Table 4 Discrepancies in harms data between CSRs, trial registries, and publications for variables that were reported in two sources
CSR and trial registries | CSR and publications | Publications and trial registries | |
|---|---|---|---|
Discrepancies | |||
Number of patients with at least one SAE | 15/32 trials (47%) | 5/13 trials (38%) | 8/14 trials (57%) |
Total number of SAEs | 5/5 trials (100%) | 0/1 trial (0%) | No trials with data from both sources |
Number of patients with at least one AE | No trials with data from both sources | 2/11 trials (18%) | No trials with data from both sources |
Total number of AEs | 5/5 trials (100%) | No trials with data from both sources | No trials with data from both sources |
Number of patients with at least one Grade 3–5 AE | No trials with data from both sources | 7/12 trials (58%) | No trials with data from both sources |
Total number of Grade 3–5 AEs | No trials with data from both sources | No trials with data from both sources | No trials with data from both sources |
Number of deaths due to AEs | 12/13 trials (92%) | 4/10 trials (40%) | 4/4 trials (100%) |
Number of patients who discontinued trial due to AEs | 23/26 trials (88%) | 18/20 trials (90%) | 11/18 trials (61%) |