An increase in clinical trial data is driving demand for enhanced clinical trial processes. Sponsors want to be able to review data which has been analysed quickly and efficiently so they can make informed decisions faster.
In this high-volume, high-demand landscape risk-based quality management (RBQM), supported by innovative new technologies, offers an opportunity to improve data quality and integrity, while still creating much-needed efficiencies.
Growth in RBQM adoption
Evolving industry guidance has combined with growing recognition of the benefits of risk-based approaches to drive an uptake in RBQM adoption.
ICH E6 (R3) tied together the concepts of quality-by-design (QBD) and RBQM, bringing together risk assessment, risk mitigation strategy planning and risk detection. In 2023, the FDA issued additional guidance to facilitate sponsors’ implementation of risk-based monitoring1.
There is growing evidence to show how RBQM can help improve data quality. Traditional manual processes are labour-intensive, prone to error and may not be effective for spotting critical issues. Source data verification (SDV) drives just 2.4% of queries in critical data suggesting it has a negligible effect on data quality2 despite being a drain on time and resources.
Analysis of key risk indicators (KRIs) from more than 200 studies conducted via RBQM software found quality improvement in 82.9% of KRIs as measured by statistical score and 81.1% for observed KRI value3.
A comprehensive assessment of RBQM adoption in clinical trials published by the Tufts Center for the Study of Drug Development found 78% of respondents trust RBQM will improve the overall quality of research and 63% trust it will enable efficiency and cost savings4.
Combining RBQM with technology
Integrating new technologies with RBQM increases its potential to improve data quality and integrity from start to submission. Artificial intelligence (AI) in particular offers opportunities to streamline data management.
It can enable faster data analysis and identification of anomalies and help sponsors and contract research organisations (CROs) make sure errors are understood and prevented or spotted earlier in the trial.
For example, a deep learning (DL) module can be used to detect duplicate patients who might have enrolled in a study more than once at different sites by using a statistical-based comparison to determine the relative likelihood that any given pair of patients might be the same person. Intelligent patient profiles can be used to score patients statistically and identify those with unusual data patterns compared to others in the study.