Table 1.
Discrepancies of registered and reported (published) outcomes
| Major outcome discrepancies between registry and publication | Number (%) |
|---|---|
| RCTs with major discrepancies | |
| all prospectively registered RCTs | 27 (100%) |
| RCTs with major discrepancies between protocol and publication | 10 (37.0%) |
| Type of major discrepancies between registry and publication | |
| all major discrepancies | 19 (100%) |
| registered primary outcome was reported as a secondary outcome in the published article | 6 (31.6%) |
| new primary outcome was introduced in the published article | 5 (26.3%) |
| timing of assessment of the registered and published primary outcomes differs | 5 (26.3%) |
| registered primary outcome was omitted in the published article | 2 (10.5%) |
| published primary outcome was registered as secondary outcome | 1 (5.3%) |
| Relation of primary outcome discrepancy on statistically significant results in RCTs | |
| all major discrepancies | 19 (100%) |
| unable to evaluate | 4 (21.1%) |
| evaluable | 15 (78.9%) |
| discrepancy favours statistical significance | 12 |
| discrepancy does not favour statistical significance | 2 |
| unclear | 1 |
| All evaluable published RCTs with discrepancies favouring statistical significance | 8 (80%) |