Table 2.
AE information typically presented by sector and overall
| CTU/public (N=46) |
Industry (N=18) |
Overall (N=64) |
||||
| n/N | % | n/N | % | n/N | % | |
| Information presented | ||||||
| Number of participants with at least one event | 4/46 | 8.7 | 4/18 | 22.2 | 8/64 | 12.5 |
| Number of events | 1/46 | 2.1 | 0/18 | 0.0 | 1/64 | 1.6 |
| Both of the above | 36/46 | 78.3 | 12/18 | 66.7 | 48/64 | 75.0 |
| None of the above | 5/46 | 10.9 | 2/18 | 11.1 | 7/64 | 10.9 |
| Other* | 16/46 | 34.8 | 6/18 | 33.3 | 22/64 | 34.4 |
| Descriptive and summary statistics† | ||||||
| Frequencies | 42/46 | 91.3 | 16/18 | 88.9 | 58/64 | 90.6 |
| Percentages | 43/46 | 93.5 | 14/18 | 77.8 | 57/64 | 89.1 |
| Risk difference | 5/46 | 10.9 | 5/18 | 27.8 | 10/64 | 15.6 |
| Odds ratio | 7/46 | 15.2 | 3/18 | 16.7 | 10/64 | 15.6 |
| Risk ratio | 6/46 | 13.0 | 5/18 | 27.8 | 11/64 | 17.2 |
| Incidence rate ratio‡ | 8/46 | 17.4 | 7/18 | 38.9 | 15/64 | 23.4 |
| Other§ | 6/46 | 13.0 | 4/18 | 22.2 | 10/64 | 15.6 |
| AE comparison† | ||||||
| Subjective comparison | 36/46 | 78.3 | 15/18 | 83.3 | 51/64 | 79.7 |
| Exclusion of null through 95% CI | 12/46 | 26.1 | 2/18 | 11.1 | 14/64 | 21.9 |
| Hypothesis test/p value | 18/46 | 39.1 | 3/18 | 16.7 | 21/64 | 32.8 |
| Other¶ | 4/46 | 8.7 | 5/18 | 27.8 | 9/64 | 14.1 |
| Awareness of any published methods specifically to analyse AEs | ||||||
| No | 25/44 | 56.8 | 4/17 | 23.5 | 29/61 | 47.5 |
| Yes | 11/44 | 25.0 | 12/17 | 70.6 | 23/61 | 37.7 |
| Don’t know | 8/44 | 18.2 | 1/17 | 5.9 | 9/61 | 14.8 |
| Undertaken any specialist AE analysis not mentioned in your previous response | ||||||
| No | 38/43 | 88.4 | 14/17 | 82.4 | 52/60 | 86.7 |
| Yes | 5/43 | 11.6 | 3/17 | 17.6 | 8/60 | 13.3 |
*Other ways of presenting AE information included presenting information on: overall number of events (n=2); number of patients experiencing 0, 1, 2, etc, events and number of AEs per patient (n=2); duration (n=1); relatedness (n=1) and severity (n=7) (full free text comments in online supplementary appendix table A1).
†Participants were able to provide multiple responses to this question.
‡Incorporates free text comments that described summaries synonymous with incidence rate ratios.
§Included a comment that a participant presents the ‘median number (IQR)’ of events.
¶Other comments related to the calculation of CIs for precision (n=2), one indicated use of a graphical summary (n=1) and four cautioned against the use of testing.
AE, adverse event; CI, confidence interval; CTU, clinical trial unit.