On-Site and Remote Monitoring

Part of Clinical Research Associate (CRA) Training: Monitoring Approaches and Reporting.

About 5 minutes. Free, in your language, no account needed.

Why this matters

Most trials now use both. Knowing what each approach can and cannot do stops you doing the wrong one: traveling to check something you could have seen from a desk, or trying to assess a refrigerator over a video call.

What the guideline says

ICH E6(R3) supports both. Section 3.11.4.1 describes site monitoring as monitoring of activities at the investigator site, including its pharmacy and local laboratories, and states that it may be performed on site and/or remotely, and may include remote, secure, read-only access to source records.

That last phrase matters: remote source data verification is expressly contemplated by the guideline, where the access is secure and read-only.

The sponsor must define the approach in the monitoring plan and justify it. Section 3.11.4.3 says the plan should focus on aspects critical to quality.

On-site monitoring

When it is necessary:

• Verification of physical source documents that cannot be accessed remotely.
• Assessment of investigational product storage conditions.
• Review of equipment and calibration.
• Direct observation of how processes actually work, including the consent workflow.
• High-risk protocols, such as those with complex procedures.
• For-cause visits.
• New or inexperienced sites.
• Where sponsor policy or regulation requires periodic in-person oversight.

What it does better: full access to physical records and facilities; direct observation, which is more reliable than description; relationship building; understanding the site's actual workflow and culture; and training a new coordinator effectively.

What it costs: travel time and money; scheduling difficulty; long gaps between visits if they are infrequent; and a substantial burden on the site, which must give a full day to hosting you.

Remote monitoring

When it works:

• The site has good electronic source or electronic regulatory infrastructure.
• The study is lower risk, or non-interventional.
• Travel is restricted.
• The monitoring strategy calls for more frequent, lighter-touch oversight.

What can be done remotely: review of data entered in the electronic system; query management and trend analysis; review of remotely accessible electronic source where permitted and where access is secure and read-only; review of regulatory documents through electronic site file systems; teleconferences with the investigator and coordinator; tracking of recruitment, retention and deviations; and investigational product accountability where digital logs are available.

What it does better: faster issue detection, because you can look weekly rather than quarterly; lower cost; less disruption to the site; more sites covered; and earlier detection of data trends.

What it cannot do: assess physical storage, inspect equipment, review paper source that is not digitized, or observe how a clinic actually runs. It is also harder to build a relationship, and investigator availability tends to be worse without a scheduled in-person visit.

What it requires: the site's technology to work; a lawful and secure access route; and attention to privacy, because remote access to medical records raises data protection questions that must be resolved before, not during, the visit.

Hybrid models

Most modern trials combine the two. A typical pattern:

1. Frequent remote review of data trends in the electronic system.
2. Periodic on-site visits to verify physical source, product and equipment.
3. Centralized monitoring flagging signals that direct targeted verification.
4. Regular teleconferences to review open issues.

The benefit is that each activity happens where it is cheapest and most effective, and that on-site time is spent on what genuinely requires presence.

The risk is that the split is never defined, and both the CRA and the site assume the other approach covers something. Write down what each approach covers.

Choosing the approach

The decision depends on protocol complexity, the site's experience and performance history, the study risk assessment, technology availability, data privacy law in that country, and sponsor requirements.

The question to ask is: which approach maximizes oversight while minimizing site burden? A quarterly full-day on-site visit and a fortnightly 30-minute remote review are not equivalent, and for many sites the second detects problems sooner.

Whatever the mix, document it. The monitoring plan defines the approach, and monitoring activities are documented regardless of where they were performed.

In practice

Leah monitors four VT-301 sites. Two have electronic health records with a secure read-only access route that the institutions have approved; two are largely paper.

For the electronic sites, she reviews data fortnightly from her desk and visits on site quarterly. The on-site visits concentrate on product storage, equipment, the site file and staff conversations, because everything else has already been checked.

For the paper sites, she visits every six weeks and does more verification in person, and holds a short call between visits to review recruitment and open queries.

At one electronic site, a fortnightly review shows two participants whose visit windows are about to expire. Leah calls the coordinator, who reschedules both within the window. On a quarterly-only schedule she would have found two deviations instead of preventing them.

At one paper site, a quarterly on-site visit finds a temperature excursion nobody had reported, because the log was on the fridge and nobody looked at it remotely — because nobody could.

Both approaches found what only they could find. That is why the plan uses both.

Key takeaways

• E6(R3) section 3.11.4.1 states that site monitoring may be performed on site and/or remotely, and may include remote, secure, read-only access to source records.
• On-site monitoring is necessary for physical source, product storage, equipment, direct observation of process, new sites and for-cause visits.
• Remote monitoring detects data and trend issues sooner and at lower cost, but cannot assess storage, equipment or undigitized paper source.
• Hybrid models must define explicitly what each approach covers, or both the CRA and the site will assume the other covers it.
• The decision question is which approach maximizes oversight while minimizing site burden, and the chosen approach must be defined in the monitoring plan and documented in practice.