Attribution Is the Hardest Problem in Adverse Event Reporting

Introduction: Grading Is Only Half the Story

Assigning a CTCAE grade answers the question of severity. Attribution answers the question of cause. In modern oncology, attribution is often the harder problem.

Combination regimens, sequential therapies, supportive medications, and comorbid disease blur causal lines. Yet attribution decisions affect protocol compliance, dose modifications, and safety analyses.

Why Attribution Is Inherently Uncertain

Patients rarely experience toxicity in isolation. A lab abnormality may reflect disease, treatment, infection, or chronic baseline variation. Symptoms evolve over time, overlapping with prior effects.

CTCAE provides structure for grading, but attribution relies heavily on clinical judgment.

Operational Consequences of Attribution Decisions

Incorrect attribution can lead to unnecessary treatment interruptions or failure to recognize true drug-related toxicity. In trials, it also affects sponsor reporting and regulatory interpretation.

Despite its importance, attribution rationale is often poorly documented.

How Decision Support Can Help

Automation can surface timelines, overlapping therapies, and relevant context to support attribution decisions. It can remind reviewers of protocol attribution rules and present alternatives without deciding.

The goal is support, not certainty.

Making Attribution Defensible

Defensible attribution requires transparency. Documenting rationale, uncertainty, and contributing factors protects clinicians and improves data quality. Attribution will never be perfect, but it can be made clearer, more consistent, and more auditable.