Source Data Verification (SDV) for CRAs

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-08-27

Quick Answer

• Source data verification (SDV) is the process of comparing clinical trial data recorded in case report forms against original source documents to confirm accuracy and completeness.
• Clinical research associates (CRAs) conduct SDV during site monitoring visits, documenting discrepancies and resolving them with site staff before database lock.
• SDV requires professional judgment about risk-based monitoring priorities, and it does not guarantee that every data point is error-free.

Understanding Source Data Verification

Source data verification is a core monitoring activity in clinical research. The U.S. Bureau of Labor Statistics describes clinical research associates as professionals who monitor clinical trials and ensure that they are conducted according to protocol and regulatory requirements. The Occupational Outlook Handbook notes that medical scientists, including those who design and conduct clinical trials, work in settings such as research and development laboratories, hospitals, and universities. CRAs operate at the intersection of these roles, serving as the primary liaison between sponsors and clinical trial sites.

The term source data refers to the original records where clinical information first appears. These records include hospital charts, laboratory reports, patient diaries, radiology images, and other documents that capture the raw clinical facts. The case report form (CRF) is the data collection instrument used by the site to record information for the sponsor. SDV compares these two sets of records to ensure that what appears in the CRF matches what exists in the source documentation.

The purpose of SDV extends beyond simple transcription checking. Verification protects patient safety by confirming that eligibility criteria were met before enrollment and that adverse events are accurately documented. It also protects data integrity, which is the foundation of any clinical trial result. If the data are not accurate, the conclusions drawn from the trial cannot be trusted.

The Role of the CRA in SDV

The clinical research associate, also called a clinical monitor, is the professional who performs SDV. The role requires a combination of clinical knowledge, attention to detail, and communication skills. According to the U.S. Department of Labor's O*NET OnLine, clinical research coordinators and related professionals perform tasks such as preparing and maintaining documents, coordinating with other departments, and communicating with individuals outside the organization. These task descriptions align with the daily work of a CRA conducting SDV.

CRAs typically work for sponsors, contract research organizations (CROs), or academic research institutions. Their work involves regular visits to clinical trial sites, where they review source documents, compare them to CRFs, and document their findings. The U.S. Bureau of Labor Statistics notes that medical scientists, a category that includes clinical researchers, often work in teams and collaborate with other professionals. This collaborative aspect is central to the CRA role, because SDV requires ongoing communication with site coordinators, investigators, and data management staff.

The CRA's work context includes significant time spent reviewing documents and entering data. O*NET OnLine describes work context factors such as the importance of being exact or accurate, the need to work with others, and the requirement to make decisions that affect other people. These factors describe the SDV environment well, because a single transcription error can have downstream consequences for the trial.

Core Principles of SDV

Accuracy and Completeness

The first principle of SDV is accuracy. Every data point in the CRF must match the source document. This includes demographic information, medical history, vital signs, laboratory values, and adverse events. The second principle is completeness. Every required data field must be filled in, and every expected source document must be present.

Attribution and Legibility

Source documents must be attributable to the patient and the date of the visit. Handwritten entries must be legible, and corrections must be made in a way that preserves the original entry. The CRA checks that the source documents are signed and dated by the appropriate clinical staff.

Timeliness

Data should be recorded in the source document at the time of the clinical encounter, not reconstructed later. The CRA verifies that the dates on source documents align with the dates of the clinical visits and that the data was entered into the CRF within the expected timeframe.

Consistency Across Documents

The CRA checks whether the source documents are internally consistent. For example, the medical history in the clinic note should match the medical history in the admission note. Inconsistent source documents are a finding that must be resolved before the data can be considered reliable.

The SDV Process Step by Step

Preparation Before the Site Visit

Effective SDV begins before the CRA arrives at the site. The CRA reviews the previous monitoring report, the protocol, and the data that has been entered since the last visit. The CRA also reviews the subject enrollment log and the screening log to identify which subjects need verification.

The CRA prepares a subject selection list based on the monitoring plan. The monitoring plan defines the percentage of data to be verified and the critical data points that require 100 percent verification. The CRA also prepares the necessary forms and templates for documenting the visit.

The Site Visit

During the site visit, the CRA meets with the site coordinator and the investigator to review the status of the trial. The CRA then conducts the SDV, comparing the CRF to the source documents. The CRA reviews the informed consent forms, the eligibility criteria, the protocol deviations, and the adverse event reports.

The CRA also reviews the regulatory documents, including the delegation of authority log, the training records, and the laboratory certifications. The CRA verifies that the site has the required approvals and that the staff is properly delegated to perform trial-related tasks.

Documentation of Findings

Every finding from the SDV is documented in the monitoring visit report. The report includes the date of the visit, the subjects reviewed, the discrepancies found, and the actions taken. The CRA also documents the follow-up items that the site must complete before the next visit.

The monitoring report is a formal document that becomes part of the trial master file. It must be accurate, complete, and signed by the CRA. The report is reviewed by the CRA's supervisor and may be reviewed by the sponsor's quality assurance group.

Follow-Up and Resolution

After the visit, the CRA follows up with the site to ensure that the discrepancies are resolved. The CRA tracks the open findings and verifies the corrections at the next visit. The CRA also communicates with the data management team to ensure that the corrections are entered into the database.

At a Glance

Common Discrepancies Found During SDV

Transcription Errors

Transcription errors are the most common type of discrepancy. A laboratory value of 14.2 is entered as 14.2 in the CRF, or a date of 2024-01-15 is entered as 2024-01-16. These errors are usually corrected by the site coordinator after the CRA identifies them.

Missing Data

Missing data occurs when a field in the CRF is left blank or when the source document is missing. The CRA must determine whether the data was never collected or whether it was collected but not recorded. The CRA also checks whether the missing data is a protocol deviation.

Protocol Deviations

A protocol deviation is any change to the protocol that is not approved by the IRB. The CRA reviews the protocol deviations and verifies that they are documented and reported. The CRA also checks whether the deviation affects the subject's safety or the integrity of the data.

Eligibility Violations

Eligibility violations occur when a subject is enrolled in the trial but does not meet the inclusion or exclusion criteria. The CRA verifies the eligibility criteria against the source documents, including the medical history and the screening laboratory values.

Informed Consent Issues

The CRA verifies that the informed consent form is signed and dated by the subject and the investigator. The CRA also checks that the consent form is the current version approved by the IRB and that the subject received a copy.

Documentation Best Practices

The Monitoring Log

The monitoring log is the primary record of the SDV visit. It includes the date of the visit, the subjects reviewed, the findings, and the actions taken. The log is signed by the CRA and the site coordinator.

The Monitoring Report

The monitoring report is a formal document that summarizes the visit. It includes the status of the trial, the findings, and the recommendations. The report is submitted to the sponsor and the CRA's supervisor.

The Site Visit Log

The site visit log is a record of all visits to the site. It is maintained by the site and signed by the CRA and the site coordinator. The log is reviewed by the sponsor and the regulatory authorities.

The Subject File Review

The subject file review is a detailed review of the source documents for a specific subject. The CRA reviews the medical history, the physical examination, the laboratory results, and the adverse events. The CRA documents the findings in the monitoring log.

Communicating with Site Staff

Effective communication is essential for successful SDV. The CRA must be able to explain the findings to the site staff and to work with them to resolve the discrepancies. The CRA must also be able to communicate the findings to the sponsor and the data management team.

The CRA should use a professional and respectful tone when communicating with the site staff. The CRA should explain the reason for the finding and the required correction. The CRA should also acknowledge the site's work and provide positive feedback when the data is accurate.

The CRA should also be prepared to escalate the findings when the site does not correct the discrepancies. The escalation process is defined by the sponsor and the CRA's organization. The CRA should document the escalation and the response.

Risk-Based SDV

Risk-based SDV is an approach that focuses the verification effort on the data points that are most likely to be incorrect or that have the greatest impact on the trial. The approach is based on the principle that not all data points require the same level of verification.

The CRA uses the risk assessment to determine the data points that require 100 percent verification and the data points that require a lower level of verification. The risk assessment considers the protocol, the disease, the subject population, and the data management plan.

The CRA also uses the risk assessment to determine the frequency of the site visits. High-risk sites are visited more frequently than low-risk sites. The CRA documents the risk assessment and the rationale for the monitoring plan.

Records and Measurements

The CRA maintains records of the SDV activities, including the monitoring logs, the monitoring reports, and the site visit logs. The CRA also maintains the records of the data corrections and the escalations.

The CRA measures the effectiveness of the SDV process by tracking the number of discrepancies found, the time to resolve the discrepancies, and the number of unresolved findings. The CRA uses these measurements to improve the SDV process and to communicate the status to the sponsor.

Common Failure Patterns in SDV

Incomplete Preparation

The most common failure in SDV is incomplete preparation. The CRA arrives at the site without a clear plan for the visit, without the subject list, or without the previous monitoring report. This failure leads to an inefficient visit and missed findings.

Inconsistent Documentation

The CRA fails to document the findings consistently. The monitoring log is incomplete, the monitoring report is not signed, or the findings are not recorded in the data management system. This failure leads to a loss of traceability and an inability to track the corrections.

Poor Communication

The CRA fails to communicate the findings to the site staff or the sponsor. The site does not understand the required correction, or the sponsor does not receive the monitoring report. This failure leads to unresolved findings and a delay in the trial.

Lack of Follow-Up

The CRA does not follow up on the findings after the visit. The site does not correct the discrepancies, and the CRA does not verify the corrections at the next visit. This failure leads to a persistent data quality problem.

Limitations of SDV

SDV is a critical monitoring function, but it has limitations. The CRA cannot verify every data point in the trial. The CRA uses a risk-based approach to focus the verification on the most important data points, but this approach does not guarantee that every error is found.

SDV also depends on the quality of the source documents. If the source documents are incomplete or inaccurate, the SDV cannot confirm the accuracy of the CRF. The CRA must document the limitations of the source documents and escalate the issues to the sponsor.

SDV is also a time-consuming process. The CRA must balance the need for thorough verification with the need to complete the visit within the allotted time. The CRA must prioritize the verification activities and document the rationale for the priorities.

Professional Escalation Criteria

The CRA should escalate the findings to the clinical team leader or the sponsor when the site does not correct the discrepancies, when the discrepancies are significant, or when the site is not responsive to the communication. The escalation should be documented and the sponsor should be informed of the status.

The CRA should also escalate the findings when the source documents are not available, when the source documents are not reliable, or when the site is not compliant with the protocol. The escalation should include the findings, the actions taken, and the recommendations.

Career Context for CRAs

The CRA role is a professional role in the clinical research industry. The U.S. Bureau of Labor Statistics describes the work of medical and clinical research professionals, including the education and training requirements. The role typically requires a bachelor's degree in a health-related field, such as nursing, biology, or chemistry.

The National Institutes of Health Office of Intramural Training and Education provides training and professional development for research careers, including clinical research. The NIH training programs are designed to prepare individuals for research careers in the biomedical and behavioral sciences.

The American Society for Microbiology provides career resources for professionals in the microbiology field, including clinical research. The ASM career resources include job listings, career development, and networking opportunities.

The Association of American Veterinary Medical Colleges provides career resources for veterinary professionals, including those who work in clinical research. The AAVMC career resources include the veterinary education and career-path context.

The National Science Foundation provides career resources for scientific and technical professionals, including the clinical research and the research administration. The NSF careers page describes the work of the scientific and technical staff in a federal research organization.

A Practical Decision Framework for SDV Discrepancy Triage

CRAs often face a common problem during site visits: they identify more discrepancies than they can resolve in the available time. Without a structured way to prioritize, the CRA may spend hours on minor transcription issues while a serious eligibility violation goes unaddressed. A practical decision framework helps the CRA sort findings by their impact on subject safety, data integrity, and regulatory compliance.

The Three-Tier Classification System

The framework uses three tiers to classify every discrepancy found during SDV. The CRA assigns a tier at the moment the discrepancy is identified, before deciding whether to stop the review, continue with the visit, or escalate the issue.

Tier 1: Critical Discrepancies. These are findings that affect subject safety, the subject's eligibility for the trial, or the primary endpoint data. Examples include an eligibility violation, an unreported serious adverse event, or a missing informed consent form. A Tier 1 finding requires immediate action. The CRA should stop the current review, document the finding, and discuss it with the site investigator before continuing the visit.

Tier 2: Major Discrepancies. These are findings that affect the accuracy of secondary endpoints or that indicate a pattern of poor documentation. Examples include repeated transcription errors in laboratory values, missing dates on source documents, or inconsistent medical history across documents. A Tier 2 finding requires correction before the database lock, but it does not require the CRA to stop the visit.

Tier 3: Minor Discrepancies. These are isolated transcription errors or formatting issues that do not affect the interpretation of the data. Examples include a single typographical error in a non-critical field or a missing initial on a correction. A Tier 3 finding is documented and corrected during the visit or at the next visit.

The Triage Decision Process

The CRA applies the framework in a consistent sequence. First, the CRA asks whether the discrepancy affects subject safety. If the answer is yes, the finding is Tier 1. Second, the CRA asks whether the discrepancy affects a primary or secondary endpoint. If the answer is yes, the finding is Tier 1 or Tier 2 depending on the severity. Third, the CRA asks whether the discrepancy is an isolated event or part of a pattern. A pattern of errors in the same field or the same subject raises the tier.

The CRA records the tier assignment in the monitoring log. The tier determines the follow-up action. Tier 1 findings are discussed with the investigator during the visit and are escalated to the sponsor immediately. Tier 2 findings are documented in the monitoring report and tracked for correction at the next visit. Tier 3 findings are listed in the monitoring report and corrected by the site coordinator.

Applying the Framework to Common Findings

The framework is applied to the common discrepancies described in the SDV process. A transcription error in a laboratory value is Tier 2 if the value is used for a safety assessment or a dosing decision. The same error is Tier 3 if the value is a non-critical exploratory measurement. The CRA uses the protocol to determine the criticality of each data point.

An eligibility violation is always Tier 1. The CRA verifies the inclusion and exclusion criteria against the source documents. If the subject does not meet the criteria, the CRA documents the finding, discusses the impact with the investigator, and escalates the issue to the sponsor. The sponsor determines whether the subject continues in the trial or is withdrawn.

A missing informed consent is Tier 1. The CRA verifies that the consent form is signed and dated by the subject and the investigator. If the consent is missing, the CRA must determine whether the subject was treated without consent. This finding is escalated immediately because it affects the subject's rights and the validity of the data.

The Escalation Path for Tier 1 Findings

The CRA uses the escalation criteria to determine the path for Tier 1 findings. The CRA documents the finding in the monitoring log and the monitoring report. The CRA discusses the finding with the site investigator and the site coordinator. The CRA then contacts the clinical team leader or the sponsor to report the finding.

The escalation is documented in the trial master file. The CRA records the date of the escalation, the person contacted, and the response. The CRA also tracks the resolution of the Tier 1 finding until the site corrects the issue or the sponsor makes a decision about the subject.

The Follow-Up Path for Tier 2 and Tier 3 Findings

Tier 2 findings are tracked in the monitoring log. The CRA lists the findings in the monitoring report and assigns a due date for the correction. The CRA verifies the corrections at the next site visit. If the site does not correct the Tier 2 findings, the CRA escalates the issue to the clinical team leader.

Tier 3 findings are tracked in the monitoring log but do not require a separate escalation. The CRA verifies the corrections at the next visit. If the Tier 3 findings are not corrected, the CRA reclassifies the finding as Tier 2 and follows the Tier 2 path.

The Impact of the Framework on the SDV Process

The framework improves the efficiency of the SDV process. The CRA does not spend the same amount of time on every discrepancy. The CRA focuses the immediate attention on the Tier 1 findings and the follow-up attention on the Tier 2 findings. The Tier 3 findings are documented and corrected without interrupting the flow of the visit.

The framework also improves the accuracy of the monitoring report. The report includes the tier for each finding, which helps the sponsor and the data management team understand the severity of the issues. The tier also helps the CRA track the resolution of the findings over time.

The framework is consistent with the risk-based monitoring approach. The CRA uses the risk assessment to determine the critical data points, and the CRA uses the tier framework to determine the critical findings. The two approaches work together to focus the monitoring effort on the data that matters most.

The CRA's Professional Judgment

The tier framework is a decision aid, not a replacement for the CRA's professional judgment. The CRA must consider the context of the finding, the subject's condition, and the protocol requirements. The CRA must also consider the site's history of compliance and the site's response to previous findings.

The CRA documents the rationale for the tier assignment in the monitoring log. The documentation includes the reason for the tier, the context of the finding, and the action taken. This documentation is important for the sponsor's review and for the CRA's own record of the visit.

The CRA also uses the tier framework to communicate with the site staff. The CRA explains the tier of the finding and the required action. The CRA uses the tier to set the expectation for the site's response. The CRA also uses the tier to acknowledge the site's work when the data is accurate and the findings are minor.

Frequently Asked Questions

What is the difference between source data and a case report form?

Source data is the original record of a clinical event, such as a hospital chart or a laboratory report. The case report form is the data capture used by the sponsor to record the data for the trial. SDV compares the two to confirm that the data matches.

How often does a CRA conduct SDV?

The frequency of SDV visits depends on the risk assessment and the trial plan. The CRA conducts the visits at the intervals defined in the monitoring plan. The risk-based approach may require more frequent visits for high-risk sites.

What is the most common discrepancy found during SDV?

Transcription errors are the most common discrepancy. A laboratory value or a date is recorded incorrectly in the case report form. The CRA identifies the error and the site corrects the data.

How does a CRA document a discrepancy?

The CRA documents the discrepancy in the monitoring log and the monitoring report. The CRA records the finding, the subject, and the required correction. The CRA also tracks the correction in the data management system.

What is the role of the investigator in SDV?

The investigator is responsible for the accuracy of the data at the site. The investigator signs the case report forms and the source documents. The investigator is also responsible for correcting the discrepancies identified by the CRA.

What is the difference between SDV and data management?

SDV is the process of comparing the source data to the case report form. Data management is the process of cleaning the data and preparing the database for analysis. The CRA conducts the SDV, and the data management team conducts the data cleaning.

What is the career path for a CRA?

The CRA role is a professional role in the clinical research industry. The career path includes the clinical research coordinator, the clinical research associate, the senior CRA, and the clinical operations manager. The U.S. Bureau of Labor Statistics provides the labor-market context for the clinical research roles.

What is the education requirement for a CRA?

The CRA role typically requires a bachelor's degree in a related field, such as nursing, biology, or chemistry. The U.S. Bureau of Labor Statistics describes the education and training for the medical and clinical research professionals. The National Institutes of Health provides the training and professional development opportunities for the research careers.

Using the Evidence

Related Articles

• How to Read a Scientific Paper Efficiently
• Bioinformatics Courses and Certifications: A Practical Guide
• How to Get a Job in the Pharmaceutical Industry: A Practical Guide
• Genomic Data Internship: How to Land and Succeed in a Bioinformatics Internship
• interpreting data inheritance patterns mastering biology

References and Further Reading

• Occupational Outlook Handbook. U.S. Bureau of Labor Statistics.
• O*NET OnLine. U.S. Department of Labor.
• Office of Intramural Training and Education. National Institutes of Health.
• Careers at NSF. U.S. National Science Foundation.
• Career Services. American Society for Microbiology.
• AAVMC Career Resources. Association of American Veterinary Medical Colleges.
• NCBI Literature Resources. National Center for Biotechnology Information.
• PubMed. National Library of Medicine.

This article is educational and does not replace institutional policy, professional advice, or applicable safety and regulatory requirements.