Clinical study reports (CSRs) are documents that provide detailed information about the design, methodology, and outcomes of clinical studies. They are created from sources such as Tables, Listings, and Figures (TLFs), among others. These documents are submitted to regulatory authorities, including the Food and Drug Administration (FDA) in the United States and the European Medicines Agency (EMA), to assess the safety and efficacy of pharmaceutical and biotech products.

Patient safety narratives represent an essential component of the CSR, providing comprehensive safety information about adverse events experienced by subjects during a clinical study.

More about patient safety narratives

Patient safety narratives help regulatory authorities assess the safety of investigational products and contribute to evaluating the risk-benefit ratio. These narratives typically include the subject’s identifier, a description of the clinical course leading to the adverse event, timing, severity, duration, concomitant medications, outcomes, laboratory measurements, and the investigator’s assessment. All information is presented in detail while maintaining patient privacy.

To produce patient narratives, data must be extracted from source files such as the Study Data Tabulation Model (SDTM), Case Report Forms (CRFs), and Council for International Organizations of Medical Sciences (CIOMS) forms. Medical writers then use this data to draft the narratives.

Challenges associated with authoring patient safety narratives

Writing a safety narrative is not a simple task. Many phase II to phase IV trials involve a high volume of narratives. In oncology studies, due to therapeutic complexity, long study durations, and large participant populations, this challenge is even more pronounced.

Authoring narratives requires substantial time investment to combine and structure data logically, chronologically, and consistently. In addition, narratives must follow specific formats, outlines, and sentence structures, adding another layer of complexity.

Narrativa’s solution can produce narratives with a few clicks

For years, Narrativa has refined its agentic AI tools to meet the needs of pharmaceutical and biotech companies. One of our key solutions is patient safety narrative automation. Our patient narrative generation technology leverages natural language processing (NLP) and natural language generation (NLG), subsets of AI that have delivered measurable value to our clients.

For example, the Leukemia & Lymphoma Society (LLS) uses our automation solution to generate patient narratives for submission to the FDA.

Clients report that narrative automation significantly reduces authoring time, with first drafts generated in just minutes. They also highlight how easily drafts can be modified and updated with a few clicks.

Another differentiating feature is the solution’s ability to detect inconsistencies within source databases. It can flag discrepancies in adverse event dates, incomplete entries, missing data, and other issues. Medical writing teams emphasize the importance of this functionality in validation and quality control workflows.

The solution can also be configured to match specific formats, sentence structures, and templates. This setup is completed once and reused, resulting in long-term time savings.

Collectively, the value of our patient narrative automation solution spans multiple stages of the production process, from drafting and quality improvement to validation and quality control.

Narrativa also provides additional regulatory submission solutions, including an anonymization and redaction tool for clinical documents.