Monitoring Visits

Every problem in your trial has three possible finders: your own site, the monitor, or the inspector. Found by you, it is a fix. Found by the monitor, it is a finding — one you can still close. Found by the inspector, it is part of the record before you say a word.

That ladder is worth translating. A monitoring visit is the sponsor checking its trial while the trial can still be corrected. Someone from outside your team, at intervals set by risk, reads your files and your data against what the protocol promised. Being found by the inspector is not the end of anything; corrective plans exist, and Module 11 walked you through them. But by then the discovery is an official finding, logged in an inspection report you do not control. Monitoring exists to keep discoveries on the ladder's first two rungs.

This module follows that work in six steps: what monitoring is for; how much of it a trial gets, and who decides; the three ways it reaches you; the four visits it arrives as; the verification at its centre; and what your site does before a visit and after its findings arrive.

Learning Objectives

After this module, you can:

• Explain what monitoring exists to protect — participants and the reliability of results — and place the monitor as quality control's field arm
• Judge how much monitoring a trial needs, and anchor the answer in the risk-based monitoring plan
• Distinguish on-site, remote and centralised monitoring — and spot what only centralised monitoring can catch
• Identify the four visit types — pre-study, initiation, routine, close-out — and what the monitor verifies at each, source data verification included
• Prepare your site for a monitoring visit, and act on what comes back so findings close instead of compounding

The Monitor's Job Is to Catch It First

A consent form dated one day after the blood draw it covers does not announce itself. Neither does a fridge that spent a weekend two degrees warm, or an eligibility criterion the whole team has quietly been reading one way when it means another. Trials rarely fail loudly while they run — they fail quietly, into the file.

Someone has to go looking. That someone is the monitor — the sponsor-appointed role Module 3 introduced, in job titles usually a clinical research associate (CRA). The looking has a name too: monitoring, the act of overseeing a trial's progress and ensuring it is conducted, recorded and reported in accordance with the protocol, the sponsor's procedures, GCP and the law. ICH E6(R3)

Monitoring exists to protect two things at once: the people in the trial, and the truth of what the trial will claim. Both halves matter to someone. The participant whose consent was mis-dated is protected by the visit that catches it; so is every future patient whose treatment will ride on this trial's numbers being real.

The work is broader than the site visit that gives this module its name. The monitor keeps a line of communication open with your site; verifies the team's qualifications and the site's resources; reviews training and trial documents; reads source records and reported data against each other; and follows analytics across the accumulating dataset. The role is also built on distance: monitoring is performed by people not involved in the clinical conduct of the trial at the site being monitored. The guideline counts monitors among the trial's qualified specialist roles. Your coordinator can check your site's work; they cannot monitor it. ICH E6(R3)

Where does this sit in the quality system? Module 11 already drew the map: quality control checks the work; quality assurance checks whether the checking works. It also placed the monitor precisely — monitoring is quality control, the sponsor's field arm, distinct from the independent audit that stands above it.

Which leaves the ladder's first rung where it always was: with you. The monitor is the trial's second finder by design, not by guarantee — whatever the first rung misses is what the second exists to catch. Your own checks — consent files reviewed before filing, the accountability log reconciled on schedule, entries read back against source — run between every visit, on your own initiative. A site that waits for the monitor to find its errors has not saved any effort.

For the participant, finally, the monitor is a protection they will never meet: an outside reader confirming that their consent was real, their data honest, their doses accounted for. And confirming it while each of those things can still be put right.

How Much Monitoring, and Why: The Risk-Based Plan

Two trials, one sponsor. The first gives a novel compound to its first-ever human cohort; the second compares two long-marketed painkillers at standard doses. If both sites get the same monitor for the same day every month, something has gone wrong — not with the visits, but with the thinking behind them.

How much monitoring a trial gets is a risk decision, made by the sponsor, and E6(R3) says so directly: the extent and nature of monitoring are determined from the identified risks. Into that assessment go the trial's objective and design, its complexity and blinding, the number of participants, the product's known safety profile, and the endpoints everything depends on. Visit frequency follows the same logic and moves as knowledge accumulates: a site that keeps producing findings earns closer attention; a stable, clean trial can earn less. ICH E6(R3)

The decision reaches into the fine grain — including how much source data verification (SDV), the checking of reported data against the original records it came from, each kind of data will get. The verification itself is this module's later ground; here it is one dial among several that the risk assessment sets.

This is risk-based monitoring, and R3 treats it as the baseline, not the exception: not less oversight — oversight aimed. The proportionality is not new to you. It is the fit-for-purpose principle Module 8 taught, processes scaled to the risks to participants and to the importance of the data, arriving here with a concrete job. ICH E6(R3)

The decision lives in a document: the monitoring plan. The sponsor writes it, tailored to the identified risks to participant safety, to data quality and to the reliability of the results. Particular attention goes to the procedures that protect participants and to the trial's endpoints. The plan describes the monitoring strategy, the activities of all the parties involved, the methods and tools to be used — and the rationale for using them. It is expected to consider what a site can carry, to focus on the aspects critical to quality, and to cover important work done away from your site: the central laboratory, the central image-reading facility. ICH E6(R3)

The plan is itself an essential record, one of the trial-specific plans expected to be in place before the trial starts. Whoever monitors follows it — the plan and the sponsor's procedures, not personal style, set what a visit covers. ICH E6(R3)

Swiss law, meanwhile, never says the word "monitor". Search the HRA and the KlinV and no monitoring article exists; the duty binds through a door you have walked through before. KlinV Art. 5(1) requires clinical trials to follow the rules of Good Clinical Practice. The ordinance's Annex 1 no. 2 names the ICH-GCP E6(R3) guideline — the version of 6 January 2025 — as those rules. The same article scales the duty: Art. 5(3) requires GCP's measures and precautions to be adapted to the risks participants actually face. Proportionate monitoring is not a loophole in Swiss law — it is Swiss law. KlinV

Three Ways to Watch: On-Site, Remote, Centralised

The site with the cleanest file in the trial — every field complete, never a missing value, never a query — may be its most dangerous one. From inside that site, nothing looks wrong. From above all forty sites, it is the one that looks like nothing else in the trial.

Monitoring reaches a trial along three routes, and R3 treats them as components of one strategy rather than rivals: site monitoring on-site, site monitoring at a distance, and centralised monitoring across every site at once. One block each. ICH E6(R3)

On-site monitoring. The monitor comes to your site — including, where the trial touches them, your pharmacy and your local laboratory — and works the visit in person. Records against source, the product count, the storage log, the conversations that show how work actually flows. How often the monitor comes is a risk decision, revised as the trial teaches the sponsor about your site. On-site is the only route that can see the physical trial: the fridge, the storage room, the team in its own corridor. ICH E6(R3)

Remote site monitoring. The same site-focused work, performed at distance where the activity's nature allows. R3 says the operative part plainly: monitoring may include remote and secure, direct read-only access to source records, to data acquisition tools and to the systems where essential records are kept. Read-only is the load-bearing phrase — the monitor can examine, analyse and verify, and the access leaves the record untouchable. That permission to examine is what the glossary calls direct access, and it exists on-site and remotely alike. What remote monitoring cannot do is walk the corridor; it trades presence for frequency. ICH E6(R3)

Centralised monitoring. No visit at all. Centralised monitoring is an evaluation of the trial's accumulating data, performed in a timely manner by the sponsor's qualified people — a medical monitor, data scientists, biostatisticians — across all sites at once. It can complement site monitoring and reduce how much of it is needed, or carry the load alone. Its analytics surface what no single visit can: systemic issues, site-specific issues, protocol noncompliance, data that may not be reliable. And it aims the other two routes — centralised findings select which sites and which processes get targeted, for-cause attention. ICH E6(R3)

That is how the too-clean site gets caught. A site reporting complete data with no missing values, ever, is invisible to its own monitor, because each individual file is fine. Across the whole trial it is an outlier in completeness itself — and unexpected lack of variability sits on R3's own list of what the data checks hunt for. ICH E6(R3)

Centralised monitoring is also how a trial watches the data that never waits for a visit: readings arriving straight from wearables and sensors, which E6(R3) counts among source records and data acquisition tools. ICH E6(R3)

Which mix your trial runs — on-site, remote, centralised, in what proportions — is exactly what the monitoring plan from the previous section exists to state and to justify. The design of the trial decides; decentralised elements push toward remote access and analytics, a first-in-human cohort pulls the monitor physically close.

Four Visits Across the Trial's Life

Your relationship with the monitor starts before your site is chosen and ends after your last participant has gone home. In between it has a rhythm — four kinds of visit, each asking the site a different question.

E6(R3) folds them into one heading: site selection, initiation, management and close-out, the monitoring activities that run across the trial's life cycle. Your site meets them as four visits, in order. ICH E6(R3)

The pre-study visit (site selection). Before anyone signs anything, the sponsor confirms your site could actually run this trial. The investigator and team hold adequate qualifications; the resources and facilities — laboratories, equipment, staff — are enough to conduct it safely and properly. The question: could this site do it? ICH E6(R3)

The initiation visit. Selection said the site could; initiation confirms it is ready now, for this protocol: the team adequately informed, working from the current approved protocol, the current Investigator's Brochure and the product information that matters. The question: is everything in place to start? The reports of both visits sit on E6(R3)'s essential-records list with the mark reserved for records generally in place before the trial starts — the paper trail of readiness precedes the first participant. ICH E6(R3)

Routine visits. The long middle, at whatever frequency the plan's risk logic sets. Visit by visit, the monitor confirms the running trial against its promises:

• the essential records are being maintained — the file growing with the trial, not after it;
• consent was obtained before each participant's trial activities began;
• adverse events are reported within the deadlines the protocol, GCP and the law set — the duties themselves are Module 13's ground;
• the protocol's requirements for source records are honoured, and the site knows where its source data lives;
• the blind, where there is one, is intact;
• recruitment and retention are reviewed and reported;
• the reports and notifications the site owes are actually being provided;
• and the investigational product's whole path — storage conditions, sufficient in-date supply, dispensing only to eligible participants at protocol doses, instructions given, counts and records kept — is confirmed against the disciplines Module 10 taught. ICH E6(R3)

The close-out visit. When the last participant's last data point is in, the monitor closes the site properly. The arrangements for retaining the essential records are confirmed; Swiss law holds those records for at least twenty years after the trial's completion or premature termination KlinV — retention depth is Module 11's ground. And the investigational product's final accountability is settled: returned, destroyed or otherwise dispositioned, and documented. ICH E6(R3)

Source Data Verification: Against the Source, by Sample

Every number your trial reports was born somewhere — in a chart note, an instrument printout, a participant's diary. Source data verification is the monitor walking a reported number back to its birthplace and checking that the two match. Module 12 followed the data's whole life cycle and kept the query machinery; at its review station it said monitoring itself was this module's ground. This is that ground.

In R3's words, the monitor checks the accuracy, completeness and consistency of the reported trial data against the source records, and whether the reporting was timely. The direction matters. The source record is the original; the reported datum is its copy at one remove; verification asks whether the copy tells the source's truth — never the other way around. ICH E6(R3)

Three design choices keep the checking sane, and all three come from the risk logic:

• Sampled, not census. Verification may run on samples of the data, supported by analytics — and the sample's size and the kinds of records in it are adjusted when earlier monitoring results or other quality signals say the site needs a closer look.
• Criticality first. The data the monitoring plan flagged as critical — primary and key secondary endpoints, the processes protecting participants — is verified against source as the priority; an administrative date is not owed the effort a primary-endpoint value is owed.
• Eligibility, always. The monitor verifies that only eligible participants are being enrolled — the one check where source records, the protocol and participant protection all meet. ICH E6(R3)

Verification also reads for shapes, not just matches. R3's list of what the data checks hunt is a fraud examiner's checklist written politely: missing data, inconsistent data, outliers, unexpected lack of variability, deviations from the protocol. It compares ranges, consistency and variability within and across sites, and it looks for significant collection errors, potential data manipulation and data integrity problems. ICH E6(R3)

When verification finds a mismatch, the monitor's job is to say so. Entry errors and omissions are communicated to the site, and the correction — made, dated, explained where necessary, its approval documented — runs through the trial's query-and-correction machinery. The monitor flags; the record changes only through the audit trail. ICH E6(R3)

Preparing for the Visit — and Acting on What Comes Back

A monitoring visit is an open-book exam, and the book is your file. The visit is confirmed in advance and its scope is known; a site that is surprised by a finding has usually been surprised by its own filing first.

Preparation, then, is mostly the discipline you already run, brought to a point. The records the monitor will read are the ones GCP already requires you to keep current — the site file, the delegation log, the product accountability, the source records for the participants under review. What the visit adds is logistics: access, space and people.

Two duties carry the visit itself, and both are the investigator's by GCP. The investigator and institution permit monitoring — the same clause that admits the auditor and the inspector. And upon request, they make all requested trial-related records available for direct access: the permission to examine, analyse and verify that monitoring is built on. ICH E6(R3)

Access travels with obligations in the other direction. Monitors and the staff around them adhere to data-protection and confidentiality requirements — regulatory, institutional and the trial's own data-security standards. Anyone with direct access takes reasonable precautions to keep participants' identities, and the sponsor's proprietary information, confidential. ICH E6(R3) The legal framework behind that duty is Module 6's ground; at your site it means the monitor's access is arranged, bounded and documented — never informal.

Then the visit ends, and the part your site is judged on begins. The monitor writes a monitoring report after site and centralised monitoring activities alike. It carries a summary of what was reviewed, the significant findings and conclusions, the actions required to resolve them — and follow-up on their resolution, including actions still open from previous reports. That report is the sponsor's record, not yours. It goes to the appropriate sponsor staff for review and follow-up, and original records generally stay with the party that generated them. ICH E6(R3)

What arrives at your site is the follow-up: the monitor informing you of deviations found, of entry errors to correct, of the actions your site owes — proportionate to how much each finding matters. That correspondence, and the corrective work it triggers, are your records. E6(R3)'s essential-records list carries both: the documentation of relevant communications, and the records of noncompliance with their corrective and preventive actions. ICH E6(R3)

Now the trap this module exists to spring: the follow-up left unanswered. An action item does not expire. The next report follows up on unresolved actions from the last one; findings that need escalation are escalated to the sponsor, whose decisions and their resolution are recorded. Ignore the letter and the record builds itself: a documented pattern of a site told, twice, and a PI who did not act. That is no longer a filing problem; it is evidence about oversight — the responsibility that stays with the investigator no matter who performs the delegated work. ICH E6(R3)

Module Summary

Three finders, one ladder — and this module put the middle rung in focus. The monitor is quality control in the field, aimed by a written, risk-argued plan, arriving on-site, remotely, and through the cross-site analytics no single visit can match. Four visits frame a site's trial: could you, are you ready, are you keeping your promises, can you close. At the centre sits verification — reported numbers walked back to the records where they were born, on a sample sized by risk. And after every visit comes the follow-up your site answers: in writing, on time.

You can now:

• Explain what monitoring exists to protect — participants and the reliability of results — and place the monitor as quality control's field arm
• Judge how much monitoring a trial needs, and anchor the answer in the risk-based monitoring plan
• Distinguish on-site, remote and centralised monitoring — and spot what only centralised monitoring can catch
• Identify the four visit types — pre-study, initiation, routine, close-out — and what the monitor verifies at each, source data verification included
• Prepare your site for a monitoring visit, and act on what comes back so findings close instead of compounding

The monitor is the only person in the system whose entire job is to find your problems while they can still be fixed. Treat the visit as the trial offering you its second chance — because the third finder does not offer one.