This SOP sets how [SITE NAME] defines, creates, corrects and protects source records so that trial data can be traced back to the participant and the person who recorded them, and can be reconstructed by a monitor, auditor or inspector.
2. Scope
This SOP applies to all source records for every study at [SITE NAME], on paper or electronic, including the electronic health record (EHR), study-specific worksheets, participant diaries and ePRO, device and instrument outputs, and records received from outside providers. It applies to all staff who create, review or correct source.
Out of scope: entry of data into the sponsor's EDC and query handling (SOP-307); validation and access control of site-owned computerized systems (SOP-401); retention and archiving (SOP-402).
Where the protocol, sponsor or IRB sets a stricter requirement than this SOP, the stricter requirement applies.
3. Responsibilities
Role
Responsibilities
Principal Investigator (PI)
Responsible for the integrity of data under the site's control. Approves the Source Data Location List for each study before the first participant visit. Reviews source data and gives evidence of that review.
Sub-Investigator (Sub-I)
Records and signs clinical assessments, eligibility decisions and AE assessments in source.
Clinical Research Coordinator (CRC)
Records study data at the time of the activity, follows correction rules, files outside records, and makes certified copies when needed.
Lead CRC
Drafts the Source Data Location List, prepares study-specific source worksheets, and keeps both current through amendments.
Regulatory Coordinator
Files the approved Source Data Location List and source templates in the ISF.
Quality Lead
Checks source quality during internal QC (SOP-700), reviews Notes to File for misuse, and trends data integrity findings.
All Research Staff
Record only what they did or observed, under their own identity, and report any suspected falsification under SOP-704.
4. Definitions
Term
Definition
Source Records
Original documents or data (including relevant metadata) or certified copies of them, in any media, in which trial data are first recorded. Examples include medical records, worksheets, ePRO data, lab and pharmacy records, and data from instruments, wearables and sensors.
ALCOA+
Attributes of reliable data: Attributable, Legible, Contemporaneous, Original, Accurate, plus Complete, Consistent, Enduring and Available. ICH E6(R3) also names secure and reliable as data integrity criteria.
Metadata
Data that describe other data, such as who entered a value, when, on which device, and any changes made.
Audit Trail
A secure, computer-generated, time-stamped record that shows the initial entry and each change to data, by whom, when and, where applicable, why.
Certified Copy
A copy of the original record, in any media, verified by dated signature or generated through a validated process to have the same information as the original, including relevant metadata.
Source Data Location List
A study-specific document, approved by the PI before the first participant visit, stating where each data item is first recorded. Some sponsors call it a source document agreement.
Shadow Chart
A research file kept alongside the EHR that duplicates or supplements medical record information for a study.
Note to File (NTF)
A signed and dated memo that explains a recordkeeping issue, a process decision or an event that is not evident from other records.
Late Entry
An entry made after the time the source record would normally have been completed, labeled as late.
5. Procedure
5.1 ALCOA+ Standard for All Entries
5.1.1
All Research Staff record study data so that each entry is:
Attributable: shows who recorded it (signature, initials listed on F-302-01, or unique electronic login)
Legible: readable and permanent; paper entries in blue or black ink
Contemporaneous: recorded at the time of the activity, or as soon as possible afterward; an entry made more than [DEFAULT: 1 business day] after the event is a late entry (section 5.5.5)
Original: recorded first in the location defined on the Source Data Location List, not on scrap paper, sticky notes or personal devices
Accurate: reflects what was done or observed, with units, times and values as measured
Complete: all required fields completed; blanks are marked "not done," "not applicable" or "unknown," with initials and date
Consistent: the same data item agrees across records, or the difference is explained
Enduring: kept on durable media for the full retention period
Available: retrievable for monitors, auditors and inspectors when requested
5.1.2
Staff do not use pencil, correction fluid, correction tape, erasable pens or labels placed over entries.
5.1.3
Staff never backdate. An entry is dated with the date it is written, not the date of the event, unless it is clearly labeled as a late entry under section 5.5.
5.1.4
Staff do not sign or initial for another person and do not share electronic logins.
5.1.5
Staff never create, alter or remove data to make a record appear complete or compliant. Suspected falsification is reported under SOP-704.
5.2 Defining Source for Each Study
5.2.1
Before the first participant visit, the Lead CRC prepares a Source Data Location List that states, for each data category in the protocol, where the data are first recorded. Typical categories:
consent process (consent form and F-301-01)
medical history, concomitant medications, vital signs, physical exam
eligibility confirmation (F-303-02)
adverse events and investigator assessments
IP dispensing and accountability (F-304-01, IRT)
lab results (central or local lab reports)
data captured directly in EDC, ePRO, IRT or devices (where the system itself is the source)
5.2.2
The PI reviews and signs the list. The Regulatory Coordinator files it in the ISF and gives the sponsor a copy if requested.
5.2.3
Where the sponsor allows data to be entered directly into EDC or ePRO without a prior paper or EHR record, the list names the system as source for those data items. Staff do not then make a paper copy as a second source.
5.2.4
The Lead CRC updates the list when a protocol amendment, new system or process change alters where data are first recorded. The PI signs each version.
5.2.5
Study-specific source worksheets are version-controlled, show the protocol version they support, and are reviewed against each amendment. Sponsor-supplied worksheets are checked to confirm they do not prompt staff to record data the protocol does not require.
5.3 EHR as Source and Certified Copies
5.3.1
Where the EHR is source, staff record study data in the EHR at the time of the visit and identify the encounter as a research visit as the site's EHR set-up allows. Research-only data that cannot be held in the EHR are recorded on study worksheets listed on the Source Data Location List.
5.3.2
The Site Director confirms that the EHR keeps an audit trail and that monitors can be given direct access (on site or remote) limited to enrolled participants, or that certified copies will be provided instead.
5.3.3
Records from outside providers (hospital discharge summaries, specialist notes, outside labs) are original source when received. Staff file the original or a certified copy in the participant's research record.
5.3.4
To make a certified copy of a paper record, the staff member compares the copy with the original, confirms it is complete and legible, and signs and dates the copy (or attaches a certification statement) stating that it is a true and complete copy of the original. A certified copy of an electronic record must keep the relevant metadata, or be produced through a validated process.
5.3.5
Staff destroy an original only if it has been replaced by a certified copy and the sponsor and the site's records policy (SOP-402) allow it.
5.4 Shadow Charts and Research Files
5.4.1
The research record for each participant may hold worksheets, certified copies and correspondence. It must not become a second, conflicting medical record.
5.4.2
Clinical information that affects the participant's routine care (for example new diagnoses, abnormal results, medications started) is also recorded in the medical record so that treating clinicians can see it.
5.4.3
When the same data item appears in more than one record, the Source Data Location List says which is source. Staff resolve conflicting entries by correcting the error under section 5.5, not by choosing a value in EDC.
5.5 Corrections and Late Entries
5.5.1
To correct a paper entry, the staff member draws a single line through the error so the original remains readable, writes the correct value nearby, and initials and dates the change. The staff member adds a short reason where the reason is not obvious (for example "transcription error," "wrong participant").
5.5.2
Electronic corrections are made through the system's correction function so that the audit trail captures the original value, the new value, who made the change, when, and the reason where the system asks for one. Staff do not delete and re-create records to avoid the audit trail.
5.5.3
Corrections must be supported by source and made within [DEFAULT: 5 business days] of the error being found, or the sponsor's query time if shorter. A correction to a clinical assessment (for example AE causality or eligibility) is made only by the investigator who made the assessment or another delegated investigator.
5.5.4
Only the person who made the original entry, or a delegated staff member with knowledge of the correct information, corrects an entry. Monitors and sponsor staff do not write in site source records.
5.5.5
An entry made more than [DEFAULT: 1 business day] after the event is a late entry. It is labeled "Late entry," dated with the actual date it is written, refers to the date of the event, states the reason for the delay, and is signed. The staff member also states the information source (for example "from participant phone call on [DATE]").
5.5.6
Staff do not rewrite or re-copy a page to make it neater. If a record is damaged and must be recreated, the original is kept with the new record, and the reason is recorded.
5.6 Participant-Reported Data, ePRO and Devices
5.6.1
Paper diaries completed by participants are source. Staff review them with the participant at the visit, clarify unclear entries and record any clarification separately, with the date and their initials. Staff never alter a participant's entry.
5.6.2
ePRO data are source in the sponsor's system. Staff train participants on the device, record the training in source, and check compliance as the protocol requires. Staff do not enter data in ePRO on the participant's behalf unless the protocol allows it and the entry is attributed correctly.
5.6.3
Where the site issues devices or wearables to participants, staff record the device serial number against the participant, so the data can be traced to the device and the participant (ICH E6(R3) Annex 1, 2.12.10(d)).
5.6.4
Problems with ePRO, wearables or other sponsor systems that may affect data are reported to the sponsor's help desk and recorded. Incidents that may have a significant or persistent impact on the data or system security are also reported to the sponsor and, where applicable, the IRB (ICH E6(R3) Annex 1, 2.12.10(e); SOP-401).
5.7 Investigator Review
5.7.1
Investigators review study data within the result review times in SOP-101, section 5.4.4, including data from outside sources that can affect eligibility, treatment or safety (central lab results, central imaging reads, outside records, ePRO alerts where the investigator has access).
5.7.2
The reviewing investigator gives evidence of the review by signing and dating the lab report or record, signing a visit note, or using the system's electronic sign-off. Clinically significant findings are assessed and acted on, and the action is recorded.
5.7.3
Review timing follows the protocol. Where the protocol sets no time, the investigator follows the result review timelines in SOP-101 5.4.4.
5.8 Notes to File
5.8.1
Staff use F-306-01 to explain something a reviewer would not otherwise understand from the records, for example:
where a record is kept when it is not in the expected place
a system outage and how data were captured during it
a sponsor instruction received verbally that changes a process
the explanation of a recurring documentation practice
5.8.2
A Note to File does not replace missing data, does not substitute for a required record, and does not correct a deviation. A missing assessment remains missing and is handled as a deviation (SOP-308).
5.8.3
Each Note to File states the study, participant (if any), the issue, the cause where known, the corrective action, and the date range it covers. It is signed and dated by the author and, where it concerns investigator responsibilities, by the PI.
5.8.4
The Quality Lead reviews Notes to File during internal QC. Repeated Notes to File on the same issue indicate a process problem and are referred for CAPA (SOP-703).
5.9 Data Governance and Protection
5.9.1
Source records are stored in secure locations with access limited to authorized staff (locked cabinets, restricted folders, role-based system access). Paper source does not leave the site except under SOP-313 or SOP-402.
5.9.2
Data reported to the sponsor are identified by participant code only. The link between code and identity is kept at the site in the confidential participant identification code list (SOP-303) and is not sent to the sponsor (ICH E6(R3) Annex 1, 2.12.8).
5.9.3
Site-owned systems that hold source data must have an audit trail that cannot be switched off, unique user accounts, and backup (SOP-401). Audit trails are not altered.
5.9.4
Staff do not transfer source data by personal email, text message, unencrypted removable media or personal cloud storage.
5.9.5
When data move between systems (for example EHR extract to EDC, or migration to a new EHR), the move is documented and checked so that data and metadata are complete and unchanged (SOP-401).
5.9.6
Privacy incidents involving source records are handled under SOP-400.
F-905-01 Cross-Entity Data Flow Register (Model N only)
SOP-301 Informed Consent Process and Documentation
SOP-307 Case Report Forms, EDC Entry and Query Resolution
SOP-308 Protocol Deviations and Noncompliance
SOP-313 Decentralized Trial Elements (Remote Visits, Home Health, Direct-to-Participant IP)
SOP-400 Participant Privacy and HIPAA in Research
SOP-401 Computerized Systems and Electronic Records
SOP-402 Record Retention, Archiving and Destruction
SOP-700 Internal Quality Control and Quality Assurance Audits
SOP-703 Corrective and Preventive Action (CAPA)
SOP-704 Research Misconduct and Fraud Reporting
SOP-901 Member Site Onboarding, Integration and Exit (Model N only)
SOP-905 Network Data, Systems, Privacy and Records Across Sites (Model N only)
Appendix B Central and Site Responsibility Matrix (Model N only)
7. References
Reference
Title
21 CFR 11
Electronic Records; Electronic Signatures
21 CFR 312.62(b)
Investigator recordkeeping: case histories
21 CFR 812.140(a)(3)
Records: participant case records
ICH E6(R3) Annex 1, 2.12
Records
ICH E6(R3) Annex 1, 4.2
Data Life Cycle Elements
ICH E6(R3) Annex 1, 4.3
Computerised Systems
FDA Guidance (Sept 2013)
Electronic Source Data in Clinical Investigations
FDA Guidance (2018)
Use of Electronic Health Record Data in Clinical Investigations
FDA Guidance (Oct 2024)
Electronic Systems, Electronic Records, and Electronic Signatures in Clinical Investigations: Questions and Answers
Independent Site SOP Library. Licensed CC BY-NC 4.0 by the American Institute for Medical Research (AIMR), with added permission for any site's internal use (license terms). Not legal advice: each adopting site confirms state law, IRB and sponsor requirements with its own counsel and owns every step it approves.