ICH M14 implementation advances as FDA publishes final guidance

Continuing its series of regulatory guidance updates, the US Food and Drug Administration (FDA) has announced the availability of the final ICH M14 guidance, which sets out internationally harmonized principles for planning, designing, analyzing, and reporting non-interventional studies using real-world data (RWD) to evaluate the safety of medicines.

Titled ‘General Principles on Planning, Designing, Analyzing, and Reporting of Non-interventional Studies That Utilize Real-World Data for Safety Assessment of Medicines’, the document was developed by the International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use (ICH) and is intended to support globally aligned approaches to the use of RWD in post-marketing safety studies.

ICH brings together regulators and industry to develop consensus-based technical guidelines that align regulatory expectations across regions. As a founding regulatory member, the FDA contributes to their development and adopts them as guidance for industry. According to the guidance, its purpose is:

“To recommend international standards for, and promote harmonization of, the general principles on planning, designing, analyzing, and reporting of non-interventional studies that utilize fit-for-use… data for safety assessment of medicines.”

The guideline was adopted by ICH in September 2025 and has now entered Step 5 implementation, during which harmonized ICH guidelines are adopted and implemented by regulatory members and observers within their respective regions. The European Medicines Agency (EMA) has already published the final guidance, and the FDA has now followed with its own announcement of availability.

The release of the guideline reflects the growing emphasis regulators are placing on the role of RWD in safety evaluation. In a recent interview with The Evidence Base, Robert Reynolds, Vice President of Epidemiology at GSK, noted that “the ICH M14 guideline has the potential for true global impact because it brings together many guidance documents describing scientific and regulatory best practice into one document.” By consolidating these principles, he explained, the guideline can serve as an entry point for developers, offering clarity on regulatory expectations across the lifecycle of non-interventional studies using RWD for safety assessment, from initial study concept through to the dissemination of results.

The M14 guideline focuses on non-interventional studies using RWD to evaluate the safety of medicines, including drugs, vaccines, and other biological products. Such studies are an established component of post-marketing safety monitoring and typically analyze healthcare data generated during routine clinical care, including electronic health records (EHRs), administrative claims data, and patient registries, to identify and evaluate potential safety signals.

While primarily intended for safety assessment, the principles may also apply to effectiveness studies and align with broader regulatory efforts to support the generation of real-world evidence (RWE). As Reynolds notes in his interview “many of the elements [of M14] – the conceptual framework, the feasibility assessment process and most of the sections on study design, analysis, reporting and dissemination – are applicable to studies of effectiveness outcomes.”

The document emphasizes that the credibility of evidence generated from RWD depends on both the relevance and reliability of the underlying data and the methodological rigor of the study design. Ensuring that data are fit-for-use for the research question is therefore critical for supporting regulatory decision-making.

The guidance outlines a structured framework for designing and evaluating non-interventional studies, beginning with clearly defining the safety concern and research question and identifying appropriate study populations, exposures, comparators, outcomes, and covariates. Importantly, it emphasizes that methodological decisions should be driven by the research objective rather than the availability of a particular dataset.

Key considerations for determining whether a study can generate adequate evidence for regulatory decision-making include the relevance and reliability of the data, the appropriateness of the study design and analytical methods, and a systematic assessment of study limitations and potential sources of bias. The guidance also recommends feasibility assessments and encourages early engagement with regulators when planning studies.

The final M14 guidance replaces a draft published in July 2024 and incorporates revisions following public consultation, including expanded discussion of feasibility assessments, protocol development, and quantitative bias analysis. The FDA has also announced the withdrawal of its 2013 guidance on pharmacoepidemiologic safety studies using EHR datasets.