Variability in the assessment of adverse events in a multicenter clinical trial

Background: Consistent documentation, characterization, and evaluation of adverse events (AEs) are needed during multicenter clinical trials to ensure accuracy of data reported to the US Food and Drug Administration and in the medical literature. Objective: The purpose of this study was to identify and characterize variations in the assessment of AEs by clinical trial personnel. Methods: During the annual meeting of personnel from a multicenter, controlled clinical trial of an investigational new drug treatment for opioid dependence, an oral presentation of procedures for AE data collection was given to 25 principal investigators and ancillary study personnel who assessed AEs for the study. A post-test using 3 hypothetical AE cases in which AEs were categorized by type of reaction, relatedness to study drug, severity, action taken, and outcome was completed by study participants. Cases and expected responses were reviewed for content and validity by clinical research pharmacists who were not involved with the study. The level of agreement with expected responses was assessed using McNemar symmetry chi-square tests. Results: Assessments of type of AE, relatedness to study drug, and severity were less frequently aligned with expected responses than were action taken and outcome (P < 0.013). Less consistency with expected responses was found in 1 case than in the other 2, suggesting that certain types of AEs may be more difficult to assess. Conclusions: There was considerable variability in categorization of AEs in an exercise following training for AE data collection. Type of report, relatedness, and severity were found to have more variability in reporting than did action taken or outcome. The results suggest that unless data are gathered to verify reliability of reporting, subcategorization of AE data should be undertaken cautiously. Further research is needed regarding methods for improving consistency in reporting of AEs.