High-quality clinical trial data is essential for a successful clinical trial. This data is the foundation for the analysis, submission, approval, labelling and marketing of a compound. A focus on data cleaning, an essential process in the collection and management of clinical data, ensures that the data collected is consistent and accurate. Jennifer Bradford and Sheelagh Aird at PHASTAR analyse the AI approach to manage clinical data quality.
Extract:
Querying the Queries – An AI Approach to Manage Clinical Data Quality
High-quality clinical trial data is essential for a successful clinical trial. This data is the foundation for the analysis, submission, approval, labelling and marketing of a compound. A focus on data cleaning, an essential process in the collection and management of clinical data, ensures that the data collected is consistent and accurate.
However, it is not unusual for data errors to occur during data entry. Some of the most common are spelling or transcription errors, range errors and text errors, which impact coding. While automated edit checks exist to prevent the entry of inaccurate information, they are not able to detect all potential data entry issues. As data quality is at the crux of clinical trial success, clinical data management teams also use a manual approach to data cleaning by raising queries to the clinical trial sites to resolve any potential safety issues or inconsistencies in the data collected.
Often, on some studies, there can be numerous manually generated queries, which are time consuming and costly. By applying AI (artificial intelligence) techniques to understand the context of these queries, it may help improve automated edit checks or even offer opportunities to put additional checks or processes in place to help identify issues earlier in the studies. By applying machine learning to historic manual queries across different studies to understand common issues across and within studies, it could enable a more targeted approach to process optimisation for clinical trial data cleaning.
Click the download button below to read the complete version of ‘Querying the Queries – An AI Approach to Manage Clinical Data Quality’ by Jennifer Bradford and Sheelagh Aird at PHASTAR