Verbatims From the EDC
Free-text answers in (S)AE reports, medical histories and any scheduled event form become verbatims in the coding overview, with subject, form and field one click away.
The moment a site submits a free-text field, the verbatim is queued for coding. AI proposes the MedDRA term with a confidence score, your medical coder approves it, and every step is recorded in the audit trail. No export, no spreadsheet, no second system.
Free-text answers in (S)AE reports, medical histories and any scheduled event form become verbatims in the coding overview, with subject, form and field one click away.
AI proposes the MedDRA term for each verbatim and rates its confidence as high, medium or low, so your coders know which proposals need a closer look.
Compound verbatims such as “chest discomfort and anxiety” are split into single terms. Each term is coded and reviewed on its own and stays linked to its siblings.
From the free-text field to the approved MedDRA code, in the same system as your study data.
Choose the dictionary version and pick the free-text fields to be coded, form by form. From then on every submitted answer is queued for coding.
AI proposes the lowest level term and its system organ class for each verbatim. The ring around the robot shows the confidence: green for high, orange for medium, red for low.
Open a verbatim to see subject, (S)AE ID, site, form and field next to the full MedDRA hierarchy from LLT to SOC, and jump straight to the source form.
Verbatims that describe more than one event are flagged for a split. Each term is coded on its own, and the siblings stay linked to the original verbatim.
Filter by status, confidence, form or site and set the status for many verbatims at once. Only users with the approve right can approve a code.
Coders review proposals instead of searching the dictionary for every verbatim.
The same verbatim gets the same term, across sites and over the whole study.
Subject, form and field sit next to the code, no export or lookup needed.
Statuses, rights and every change are recorded in the audit trail.
Medical coding in Climedo is AI-assisted, not AI-alone. Every code stays a proposal until a coder with the approve right accepts it, and removing a code sets the verbatim back to uncoded. Rights are set per role: read, update and code, approve, export.
In clinical research, data privacy is non-negotiable. Climedo’s AI runs inside our own EU cloud perimeter, so your study data stays where your trial does, never in a public AI service.
Documents, voice notes and verbatims are processed inside our secure environment and deleted from our servers within seconds. Your data is never used to train public AI models.
Advanced language models run inside our private, EU-hosted architecture in Frankfurt, with no processing outside the EU: the power of modern AI without the risk of external exposure.
At Climedo, data security comes first. Our hosting partner, from whom we lease servers, is ISO 27001 certified (Standards for Information Security Assurance). In addition, we rely on strong data separation and state-of-the-art encryption methods. At the federal level, our system has also been tested by the BSI and the BfDI.

Prof. Dr. Christoph Schmitz
Chief Medical Officer, Auto Tissue Berlin
