Data management is crucial for the successful execution of clinical trials. Involving the collection, organisation, and analysis of data generated during the trial process, it ensures accuracy, integrity, and reliability, enabling informed decisions and valid conclusions from trial results. It also helps identify trends, patterns, and potential risks associated with the trial.
Accurate and reliable data is also essential for regulatory submissions and safety monitoring and ensures that the trial is conducted in compliance with regulatory requirements and Good Clinical Practice (GCP) guidelines.
Post-trial, data management continues to play a crucial role. Collection of real-world data from routine medical care can provide insights into the long-term risks and benefits of treatments in real-world settings.
However, collecting, collating and analysing such vast amounts of data is costly and time consuming. This is where artificial intelligence (AI) comes in.
How can AI help with clinical trials?
AI has permeated practically every industry since its introduction a few short years ago, and the clinical trials sector is no exception.
AI tools can potentially improve the success and efficiency of clinical research by analysing vast amounts of patient data and medical records, quickly and accurately. In particular, AI can help with clinical trials in several ways:
Patient recruitment
According to research collated by analytics firm GlobalData, around 80% of global clinical trials fail to recruit and retain enough patients to enrol on time. AI can efficiently identify clinical trial candidates by analysing patient data and medical records, narrowing the search for optimal cohorts and accelerating trial recruitment. It can also simplify entry criteria by analysing patient genetic makeup, physiological data, and lifestyle factors.
Patient adherence and retention
Traditional clinical trials have an average 30% dropout rate due to inconvenience, complex protocols, and lack of support. AI can improve the patient experience, and thereby retention, by enabling remote health assessments and real-time medication tracking, reducing non-adherence. AI-assisted apps can also provide reminders and allow patients to track progress, ensuring proper engagement with the study.
Data analysis
Clinical trials generate vast amounts of data that researchers manually review to uncover meaningful insights. AI can efficiently analyse this data, finding patterns that may be missed by human analysis. AI-based models can predict drug toxicity, help select suitable compounds, and even find data for new trials.
Caution is needed, however. Just as in more traditional analysis, inherent bias in data needs to be addressed. AI-based systems also raise data security and privacy concerns, so maintaining confidentiality of medical records is crucial.
Using AI to manage your data
During clinical trials, data is often collected in the form of free text, submitted by patients or staff, but these texts can be unclear or contain errors. When this goes uncorrected, data quality and patient safety are impaired.
However, manual review of these texts from the electronic Case Report Form (eCRF) presents a challenge requiring experts to understand and correct documentation errors, leading to time-consuming but avoidable tasks.
Alcedis (a Huma company) has developed an AI solution to help. Meteor is an AI application that streamlines the free text review process and automatically checks all texts for adverse events, drugs, treatments, and proper documentation language using AI models.
This increases patient safety, while also improving the quality of the data collected within the trial. This solution is GAMP-5-validated and compliant with relevant EMA guidelines.
Dr Daniel-Timon Spanka, is product manager for data analytics at Alcedis, and says that patient safety is the top priority in clinical research.
“Meteor helps to identify potential problems, such as hidden adverse events, in a timely manner,” he says. “Problems that patients may develop during trials are recognised more reliably, significantly improving the safety and integrity of clinical trials.”
The improvement in the quality of data gathered is another significant benefit of Meteor, says Dr Spanka:
“With Meteor, we provide the user with an AI toolbox. We use it to support experts in reviewing texts and making the best possible decisions. This allows incorrect documentation to be found quickly and effectively.”
Further, Meteor displays the complete patient history at a glance, removing the need for time-consuming searches in the eCRF for specific patient data. Dr Spanka summarises the benefits of the Meteor system as follows:
- Increased efficiency: By optimising the data review process and providing fast, accurate insights, experts can focus on strategic decision-making and keep their minds free for the essential aspects of the process.
- Higher data quality: AI reduces error rates in documentation, increasing data quality and improving patient safety.
- Standardisation: An intuitive user interface enables standardisation of the review of clinical free texts – across departmental boundaries.
- Compliance: Each user action is tracked in an audit trail to comply with the latest guidelines for computerised systems and electronic data in clinical trials.
Additionally, Meteor is a browser-based software and thus accessible from anywhere.
“Our Meteor software shows that the synergy of man and machine is shaping the future of clinical research,” says Dr Spanka. “This AI supported process not only saves costs and time, but also significantly increases the quality of clinical trial data and contributes to patient safety.”
Digital technologies like wearables and patient apps are revolutionizing clinical trials, reshaping data collection and analysis processes. Download this free whitepaper for more information.
