Table 4.
Reasons for omitting one or more outcomes per trial; based on 69 survey responses with unreported efficacy or harm outcomes in randomised trials published in December 2000 and cited in PubMed
|
No (%) of trials
|
||
|---|---|---|
| Reason* | Efficacy outcomes (n=59) | Harm outcomes (n=16) |
| Space constraints: | 28 (47) | 4 (25) |
| Journal imposed | 11 (19) | 1 (6) |
| Author imposed | 21 (36) | 3 (19) |
| Not clinically important | 22 (37) | 12 (75) |
| Not statistically significant† | 14 (24) | 8 (50) |
| Not yet submitted | 13 (22) | 1 (6) |
| Not yet analysed | 10 (17) | 1 (6) |
*
Reasons are not mutually exclusive (>1 may have been given per trial).
†
P≥0.05 in all inter-group comparisons.