Safety Reporting in RBQM: Bridging Risk Assessment, Monitoring, and Action

Applied Clinical Trials Online

Safety monitoring generates substantial signal volume across risk-based quality management tools, but only 30% to 36% of signals correspond to confirmed issues, suggesting the need for better prioritization, signal consolidation, and alignment between detection and proportional action.

Safety reporting is a critical aspect of clinical trial quality, directly impacting patient safety and regulatory compliance. While risk-based quality management (RBQM) provides multiple tools to manage this risk, ensuring these tools operate in a coordinated and effective manner remains challenging. In line with International Council for Harmonisation (ICH) guideline E6(R3), which emphasizes proportionate and integrated risk management, it is essential to assess how safety reporting is identified, monitored, and controlled across RBQM components.

This article is based on aggregated CluePoints data from more than 2000 clinical studies, offering a cross-study view of safety reporting in practice.

Safety reporting is consistently identified as a key risk. On average, studies define approximately 5 safety-related risks, yet only 16% are classified as high risk, suggesting limited prioritization at the risk assessment (RA) stage.

These risks are operationalized through quality tolerance limits (QTLs), key risk indicators (KRIs), and statistical data monitoring (SDM)—a set of complementary tools forming a continuum from risk identification to detection and escalation. In practice, studies generate approximately 10 safety-related signals, leading to approximately 8 actions per study, primarily driven by KRIs (78%) and complemented by SDM (21%). These actions reflect concrete interventions such as site internal process review, protocol clarification, targeted data review, or site retraining.

Signals represent potential issues requiring investigation, while confirmed issues correspond to validated deviations impacting safety or data quality. However, high activity does not necessarily translate into efficiency: Only 30% to 36% of safety-related signals correspond to confirmed issues. While many signals trigger action, these actions are not always proportional to confirmed problems, reflecting a precautionary and reactive model.

This pattern also suggests overlap across monitoring approaches, where different tools may detect signals linked to the same underlying issue, contributing to signal volume without necessarily increasing insight.

At a higher level, QTLs confirm that safety deviations remain frequent, with approximately 30% exceeding tolerance limits and impacting ~31% of studies, typically once issues have already accumulated.

Taken together, these findings highlight a key dynamic: Safety reporting is actively monitored, but the transition from risk identification to efficient detection and timely escalation is not fully aligned. The system is highly sensitive, generating substantial activity, but not always optimized for prioritization and early intervention.

To strengthen safety monitoring, a more integrated approach is needed:

• Prioritize safety risks in RA to drive monitoring focus
• Continuously update RA based on recurring signals and confirmed issues
• Consolidate signals across KRI, SDM, and QTLs to address root causes
• Triage signals based on impact and recurrence, avoiding unnecessary actions
• Differentiate isolated vs systemic issues and prioritize broader risks
• Act on patterns and escalate consistently, including updates to study documentation

A more integrated use of RBQM tools—aligned with ICH E6(R3)—can help sponsors move from a reactive, signal-driven approach to a more focused and proactive model, ultimately strengthening patient safety and study quality.