I suggest you follow the PRISMA statement to perform and report your meta-analysis.

Title: Identify the report as a systematic review, meta-analysis, or both.

Abstract: When "PE" first appeared, a full name is needed. "to systematically review and the efficacy and harm of dapoxetine", no "and". Statistical methods need to be described. Notice how to use short name and full name. "95% CI" should appear. 

too many keywords.

Methods: I think that inclusion and exclusion criteria and clinical outcomes should be clearly described here. I think search strategies can be put in supplemental materials. Characteristics of included studies are important information and should be in the main text.

Results: "safety" is more common used than "harm".

I am sorry that I can only give you a little advice, as I am not an expert in the relevant clinical field and my English is not good enough. I hope the comments above may help you. 

Great suggestions Wenjun! I'll use these when revising the manuscript.

Hi Hitesh, 

Thanks for your thoughts. I believe many eyes on the manuscript will be a good thing. With all the comments, we all help this manuscript becoming better. A subsequent publication submission (for regular peer review) may be possible in a higher impact journal. Feel free to give your comments too! You can find a list of comments already made by peers in the additional resources box. You can email me at [email protected]

My concern is that, although PLoS One is an open-access journal, do they allow for this type of review process. Generally, the peer-review process is supposed to be held confidential and not discussed with others.

one approach might to be suggest names of external peer reviewers who you do not have a personal knowledge of, but who are familiar with either the methodology or field.  One of my colleagues would be able to peer review for you, but im dubious about asking him to present open comment outside the conventional system.  If you provide One with some names of international researchers, that would i think give the greater appearance of rigor.

I discussed this open review with PlosONE. They encourage but say "In the end, the academic editor chooses wether or not to use this information for evaluating the manuscript."

Hi Frederik,

Your approach sounds to me a very modern way of disseminating research results. I would not really be too surprised if PlosONE rejects the paper on basis of the unconventional review process. However, medical journals should not keep holding conservative these approaches. The traditional peer review process has obvious drawbacks, including the limited time and availability of reviewers. I suggest you'll be the first to publish a paper yourself with open, public peer review comments addressed. Reseachgate appears to me like a good vehicle for this experiment, the developers of Researchgate would be wise to implement a tool that facilitates this process. I wish you good luck.

Thank you Wilco, I agree with you. ResearchGate has the potential to facilitate many aspects of modern scientific conduct. In cases of paradigm transformation, my strategy is to use both the old and the new. So, I would publish the paper on ResearchGate and in a regular (open access) journal. I invite you to review my current manuscript. We currently have two reviews. A third would be nice. Your would help to make a statement that 'new style' reviewing already is possible and feasible. Different scientist engaging into peer review voluntarily can also stir media attention. This in turn activates the debate as to what kind of review process would benefit science and humans best. Thank you.

Dear all, how interesting!

I am a Section Editor and Academic Editor for PLOS ONE.

I strongly disagree with Wilco's comment that:

"I would not really be too surprised if PlosONE rejects the paper on basis of the unconventional review process."

The criteria for publication in PLOS ONE are clear and are available here:

http://www.plosone.org/static/publication

One of the criteria here is that results should not have been published elsewhere, BUT it is also noted that:

"We will consider manuscripts that have been deposited in preprint servers such as arXiv or published in a thesis. "

I would argue that ResearchGate is such a forum.

I agree with the advice that you received from the editorial office, that at the end of the day it is up to the academic editor to decide whether the manuscript should be published or not, but there is certainly no clear rationale for rejecting a manuscript that has been subject to open review on a preprint server (as far as I am aware). 

Good luck Frederik!

Frederik, all is not lost, let's engage in your new style reviewing. Mine is underway. Still hope PlosONE finds the reviewers (But why do it again...?). And thanks for your clarification Andrew. 

Thanks Andrew. I hope to submit an even better manuscript thanks to the open review process. Ideally, this may then speed up conventional review. I am wondering, if I revise my manuscript now and put it on open review again. Would the same reviewers be willing to give -if any- comments? What if a new reviewers pops up with good comments? When is the review process to end? Who decides this? Interesting stuff...

Frederik: what is the status of your ms with PLOS ONE? I would liaise with the editorial office and highlight how you have improved the manuscript by open review and perhaps even submit a traditional rebuttal  to reviewers (from ResearchGate) to the PLOS ONE academic editors  for their review via the office, I.e. Comment - response - comment - response. Perhaps the AE would then make a decision by themselves.  Make sure you document everything carefully and even direct PLOS ONE to ResearchGate to expedite the decision making process. You're breaking new ground, that's for sure! 

Frederik, Just uploaded my review. Good luck with the submission, I would certainly be available for a second review, which I honestly think is necessary considering the improvements that are required. I don't mind, either for PlosONE (@Andrew) or a second presubmission round.

Thanks for your review Wilco! @Andrew That seems like an excellent suggestion for a modern workflow :)