Skip to main content

Table 2 Registry-publication concordance and discrepancies in discordant studies

From: Registration and primary outcome reporting in behavioral health trials

Studies

n (%)

Concordant vs. discordant studies ( n =53)

 Concordant

16 (30)

 Discordant

37 (70)

Discrepancies identified in discordant studies ( n =37)

 Primary outcome(s) described in registration was/were not explicitly defined

19 (36)

 Registration did not state how primary outcome(s) would be measured/codified

16 (30)

 Published manuscript(s) reporting results of registered trial include at least one unregistered primary outcome

16 (30)

 Primary outcome(s) described in registration was/were not measured according to registration prescribed timing protocol

14 (26)

 Published manuscript(s) reporting results of registered trial did not report on all registered primary outcomes

11 (21)

 Secondary outcome(s) described in registration was/were promoted to primary outcome(s) in published manuscript(s) reporting trial results

5 (9)

 Registration prescribed timing protocol for measuring primary outcome(s) was not adequately defined

4 (7)

 Primary outcome(s) described in registration was/were de-emphasized or demoted to secondary outcome(s) in published manuscript(s) reporting trial results

4 (7)