Monitoring Findings, Source Data, and the Escalation Ladder
A monitor conducting a routine visit at a site in a trial of an orthopedic implant finds: (1) source documents for 4 of 15 participants are missing the implant serial number, which the protocol requires to be recorded; (2) the site's delegation log does not list a study nurse who has been performing outcome assessments for three months; (3) the electronic case report form shows a pain score of 3 for one participant while the source shows 8, and the site cannot explain the discrepancy; (4) the site has not reported two adverse events to the sponsor within the protocol's 24-hour window. The site investigator is also the owner of the clinic and has a financial interest in the device's commercial success, which is disclosed in the trial's registry but not in the informed consent form. The monitor's visit report is due in 10 days.
1. Rank these findings by severity and explain the reasoning behind the ranking.
Read the worked answer
Severity should be ranked by potential to affect participant safety or data integrity, not by ease of correction.
Unreported adverse events (highest). Failure to report within 24 hours means the sponsor's safety surveillance is blind. If either event is serious or device-related, the delay may have delayed a safety signal affecting other participants. This is time-sensitive.
Undelegated nurse performing assessments. Assessments performed by staff not on the delegation log are, from a regulatory standpoint, not attributable to a qualified investigator. This affects the validity of every assessment that nurse performed over three months - potentially many participants - and raises a qualification/training question.
Source-to-CRF discrepancy (3 vs. 8). A single unexplained discrepancy is concerning but bounded; however, a 5-point difference on a pain scale is not a transcription slip and may indicate data fabrication or a systematic entry problem. It requires source verification of the other participants' entries.
Missing serial numbers. This compromises device traceability and the ability to link outcomes to specific units - important for a device trial, especially if a recall or failure analysis is later needed - but it does not by itself endanger participants.
Undisclosed financial interest in consent. This is a consent and conflict-of-interest issue. It may be a reportable deviation depending on what the IRB required, but it does not corrupt the clinical data in the way the others do. It is nonetheless serious for trust and for the validity of consent.
2. What should the monitor do before, during, and after the visit report, and where does this escalate?
Read the worked answer
Before the report is finalized:
- Confirm the two unreported AEs immediately - obtain the source, determine seriousness and device relationship, and get them into the sponsor's safety system the same day. Do not wait for the report.
- Obtain the nurse's training records and CV, and determine the full set of participants whose assessments the nurse performed.
- Pull source for the discrepant participant and for the other participants assessed by the nurse; expand source verification beyond the sampled 15 if the discrepancy pattern suggests a systemic issue.
In the report:
- Document each finding with the specific evidence, the participants affected, and the corrective action requested.
- Distinguish findings the site can correct (serial number capture, delegation log) from those requiring investigation (the discrepancy).
- State the escalation path explicitly.
Escalation:
- Sponsor medical monitor and quality: unreported AEs, delegation failure, discrepancy.
- IRB: the discrepancy may meet the definition of an unanticipated problem (possible fabrication affects the integrity of the data and the rights of participants); the undisclosed financial interest may be a consent deviation if the IRB required disclosure.
- If fabrication is suspected, follow the sponsor's and institution's research-integrity procedures; do not confront the site with an accusation in the report - state the facts and the unresolved question.
Follow-up: a targeted visit within a defined interval to verify correction, with the interval justified by the severity of the findings.