Comprehensive Assessment of Risk-based Quality Management Adoption in Clinical Trials
Abigail Dirks, Data Scientist, Tufts Center for the Study of Drug Development
The FDA and EMA have encouraged the use of risk-based monitoring (RBM) and risk-based quality management (RBQM) to mitigate risks related to essential safety and efficacy data in clinical trials. Insights from the Tufts Center for the Study of Drug Development’s comprehensive assessment of RBQM adoption inform adoption levels, barriers, and industry perceptions.
1. What is the current state of RBQM adoption in the industry?
Many companies have implemented some components of RBQM in some trials, but full implementation of RBQM across a portfolio is rare. In the average company, a little more than half of clinical trials are utilising RBQM. Certain components are widely adopted, including risk assessment and risk control planning at 79 per cent of trials, while others, like the solicitation of input from the patient community to support optimal trial design are only being used in 27 per cent of trials on average. Companies expect to increase usage of all components, besides those related to the documentation of SDV and SDR.
2. How has RBQM adoption been measured in the past, and how is it measured in this recent study?
Other assessments of RBQM are important foundational tools for understanding RBQM implementation. In one study, about 6000 clinical trials were assessed for the implementation of eight main components. RBQM adoption was quantified as the percent of those trials that had adopted at least one of the 8 components. CROs reported usage of each component.
In this recent assessment, it is estimated that 125 distinct companies reported usage for over 12,000 total clinical trials. Each respondent was asked to estimate the proportion of trials within their company that had been implemented across 32 discrete RBQM components, gaining insight into the extent to which adoption is occurring as well as more granularity into where there are gaps in adoption, based on which components are more widely adopted.
3. How does the current assessment and results align with regulatory guidance, mainly FDA’s ICH E6 (R3)?
ICH E6 (R3) emphasises the adoption of a risk-based approach to the identification, assessment, control, and review of risks throughout a clinical trial, aligning with components included in the three stages outlined in the assessment: Planning & Design, Execution, and Documentation. More specifically, the guideline mentions the use of centralised monitoring, remote monitoring, and advanced statistical methods. The study found that on average, a centralised monitoring plan was developed and the deviations or updates made to the plan were documented in about half of clinical trials. Remote monitoring was utilised in 60 per cent of trials on average and statistical data monitoring is reported to be utilised in 49 per cent of trials on average.
The use of source data verification (SDV) was also assessed and is encouraged in the guidance. Looking forward to 2027, companies reported that they expect to reduce SDV or target SDV in about 50 per cent more trials than currently. 56 per cent of trials currently reduce or target SDV during Execution. They also report that the documentation of SDV will decrease by 13 per cent less trials by 2027, presumably as more trials reduce SDV or conduct more targeted SDV. Currently, 76 per cent of trials are documenting SDV.
Although not explicitly mentioned, the most recent guidance outlines quality tolerance limits (QTLs) as acceptable ranges that are set to detect deviations in support of risk mitigation. While 48 per cent of trials per company identify QTLs, 46 per cent utilise QTLs during clinical trial execution, 46 per cent identify important QTL deviations, and 44 per cent evaluate, follow-up, and resolve each important QTL deviation.
Although the industry has implemented some components of RBQM, compliance with ICH E6 (R3) will still require further adoption within companies, and many expect this to happen, reporting increased adoption by 2027.
4. What challenges is the industry experiencing in adopting RBQM?
The top challenge across stages, within Planning & Design, Execution, and Documentation & Resolution, is the lack of knowledge and awareness of RBQM, although 76 per cent of respondents did feel their organisation understands the meaning of RBQM somewhat or very well. A key theme from a roundtable of industry experts was the importance of change management strategies to ensure cross-functional understanding of RBQM, especially the purpose of RBQM. A meaningful change management plan that includes a variety of functions is an important step to implementation.
Other challenges include lack of technology and skills. For instance, many companies report using Microsoft Excel for risk assessment. It appears that RBQM professionals are looking for more advanced technology to handle advanced statistical methods for risk assessments. Respondents also report challenges with skills required for RBQM. Industry experts emphasised the need for analytical and critical thinking skills, along with communication skills. These skills may need to be further developed, as they were not critical for those conducting traditional monitoring.
5. Do companies trust RBQM and how can trust be increased?
Almost 80 per cent of companies trust RBQM to improve the overall quality of clinical research, while 63 per cent believe it will increase efficiency and cost savings, and a little over half trust RBQM to reduce study timelines. Respondents felt that within their organisations, site management/ site monitoring, clinical development, and clinical operations/study management have the lowest trust in RBQM. The functional areas that were least mentioned for low trust in RBQM were data management/data sciences and biostatistics. When asked how trust can be increased, 75 per cent of respondents chose “More experience with successful implementations”.
As mentioned, a top challenge is lack of knowledge and awareness, emphasising the need for change management strategies. A key part of a change management strategy is cross-functional education. Case studies from other companies may be a great way to spread knowledge to all employees on how RBQM may work and in turn, increase trust across an organisation.
6. What is the current level of understanding of RBQM in the industry?
As noted above, a key challenge for companies is the lack of knowledge and awareness of RBQM in the industry. Definitions of RBQM varied, although these definitions were collected before the release of ICH E6 R3. Regulatory guidance may help companies further define RBQM. The study also asked each respondent about their organisation’s level of familiarity with RBQM, understanding of what RBQM means, and commitment to RBQM.
7. What are the expected outcomes from implementation of RBQM?
A vast majority of respondents (91 per cent) noted that RBQM would improve reliability of trial outcomes and improve patient safety. Regarding cost savings, 61 per cent expected lower monitoring costs, while only 28 per cent expected lower data management costs. In terms of speed, 60 per cent felt that RBQM would cause faster time to database lock, but only 18 per cent felt that RBQM would cause faster patient enrollment.
8. Does your study cover insights into how RBQM adoption typically unfolds?
A deep dive into the data collected revealed certain subgroups that represent different stages of adoption, based on the number of components that have been implemented and the average percent of trials across the company that they are implemented within. Those who are more advanced in adoption tend to be larger companies, conducting 100+ trials per year. The most advanced are conducting duplicate patients detection in RBQM execution. About 70 per cent of the lowest adoption grouping is those conducting less than 25 trials annually. Most of the least advanced groups are soliciting input from the investigative research community to support optimal trial design, conducting risk assessment and risk control planning, identifying risks detected, documenting data SDV’ed, and documenting updates to and/or deviations from the site monitoring plan. Throughout the adoption process, companies begin to incorporate components like critical-to-quality (CTQ) factors, key risk indicators (KRIs), quality tolerance limits (QTLs), remote site monitoring, evaluation, follow-up and resolution of each risk, and eventually implement all 32 components.
Author Bio
Abigail Dirks is a Data Scientist at the Tufts Center for the Study of Drug Development (Tufts CSDD) where she specialises in analysing large datasets pertaining to industry drug development performance, including risk-based quality management (RBQM) adoption, investigative site burden and experience, impact of decentralised clinical trials (DCT), and clinical trial participation. She recently received her MS in Health Informatics and Analytics from Tufts University School of Medicine, Boston, MA and has a BS in Data Science and Mathematics from St. Michael’s College, Colchester, VT.
