The clinical data manager is often described as the custodian of research data, but this phrase should be understood broadly. Custodianship is not passive storage. It includes stewardship, design, oversight, communication, governance, and technical execution. A clinical data manager helps ensure that the data collected by a study are fit for their intended scientific, ethical, regulatory, and operational purposes.
In a typical study team, the data manager works with principal investigators, study coordinators, clinicians, nurses, laboratory teams, field workers, statisticians, monitors, regulatory staff, software developers, and sometimes community engagement teams.

The data manager must understand enough of the science to interpret the protocol, enough of the operations to design workable systems, enough of statistics to prepare analysis-ready data, and enough of governance to protect participant confidentiality and institutional accountability. At the planning stage, the data manager may help develop the data management plan.

This document describes data sources, collection tools, database systems, user roles, validation rules, quality control procedures, query management workflows, backup arrangements, coding standards, export procedures, and archival plans. A data management plan is especially important in multisite studies because it provides a shared reference for teams working across different facilities or regions.

At the design stage, the data manager develops CRFs and electronic databases. This requires attention to how data will actually be collected. A form that looks logical to the central team may be difficult to complete during a busy clinic. A variable that seems useful may be impossible to measure reliably at all sites. A free-text field may appear flexible but create difficulty during analysis. Good data managers balance scientific requirements with user experience and data quality.

During implementation, the data manager monitors data flow. They may review reports showing missing forms, delayed entry, unresolved queries, unexpected values, or site-level differences. They may train users, clarify completion guidelines, investigate data inconsistencies, and coordinate with statisticians before interim analyses or final database lock. In regulated studies, they may also support monitoring visits and inspections by providing evidence of traceability and data integrity.
The role increasingly requires technical competence. Data managers may use REDCap for database development, R for cleaning and quality checks, R Markdown or Quarto for automated reporting, Shiny for monitoring dashboards, and version control systems for documenting scripts and changes. However, technical skill is most valuable when guided by good judgment. The data manager must know not only how to build a validation rule, but why the rule matters and what risk it controls

Ensuring compliance with regulations and standards. Modern data managers often collaborate with:

• Principal Investigators
• Study Coordinators
• Statisticians
• Laboratory Teams
• Monitors
• Software Developers
• Ethics Committees

Their role bridges scientific, operational, and technical domains.