Physician experience and understanding of clinical trial adverse event attribution.

Abstract

e14524 Background: The NCI and FDA mandate the collection of adverse events (AEs) in cancer clinical trials to ensure patient safety. Clinicians are also required to declare whether or not each AE is related to the cancer drug patients received (known as AE attribution). However, prior research from our group demonstrates that attribution is unreliable. An overhaul of the current system is thus needed to ensure new drugs are safe for patients and that the development of promising cancer drugs is not halted for unfounded safety concerns. Hence, we conducted a survey to understand the current AE attribution process and to help formulate recommendations for an improved system. Methods: An electronic survey was developed and conducted among 165 medical oncologists, hematologists, and cancer prevention specialists at the Mayo Clinic. It included 21 items that queried clinicians about their clinical practice and trial experience, their training in AE attribution assignment, and their recommendations for improving the current AE attribution system. Results: 61 (37%) clinicians responded to the survey. The median survey completion time was 3.7 minutes. 43% of responders were younger than 45 years of age and 28% were female. The median years of clinical practice experience was 15 (range: 1 – 64) and of trial experience was 12 (range: 2 – 45). 89% had served as a trial PI. 15% of responders received formal AE attribution training, 70% only received informal training, and 15% received none. Other AE attribution experience is summarized in Table 1. Conclusions: The majority (66%) of experienced clinicians/clinical trialists from a large academic institution suggest formal training as the most effective means to improve attribution accuracy. This highlights the need for formal attribution training considering investigators in multi-center trials are typically from smaller institutions with fewer colleagues to consult and who enroll only a few patients to each trial. Table 1. Consulted when assigning attribution colleagues 74% study PI 72% Information used baseline AEs 85% AEs from previous cycles 84% Confidence in attribution assessment self 80% literature 60% Suggested training formal 66% regulatory guidance 38% informal or mentoring 43%