Source data in a clinical trial is all the information held in original records, and in certified copies of those records, that is needed to reconstruct and evaluate the trial. A blood pressure reading, a consent date, a dispensing entry and a laboratory result are each source data at the place the value was first written down.
This guide sets out the ICH definition, the difference between source data and a source document, the test that separates an original from a copy, and which record holds the source for each common data type at a UK research site. It sits under our complete guide to eSource in clinical trials, which covers how that first record is captured electronically.
A record is source because of when it was written, whatever it is written on.
What Is the ICH Definition of Source Data?
ICH E6(R2) defined source data as “all information in original records and certified copies of original records of clinical findings, observations, or other activities in a clinical trial necessary for the reconstruction and evaluation of the trial”. Three parts of that sentence do the work.
- Original records: the first place each finding or observation was recorded.
- Certified copies: copies verified, by a dated signature or a validated process, to hold the same information as the original, including relevant metadata.
- Necessary for reconstruction and evaluation: the test of scope, which limits source data to what the trial needs.
ICH E6(R3) took effect in the UK on 28 April 2026 and uses the term source records in place of source documents. Its glossary defines them as “original documents or data (which includes relevant metadata) or certified copies of the original documents or data, irrespective of the media used”. The wording recognises that a first capture is often a database entry. Our guide to what the new UK CTR requires of research software covers the wider set of duties that arrived on the same date, and the full text sits in the ICH E6(R3) guideline.
What Is the Difference Between Source Data and a Source Document?
Source data is the information. A source document is the record that carries it. ICH E6(R2) listed source documents as “original documents, data, and records (e.g., hospital records, clinical and office charts, laboratory notes, memoranda, subjects’ diaries or evaluation checklists, pharmacy dispensing records, recorded data from automated instruments” and continues through microfiche, x-rays, subject files and records kept at the pharmacy, the laboratories and medico-technical departments.
| Term | What it means | Example at a UK research site |
| Source data | The information itself | The value 142/88 mmHg observed at a study visit |
| Source document or source record | The record that holds the information | The clinic observation chart entry, or the electronic visit form |
| Metadata | Data that describe the attributes of other data | The user, date and time the system stamped on that entry |
| Certified copy | A verified copy of the original record | A scanned ECG trace verified against the paper trace |
| Case report form entry | The sponsor’s structured extract of the data | The blood pressure field in the sponsor’s EDC system |
The MHRA describes an original record as “the first or source capture of data or information e.g. original paper record of manual observation or electronic raw data file from a computerised system, and all subsequent data required to fully reconstruct the conduct of the GXP activity”. The definition covers the entry and the metadata around it, because a value with no author and no time cannot be reconstructed. Our explainer on what EDC is in clinical trials covers the sponsor side of that boundary.
When Is a Record the Original and When Is It a Copy?
The original is the first capture of the information. The MHRA states this directly: “Raw data is defined as the original record (data) which can be described as the first-capture of information, whether recorded on paper or electronically.” Every later version of that value is a copy, whatever its format.
Consider a hypothetical oncology visit at an NHS trust. A participant is weighed in clinic, and the weight is typed into the electronic patient record by the clinic nurse. The research nurse later writes the same weight on a study worksheet, and a coordinator types it into the sponsor’s EDC system three days after that.
- The electronic patient record holds the original, because the value was captured there first.
- The study worksheet holds a copy, and it carries no evidence that the value was read correctly.
- The EDC entry holds a second-generation copy, two steps away from the observation.
- A discrepancy between the three versions can only be resolved against the electronic patient record.
Three tests separate an original from a copy in practice.
- Timing: the original is created at or near the moment of the observation.
- Authorship: the original records who made the entry, without relying on a handwritten name.
- Completeness: the original keeps its metadata and its audit trail, so earlier values remain visible.
Certified copies are the single exception, and they qualify as source data only where a verification step has confirmed that the copy carries the same content and meaning as the original. The MHRA also expects a record captured in a dynamic electronic state to remain available in that state, so a printed copy of an electronic record does not displace the electronic original.
Also Read: CTMS vs eTMF vs EDC vs eISF: Which System Do You Actually Need?
Which Records Hold Source Data by Data Type?
Each data type has a usual original, and the site confirms it study by study. The table below sets out where the first capture normally sits at a UK research site, and the point that most often causes a query.
| Data type | Usual original record | Point to watch |
| Vital signs and physical examination | The clinic observation chart or the study visit form | Values written on a scrap sheet first make that sheet the source |
| Medical history and concomitant medications | The electronic patient record | Entries added by the care team sit outside the research team’s control |
| Eligibility assessment | The signed eligibility checklist or the investigator’s note | The assessment needs a date that proves it preceded enrolment |
| Informed consent | The signed and dated consent form | The version number on the form must match the approved version in force |
| Laboratory results | The report issued by the analysing laboratory | Transcribed values lose the reference ranges and the accreditation detail |
| ECG, imaging and instrument output | The trace, image or data file produced by the instrument | Instrument clocks drift, so timestamps need periodic checking |
| IMP dispensing and accountability | The pharmacy dispensing record | Pharmacy records sit in a separate system with separate access |
| Adverse events | The record made when the event was assessed | Onset dates reconstructed later carry the date of the reconstruction |
| Participant-reported outcomes | The participant’s device entry or completed paper diary | Device entries made outside the window are visible in the metadata |
| Randomisation | The confirmation issued by the randomisation system | The confirmation is the original, and any worksheet entry copies it |
| Protocol deviations | The record made when the deviation was identified | A deviation found during monitoring still dates from the day it occurred |
A value that exists in two systems needs a written decision about which one holds the source, taken before the data is collected and recorded in the source data location log that is filed in the investigator site file.
Why Does the Source Definition Decide What an Inspector Reads?
An inspector reads the record the site has declared as source, and compares it with what the sponsor reported. ICH E6(R3) puts the duty on the investigator to “define what is considered to be a source record(s), the methods of data capture and their location prior to starting the trial and should update this definition when needed”.
The same guideline sets the quality bar for those records: “Source records should be attributable, legible, contemporaneous, original, accurate and complete. Changes to source records should be traceable, should not obscure the original entry and should be explained if necessary (via an audit trail).” Those attributes are the ALCOA+ principles applied to the first record itself.
The declaration settles four operational questions before the first participant arrives.
- It tells the monitor which record to open for each field, which shortens source data verification.
- It tells the site which records must be retained, which sets the scope of archiving.
- It tells the data team which discrepancies are real, because only one record can be the reference.
- It tells the sponsor where the audit trail lives, which is the evidence an inspection asks for.
Control over these decisions belongs to the site’s quality management system, which holds the standard operating procedures for source declaration, correction and record retention. The MHRA sets out its wider expectations in its GXP data integrity guidance and definitions.
Also Read: The Investigator Site File Essential Documents Checklist
A source declaration is easier to hold when the first capture and its audit trail sit in one controlled place. AQ is launching eSource soon as part of the AQ platform. Book a live demo to see the AQ platform today.
