Coding Pregnancy: How Classification May Shape Clinical Research and Practice

The underrepresentation of pregnant people in clinical research limits the availability of reliable evidence for clinical practice. While this gap is often attributed to ethical and regulatory constraints, less attention has been paid to the role of medical classification systems in shaping the utility of Real-World Data (RWD). This paper examines how the coding of pregnancy within the International Classification of Diseases (ICD) and the Medical Dictionary for Regulatory Activities (MedDRA) may influence evidence generation and use.

We conducted a conceptual analysis of ICD and MedDRA, focusing on how pregnancy and related events are encoded. Drawing on literature from medical classification, pharmacovigilance, and evidence-based practice, we assess how classificatory fragmentation may affect data identification, retrieval and interpretation. An example from multiple sclerosis care is used for contextual illustration.

Pregnancy-related information is fragmented across different coding elements, without explicitly capturing the crucial temporal and clinical context. This structural limitation may hinder the reconstruction of pregnancy episodes, affecting the capacity of observational research to compensate for interventional research gaps. To overcome this, we propose a three-step approach: (1) providing dedicated training on the correct use and interpretation of existing classification systems; (2) complementing these infrastructures with a standardised minimum dataset to systematically link pregnancy variables; and (3) establishing a feedback loop from secondary research to continuously improve clinical documentation. Ultimately, this strategy can support robust RWD generation and more inclusive, evidence-based care.