Adequate Reporting in Published Study Protocols of Randomized Controlled Trials
Conditions
Brief summary
Transparent and accurate reporting is key, so that readers can adequately interpreting the results of a study. The aim of this project is to evaluate whether reminding peer reviewers of the most important SPIRIT reporting items (including a short explanation of those items) will result in higher adherence to SPIRIT guidelines in published protocols for RCTS. During the standard peer-review process, peer-reviewers will be randomly allocated to use either (i) a short version of the SPIRIT checklist including the ten most important and poorly reported SPIRIT items ; or (ii) no checklist. The aim is to find an intervention which improves the reporting, making it easier for readers to adequately interpret the presented articles.
Detailed description
The full protocol is available on Open Science Framework where the study was prospectively registered: https://osf.io/z2hm9
Interventions
OTHERSPIRIT checklist plus usual practice
Peer reviewer will be reminded of the following 10 CONSORT items (including a short description): Outcomes (12) Sample size (14) Recruitment (15) Allocation implementation (16 c) Blinding (17 a) Data collection methods (18 a) Data collection methods - retention (18 b) Statistical methods (20 a) Population analysed (20 c) Access to data (29)
OTHERUsual practice
Peer review as it is usual practice at the journal
Sponsors
University of Oxford
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
OTHER
Masking
SINGLE (Subject)
Masking description
Authors will be blinded to the intervention allocation. Editors will not be informed about the randomisation. To avoid potential bias, peer reviewers and authors will not be informed of the study hypothesis, design and intervention. All submitting authors and invited reviewers are routinely told that BMJ has a research programme and their article/review may be included in this research. Outcomes will be assessed in duplicate by blinded outcome assessors
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
for journals: It was planned from the beginning that this study will only be conducted at the BMJ Open as it publishes numerous protocols. Hence, no inclusion criteria for the journal were specified. Inclusion criteria for manuscripts: All submitted manuscripts sent out for external review that described protocols for RCTs.
Exclusion criteria
l for manuscripts: We excluded (i) Manuscripts which were clearly labelled as a pilot or feasibility study (ii) Studies randomizing animals or cells (iii) Separate publications of data analysis plans Included peer-reviewers: -Peer reviewers that were invited following usual journal practice
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Completeness of reporting | Through study completion, an average of 1 year | The primary outcome of this study will be the difference of the mean proportion of adequately reported items of the 10 most important and poorly reported SPIRIT items between the two intervention arms. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Completeness of reporting | Through study completion, an average of 1 year | Mean proportion of adequate reporting of the 10 most important and poorly reported SPIRIT items, considering each sub-item as a separate item. |
| Time from assigning an academic editor until the first decision (as communicated to the author after the first round of peer-review). | Through study completion, an average of 4 months; will be assessed from routinely collected data | — |
| Proportion of articles directly rejected after the first round of peer-review | Through study completion, an average of 4 months; will be assessed from routinely collected data | — |
| Proportion of articles published | Through study completion, an average of 9 months; will be assessed from routinely collected data | — |
Other
| Measure | Time frame | Description |
|---|---|---|
| Peer reviewer comments for any reference to SPIRIT and trial reporting | Assessed from available peer-reviewer responses; on average 3 months after randomisation | For journals where peer reviewer comments are subsequently published alongside the published article, we will examine the peer reviewer comments for any reference to SPIRIT and trial reporting. |
Countries
United Kingdom
Outcome results
None listed