News & Updates

May 15, 2023 Research Breakthroughs

New Study Reveals Insights on Post-Marketing Drug Surveillance

A recent study published in the Journal of Pharmacoepidemiology highlights innovative approaches to post-marketing surveillance of novel therapeutics. The research, conducted by an international team of investigators, developed and validated a machine learning algorithm capable of identifying potential adverse drug reactions from electronic health records with improved sensitivity compared to traditional methods.

The study analyzed over 2 million patient records across three different healthcare databases to detect signals for a diverse set of medications, including recently approved biologics and small molecules. Results showed a 30% increase in signal detection efficiency while maintaining specificity comparable to conventional approaches.

April 10, 2023 Regulatory Updates

Regulatory Authorities Update Guidelines on Real-World Evidence

The FDA and EMA have published updated guidance on the use of real-world evidence (RWE) in regulatory decision-making processes. The new guidelines, released simultaneously by both agencies, provide a harmonized framework for evaluating the quality and reliability of real-world data sources and study designs used to generate evidence for regulatory submissions.

Key updates include specific considerations for different data sources (claims, electronic health records, registries, and patient-generated data), methodological approaches for addressing common biases in observational studies, and expectations for documentation and transparency. The guidelines also outline scenarios where RWE may be considered sufficient for regulatory purposes, including rare disease indications and certain post-approval requirements.

March 22, 2023 Methodological Advances

Novel Approach for Handling Time-Varying Confounding in Longitudinal Studies

Researchers at the Center for Pharmacoepidemiology Research have developed a new statistical approach for addressing time-varying confounding in longitudinal studies of medication effects. The method, described as "Dynamic Marginal Structural Networks," extends traditional marginal structural models by incorporating machine learning algorithms that adaptively identify and adjust for complex time-dependent confounding patterns.

In simulation studies and a practical application examining incretin-based therapies and cardiovascular outcomes, the approach demonstrated superior performance compared to conventional methods, particularly in settings with numerous potential confounders and complex treatment patterns.

March 5, 2023 Technology & Tools

OHDSI Releases Enhanced Suite of Pharmacoepidemiological Tools

The Observational Health Data Sciences and Informatics (OHDSI) community has announced the release of a substantially upgraded suite of open-source tools for conducting pharmacoepidemiological studies using common data models. The update includes enhancements to the ATLAS web-based interface, HADES R package ecosystem, and underlying database infrastructure.

Notable new features include automated quality control checks for phenotype definitions, advanced methods for propensity score estimation and matching, distributed regression capabilities for multi-site studies, and improved visualization of study results. The tools maintain their commitment to reproducibility, transparency, and collaborative research across a global network of data partners.

February 18, 2023 Education & Training

Launch of Global Pharmacoepidemiology Certification Program

A consortium of leading academic institutions has announced the launch of a comprehensive Global Pharmacoepidemiology Certification Program designed to address the growing demand for specialized expertise in this field. The program, developed in collaboration with industry and regulatory partners, offers a structured curriculum covering foundational principles, advanced methodologies, and practical applications of pharmacoepidemiological research.

The certification consists of online modules, virtual workshops, and a capstone project supervised by experienced mentors. Designed for flexibility, the program accommodates working professionals while ensuring rigorous training and assessment. Scholarships are available for participants from low and middle-income countries to promote global capacity building in medication safety research.

January 30, 2023 Research Breakthroughs

International Collaborative Study Confirms Vaccine Safety Profile

An international network of research institutions has completed one of the largest post-authorization safety studies of COVID-19 vaccines to date, analyzing data from over 30 million vaccinated individuals across 12 countries. The study, which employed a common protocol implemented across different healthcare databases, evaluated the incidence of rare adverse events following immunization with various vaccine platforms.

The findings, published in The New England Journal of Medicine, provide precise risk estimates for several safety outcomes of interest, confirming the favorable benefit-risk profile of authorized vaccines. The study also demonstrated the value of international collaboration and distributed data networks for rapidly generating robust evidence on vaccine safety.

January 12, 2023 Regulatory Updates

Regulators Announce Framework for Collaborative Pharmacovigilance Activities

Major pharmaceutical regulatory authorities including the FDA, EMA, PMDA, and Health Canada have announced a new framework for collaborative pharmacovigilance activities. The initiative aims to enhance efficiency and consistency in post-marketing safety surveillance through coordinated signal detection, data sharing, and joint assessments of emerging safety concerns.

Key elements of the framework include harmonized definitions and methods for safety signal management, protocols for secure exchange of pharmacovigilance data, and mechanisms for aligning regulatory actions across jurisdictions. The collaborative approach is expected to reduce duplication of efforts, facilitate more rapid identification of safety signals, and promote global consistency in benefit-risk assessments.

December 8, 2022 Technology & Tools

Breakthrough in Privacy-Preserving Record Linkage for Health Data

Computer scientists and health data researchers have developed a novel approach for privacy-preserving record linkage (PPRL) that allows accurate integration of patient-level data across multiple sources without compromising confidentiality. The method, which combines advanced cryptographic techniques with machine learning algorithms, achieves linkage accuracy comparable to traditional methods while providing formal privacy guarantees.

This breakthrough addresses one of the most significant challenges in pharmacoepidemiological research: combining complementary data sources (such as claims, electronic health records, and registries) to create more comprehensive patient profiles while respecting privacy regulations and ethical considerations. The approach has been successfully implemented in a pilot project linking pharmaceutical claims with hospital outcomes data across three healthcare systems.