviewed using targeted literature search, including electronic health records (patient data), claims (large samples), patient registries (specific conditions), social media (qualitative data) and digital health technologies (real-time data).
Results
Healthcare databases are valuable for epidemiological studies but have limitations for hypothesis-free signal detection due to costs, access issues, and delays. RWE collects data from diverse sources like healthcare databases, registries, claims databases, mobile devices, social media, and patient platforms. Text mining and natural language processing help transform unstructured text into structured data in healthcare databases, mostly includes Electronic Health Records (EHRs) which provide comprehensive patient information, including medication history, diagnoses, lab results, and outcomes, useful for identifying potential safety signals. Claims data, despite showing phenomena rather than causal relationships, offer advantages like national representation and reduced selection bias. Registry data provide insights into prescribing practices and long-term follow-up reveals delayed risks. Social media platforms offer patient perspectives on health topics. Wearables and digital health technologies provide daily health insights. Multiple methods are needed for analyzing data from various sources, including cohort studies, propensity score matching, self-controlled case series, Bayesian approaches, and more. Longitudinal datasets yield accurate risk estimates for adverse drug reactions. Data-mining methods like sequence symmetry analysis (SSA) and tree-based scan statistics software are used for proactive screening. The European Medicines Agency (EMA)’s DARWIN EU platform and the United States Food and Drug Administration (FDA)'s Sentinel system enhance drug safety surveillance using RWE, while guidelines are being developed to standardize pharmacoepidemiologic studies. RWD may not replace spontaneous reporting systems (SRS); both sources complement each other. SRS remains useful for detecting rare events early after launch, while RWD may be better for events requiring longer induction periods or acute cases, especially with self-control case series.
Conclusion
Utilizing real-world healthcare databases for signal detection in pharmacovigilance has the potential to enhance current activities and potentially replace outdated methods. With improved accuracy and efficiency, this hypothesis -free signal detection allows discovery of previously unknown safety concerns with broader, relevant, real time, longitudinal and timely data. With its continuous flow from real-world sources enables real-time monitoring of drug safety, allowing for early detection and intervention in cases of emerging safety concerns. Outputs from large-scale approaches may contain more false positives, requiring routine signal management and clinical review. Data mining methods uncover hidden patterns and relationships in large datasets that might indicate potential drug-related adverse events. These methods go beyond manual review, providing early warnings and enhancing the efficiency of identifying potential safety concern. Various health authorities initiated active surveillance network with real world evidence, however, needs to collaborate multiple stakeholders globally in integrating the real-world evidence sources and process for its evolving needs and ensure its long-term sustainability. Signal detection in Real world data holds promise for improving pharmacovigilance, and collaboration is essential to advance this field and ensure patient safety.
