'25 at 25': Improving clinical trial monitoring

18 Jul 2024

Clinical trial monitoring involves overseeing how a trial is run, to check that the study protocol and good clinical practice guidelines are being properly followed, and that the data collected are reliable. Monitoring is crucial to protect patient safety and ensure trial results are robust. It also requires a lot of time and resource, so it is important to know how best to do it. However, this area of trial conduct has historically been under-researched.

MRC Clinical Trials Unit at UCL methodology researchers work to improve trial monitoring, and develop resources that help other researchers take a more structured approach to monitoring.

We have run studies which helped build a solid evidence base for the best approaches to monitoring. The Unit’s TEMPER study was one of the largest ever pre-planned studies in clinical trial conduct, focusing on ‘triggered’ monitoring.

Monitoring of clinical trials began with examining trial procedures and double-checking that data on case report forms filled in by site staff matched the participant’s hospital notes at frequent on-site monitoring visits. But this is expensive and resource-intensive, requiring dedicated monitors to comb through hospital notes.

Triggered monitoring is alternative approach. Trial sites are prioritised for visits according to an overall risk score, based on factors or ‘metrics’ such as sending data on time, completeness of data, or correctly following trial procedures. Each metric has an associated trigger, and the more triggers that fire at any time, the higher the site’s risk score. This approach should be more efficient, as some visits are avoided, yet still effective, as high-risk sites are investigated. But until the TEMPER study, there was no empirical evidence to show how well these approaches worked.

TEMPER provided this evidence by assessing triggered monitoring in three large cancer trials run by the MRC CTU at UCL, across multiple UK sites between 2013 and 2016. Any high-risk site that was triggered for a monitoring visit was paired with a control site, similar in aspects such as number of trial participants and time since its first participant joined the trial, but with a low risk score.

The team visited both sites at a similar time and compared their findings. If triggered monitoring worked, then researchers would expect to find more serious issues at the high-risk sites. But in a 2018 paper published in Clinical Trials, the results revealed that there was no statistically significant difference between high- and low-risk sites.

While TEMPER found that triggered monitoring was not as effective as expected, the team also provided evidence for how to improve this approach by focusing on the most useful metrics used to assess risk. This included time taken for sites to resolve questions about data quality, known cases where sites had not followed trial procedures, as well as site staffing such as the presence of a research nurse with wide-ranging responsibilities.

TEMPER generated important evidence on the key trial process of monitoring using a ‘study within a trial’ approach. This kind of study is vital for helping clinical trials run as efficiently and effectively as possible and form a core part of the MRC CTU at UCL’s methodology research.

A 2019 study nested within the Unit’s START trial (which tested the best time to start treatment for HIV) randomised sites to on-site or no on-site monitoring, to assess its value. The team found that, given the robust central and local monitoring programmes used in START, the added value of on-site monitoring was minimal considering its high cost.

Our research also involves assessing how monitoring is carried out across other clinical trials units (CTUs). In a survey sent to all 50 CTUs in the UK Clinical Research Collaboration (UKCRC) Registered CTU Network, we found wide variation in monitoring practice within UK academic trials. The team identified an urgent need for evidence-based guidance on best practice for monitoring, as well as tools for trialists to use.

In March 2020, the COVID-19 pandemic hit and forced monitoring to change. Trial staff began working from home, clinical research staff were redeployed to the COVID-19 response, and on-site monitoring visits were restricted.

As members of the UKCRC Monitoring Task and Finish Group, our researchers documented how trial monitoring had adapted to the pandemic. In a paper published in Trials, they analysed how trial teams found workable solutions to the lack of on-site monitoring by using central only or a combination of central and remote monitoring. The group also issued guidance for the wider trials community on transitioning to remote monitoring while minimising the burden on site staff.

Furthermore, the group recommended that some adjustments could remain in place beyond the pandemic to improve efficiency, such as the use of online ‘eConsent’ forms and allowing site staff to book onto and attend virtual protocol training at a time convenient to them.

The Unit now shares a collection of freely available resources and tools in our online monitoring toolkit. We aim to make the toolkit as accessible as possible by defining key terminology at the start. We have also developed five key principles which define the purpose of monitoring in plain language, published in the journal Trials:

Keeping participants safe and respecting their rights
Having data we can trust
Making sure the trial is run as it was meant to be
Improving the way the trial is run
Preventing problems before they happen.

In addition, our researchers contributed to the UKCRC Monitoring Task and Finish Group’s monitoring handbook, which answered the call for evidence-based guidance by providing information on the theory of monitoring, tips on conduct and real-life examples. The Task and Finish Group has also prepared a series of four training webinars on key principles of monitoring. These tools help trial teams understand the importance of monitoring and how to carry out monitoring successfully.

The toolkit is always evolving as new papers are published and resources are developed. Upcoming additions include guidance on more environmentally friendly monitoring practices, examples of useful metrics and triggers for risk-based monitoring, and a new trial monitoring plan template.

Further information: