The Clinical Trial Risk Tool has been featured in a guest column by Thomas Wood, director of Fast Data Science, in Clinical Leader, titled A Tool To Tackle The Risk Of Uninformative Trials, in cooperation with Abby Proch, Executive Editor at Clinical Leader.
In this article, Thomas Wood discusses the issue of uninformative clinical trials in pharmaceutical research. Uninformative trials fail to provide meaningful results, either because they don’t address key questions or are poorly designed. These trials waste resources, expose participants to unnecessary risks, and hinder progress in medical knowledge.
Wood uses the definition of “uninformative trials” from Zarin et al (2019)[1] as trials that do not deliver informative results, even though the investigated treatment may be effective or ineffective. An informative trial, according to experts, must meet five conditions: addressing an important question, providing meaningful evidence, being feasible, conducted scientifically, and reporting methods and results promptly.
The Clinical Trial Risk Tool was created to prevent such trials by providing features like a clinical trial budget estimator based on real-world cost data. The development of the first version of the tool was published in Gates Open Research[2]. In future, the tool could retrieve past trials with similar endpoints or inclusion/exclusion criteria to improve protocol design. It could also generate personalised feedback for different stakeholders, such as medical professionals, financial planners, and patient advocates.
Fast Data Science’s Clinical Trial Risk Tool offers a solution by allowing users to upload trial protocols, where AI analyzes the document for risk and cost factors. The tool estimates trial costs and identifies risks, helping to avoid uninformative outcomes early in the design process.
Ready to take the next step in your NLP journey? Connect with top employers seeking talent in natural language processing. Discover your dream job!
Find Your Dream Job
A person has recently returned from a camping trip and has a fever. Should a doctor diagnose flu or Lyme disease? Would this be any different if they had not mentioned their camping trip? Here’s how LLMs differ from human experts.
How can you predict customer churn using machine learning and AI? In an earlier blog post, I introduced the concept of customer churn. Here, I’d like to dive into customer churn prediction in more detail and show how we can easily and simply use AI to predict customer churn.
A generative AI-based legal advice tool for small businesses in distress We are pleased to announce the publication of our paper A generative AI-based legal advice tool for small businesses in distress, in collaboration with an interdisciplinary team based in the UK and Hungary.
What we can do for you
