Updates to the Clinical Trial Risk Tool

Updates to the Clinical Trial Risk Tool

We have improved the Clinical Trial Risk Tool in the last 6 months, making it more user friendly and taking on board the feedback that we’ve received. We’ve improved the accuracy of the machine learning components too.

The tool now outputs its key figures such as risk levels and estimated cost in easily readable cards, so you can see at a glance the key takeaways from your protocol:

easily readable cards

The risk factors are now organised into collapsible categories, so you can explore them easily without an information overload.

The risk factors are now organised into collapsible categories

The tool identifies endpoints and inclusion and exclusion criteria. In future, we are hoping to use this to retrieve trials with comparable endpoints or criteria from the registries such as ClinicalTrials.gov.

endpoints and inclusion and exclusion criteria

Check your trial design

Check your protocol

Upload your clinical trial protocol in PDF form to the Clinical Trial Risk Tool and check the design against our checklist. Soon we will also support CONSORT and SPIRIT criteria.

We have an easily digestible set of recommendations for the user to improve the trial, so you can see what are the high priority actions that you need to take with your protocol.

recommendations for the user to improve the trial

What’s coming up soon in the Clinical Trial Risk Tool

Currently we’re working on a feature to allow the user to generate an itemised budget for the trial in Excel from the schedule of events. We expect this to take a few months and to be finished in Q4 of 2025.

We are also working on making the tool output the probability of success for the trial. A number of groups have calculated aggregate statistics around trials proceeding from Phase 1->Phase 2, or Phase 3->approval, etc, e.g. MIT’s Project Alpha or Chufan Gao, Automatically Labeling Clinical Trial Outcomes, so these statistics or machine learning models could be integrated into the tool, and in addition to outputting the “risk score” it could also output the “probability of success” or similar, based on past trials.

We hope to also produce an accountable budget range (where should a reasonable bid fall), based on past trials.

References

  1. Gao, Chufan, et al. Automatically Labeling Clinical Trial Outcomes: A Large-Scale Benchmark for Drug Development. arXiv preprint arXiv:2406.10292 (2024).

  2. Shomesh Chaudhuri, Joonhyuk Cho, Andrew W. Lo, Manish Singh, and Chi Heem Wong, Debiasing Probability of Success Estimates for Clinical Trials (2022)

Clinical trial team structure and best practices

Clinical trial team structure and best practices

Guest post by Safeer Khan, Lecturer at Department of Pharmaceutical Sciences, Government College University, Lahore, Pakistan Introduction The success of a clinical trial is strongly dependent on the structure and coordination of the teams managing it. Given the high stakes and significant impact of every decision made during the trial, it is essential for each team member to collaborate efficiently in order to meet strict deadlines, comply with regulations, and ensure reliable results.

How to read and extract value from a clinical trial protocol

How to read and extract value from a clinical trial protocol

Guest post by Youssef Soliman, medical student at Assiut University and biostatistician Clinical trial protocols are detailed master-plans of a study – often 100–200 pages long – outlining objectives, design, procedures, eligibility and analysis. Reading them cover-to-cover can be daunting and time-consuming. Yet careful review is essential. Protocols are the “backbone” of good research, ensuring trials are safe for participants and scientifically valid [1]. Fortunately, there are systematic strategies to speed up review and keep it objective.

How accurate is the Clinical Trial Risk Tool?

How accurate is the Clinical Trial Risk Tool?

Introduction People have asked us often, how was the Clinical Trial Risk Tool trained? Does it just throw documents into ChatGPT? Or conversely, is it just an expert system, where we have painstakingly crafted keyword matching rules to look for important snippets of information in unstructured documents? Most of the tool is built using machine learning techniques. We either hand-annotated training data, or took training data from public sources. How We Trained the Models inside the Clinical Trial Risk Tool The different models inside the Clinical Trial Risk tool have been trained on real data, mostly taken from clinical trial repositories such as clinicaltrials.