’ s response to reviews Title : Professional medical writing support and the quality , ethics and timeliness of clinical trial reporting : a systematic review

Title: Professional medical writing support and the quality, ethics and timeliness of clinical trial reporting: a systematic review Authors: Obaro Evuarherhe (obaro.evuarherhe@pharmagenesis.com; obarowbin@gmail.com) William Gattrell (will.gattrell@ipsen.com) Richard White (Richard.White@pharmagenesis.com) Christopher Winchester (chris.winchester@pharmagenesis.com) Version: 1 Date: 17 May 2019 Author’s response to reviews:

Response: We have amended this statement in the conclusion section in the abstract, replacing 'increases' with 'is positively associated with measures of…' Background: I think the background could benefit from some few more words on why the timely publication of trial results is an ethical imperative. I would therefore suggest to link this study in the broader context of the "Value and waste"-debate (see for instance Macleod MR, Michie S, Roberts I, et al.: Biomedical research: increasing value, reducing waste. Lancet. 2014;383(9912). There are currently many initiatives and research groups working on improvement of trials results reporting, some of them are already cited. It would underline the importance of this systematic review.
Response: We have amended the first sentence to highlight the waste reduction element of timely reporting and included a reference to Moher et al, 2016 (Increasing value and reducing waste in biomedical research).
Methods: Is there a reason why you limited the supplementary searches from 2014-17 Response: Supplementary searches were limited to focus on the most recent congresses and keep the workload manageable for this manual element of the search strategy. Supplementary searches of the previous 3-4 years is generally considered reasonable for systematic reviews. Additionally, the supplementary searches were carried out alongside the main searches on 8 March 2018. The manuscript has been amended accordingly.
Methods: Overall, it is not entirely clear how you define quality, ethics and timeliness. How did you extract information from the included studies on e.g. ethics? What would have been ethical aspects? As your results suggest, the only issue in "ethics of publication" is the reporting of nonpre-specified outcomes. How is this distinct from quality (e.g. the CONSORT items)? If you were not able to pre-define what you have been looking for when you searched for "ethics of publication", I would suggest to omit "ethics" as an individual result, but instead subsume the of non-pre-specified outcomes to "quality". To me, even adherence to CONSORT is part of an "ethics of publication". Later in the conclusions section, however, you list some more ethical issues (e.g. transparency on conflict of interest etc.) and you cite two studies. Why have these studies not been included in your review?
Response: For the protocol of this study, we considered 'quality' as overall completeness of reporting. We agree that quality reporting and timely dissemination of clinical trial results are an ethical imperative. Although quality, ethics and timeliness are semantically distinct outcomes, it can be difficult to objectively separate quality and ethics with regards to research reporting. As such, we have combined the quality and ethics results in one section entitled 'quality and ethics of reporting'. However, we think that 'timeliness' is a distinct outcome that should be kept separate.
The two 'ethics' studies described in discussion (Desai et al (2017) and Woolley et al (2011)), were not focused on studies reporting clinical trial outcomes and as such, did not fit the inclusion criteria for this systematic review.
Methods: Page 4, line 93: I would suggest to say "effect" instead of "influence", as the latter is too causal.

Response: We have made this amendment
Results: The results are presented in a nice and readable manner. Only point is the unclear operationalisation of "ethics of publication". However, I wonder if in any of the reviewed studies the authors also investigated how quality of reporting with or without PMWS differs in relation to study characteristics (e.g. phase I/II versus phase III trials). Is there any evidence on this?
Response: We agree that this is an interesting question and could be the focus of a further study. However, none of the included studies separated RCTs by phase. Additionally, variations in quality of reporting with study characteristics was beyond the scope of this study and has such has not been discussed.
As mentioned earlier, we have combined the quality and ethics results in one section in line with their overlapping definitions.
Methods: Were there no suitable contributions from the AMWA journal, or didn't you look? I think you should mention AMWA either way.
Response: The focus of the supplementary searches was on ISMPP congress proceedings and the journals Medical Writing and The Write Stuff (which are available via the EMWA website), Searching the AMWA journal was not part of the search strategy. We have now mentioned this in the manuscript. Conclusions [previously line 162, currently lines 163-164]: Would you like to speculate on why this may be? Arguments may include word count constraints, desire to get the key results in the abstract at the expense of methodology...? Response: We have added an explanation of why adherence to CONSORT-A may be affected by word-count constraints.

Results
Conclusions [previously line 190, currently lines 192-194]: Again, would you like to speculate on potential reasons for this? Do you think that industry-sponsored research (as indicated by the use of PMWS) undergoes more intense scrutiny by peer reviewers? There are some studies to support this view ( Response: Although this is an interesting question, the results of the current study show a positive effect of PMWS on timeliness and quality/ethics of publication, including adherence to CONSORT and publication guidelines. By specializing in preparation of clinical trial publications, professional medical writers are well placed to aid in the rapid dissemination of trial findings under the direction of the authors. This suggests that PMWS may improve overall publication rates; it is important to note that the role of PMWS is subject to strict publication guidelines that should prevent inappropriately prolific authorship from clinicians who work with PMWS.
Conclusions [previously line 230, currently line 235]: Thank you for mentioning this important point. Rather self-servingly (!) there is a similar analysis of the European ISMPP meeting output, though it has remained a preprint (https://peerj.com/preprints/2499/