Reader Comments
Post a new comment on this article
Post Your Discussion Comment
Please follow our guidelines for comments and review our competing interests policy. Comments that do not conform to our guidelines will be promptly removed and the user account disabled. The following must be avoided:
- Remarks that could be interpreted as allegations of misconduct
- Unsupported assertions or statements
- Inflammatory or insulting language
Thank You!
Thank you for taking the time to flag this posting; we review flagged postings on a regular basis.
closeMisleading abstract
Posted by AdamJacobs on 25 Jul 2013 at 13:00 GMT
The abstract of this article states:
"Our results show that most trials do not report results"
This is highly misleading. What the results show is that most trials do not link to results in a structured manner. As was acknowledged in the full paper, it is possible that many trials did not link to results in a structured manner, and yet the results still exist and could be found by a manual literature search.
This study has not shown what proportion of trials do or do not report results.
RE: Misleading abstract
vojtech_huser replied to AdamJacobs on 02 Aug 2013 at 15:02 GMT
Author reply:
We acknowledge that the sentence, when taken out of the abstract context, may be viewed as misleading; however, in the second sentence of the abstract, we clearly define that we only considered linked articles in our analysis.
We also discuss this issue (as pointed by the commented) in the full article.
Furthermore, we tried to estimate how often an unlinked result article exists. In our earlier work (Ref 13; Huser V, Cimino JJ (2012) Precision and Negative Predictive Value of Links between ClinicalTrials.gov and PubMed. AMIA Annu Symp Proc, 2012: 400-408. PubMed: 23304310; available at: https://www.researchgate....) we measured how often an unlinked result article exists. This is also mentioned in the ‘Limitations section’). The article says:’ A prior study of negative predictive value of trial-article links [13] shows that 44% of trials with no linked result articles indeed have an unlinked result article that can be found by manually searching PubMed.’
In addition, it is possible to calculate a corrected estimate: If we extrapolate the unlinked article rate of 44% to our linked-only results - that for the trials without result article (2477/8907; 27.8%), indeed 44% have unlinked result article ((2477*0.44=1089 trials), we would get a corrected rate of having a linked or unlinked result article of 40% [(2477+1089)/8907] instead of reported linked-article only rate of 27.8%.