The process for the external evaluation of scientific articles (also known as peer-to-peer review, peer review, or refereeing) refers to the editorial practice established by many scientific journals of submitting manuscripts they receive for expert review on the topic at hand. These experts assess the merits and deficiencies of the manuscript. and help the editor to make a decision as to whether or not to publish the manuscript, thereby also helping to boost its quality. Although empirical knowledge of this process is still limited, there exists consensus that it is the best possible option to scrutinise the quality of scientific publications.

Although the discussion on the expedience of the process is far from over, at Archivos de Prevención de Riesgos Laborales (Occupational Hazard Prevention Archives) all the original manuscripts received, and a good part of the manuscripts with other formats, are subject to expert review, a practice embraced by the main national and international journals.

In order to optimise the review process at Archivos, the general guidelines that external manuscript reviewers should follow for our journal are set forth below.

General aspects

The external evaluation of manuscripts is a voluntary process that is recognised by the annual publication in the journal with the names of the reviewers who participated during the period. In addition, Archivos reviewers will receive a certificate at the end of the year recognising their collaboration with the journal.

A request for the evaluation of a manuscript may be freely accepted or declined. However, if one accepts, it is essential to meet the deadlines established for the review (three weeks from reception of the manuscript) and, if one declines, it is also essential to immediately notify the journal's secretariat (within 48 hours). In the event one declines, referring the author to another expert or experts who may be appropriate for the evaluation of the manuscript in question will be useful, although it will be the editor who ultimately decides whether to remit the manuscript to them or not.

The reasons for declining an invitation to evaluate a manuscript may include not being able to dedicate the necessary time to complete the work on time, a lack of familiarity or experience with the subject of the manuscript, or conflicts of interest that could compromise neutrality and rigor necessary for the review process.

Until published, a manuscript is a confidential document. Therefore, its contents, and any aspect related to evaluation by third parties, should not be discussed. The only authorised contact people for any matter related to the evaluation process are the director and editors of Archivos.

The evaluation process at Archivos is anonymous; that is, the evaluator does not know the author of the manuscript's identity or institutional affiliation. Evaluators may remain anonymous to the authors, or, if they prefer, sign their comments.

The editorial decision concerning the manuscript (acceptance, rejection contingent upon changes, rejection), once the comments of the external evaluators have been received, is made by the corresponding editor of Archivos, and conveyed to the external evaluator, who also receives a copy of the comments of any other evaluators who may have reviewed the manuscript. This decision is the editor's ultimate responsibility, and is based on both the comments and suggestions of the external evaluators, and on the journal's own editorial criteria.

Completing the evaluation

The external evaluation process at Archivos involves the completion of three documents by the evaluator:

1. The evaluation sheet, which is sent to the evaluator along with the manuscript, and which must be completed in full; that is, all the relevant parts, depending on the nature and characteristics of the work (this sheet is not shown to the authors and is used only internally by the editors of Archivos).
2. Comments for the editor (like the evaluation sheet, this part is not disclosed to the authors either).
3. Comments for the authors (this part is the only one sent to the authors, without the identity of the evaluator appearing on it, unless the evaluator expressly authorises this).

The steps necessary to successfully complete the assessment and to properly fill out all of these documents include:

1. The complete initial reading of the manuscript, including title, abstract, text, tables and figures, and bibliography, just as one would if it were a published work. At the same time as this first reading is carried out, it will be beneficial to write down, on a separate paper, or indicate on the manuscript, the main aspects of the study (for example, the statement of objectives and main details of the method) and any questions (for example, sentences that are difficult to understand, or poorly founded conclusions). On the second reading you will be in a position to delve into the actual evaluation and the different aspects of it.
2. The actual evaluation of the manuscript. The issues that should, generally speaking, be addressed by an evaluation, although their relevance may vary depending on the type of manuscript, include:

• Appropriateness for Archivos' readers, presenting and justifying, where appropriate, other journals or readerships for which the work is better suited.
• The originality of the work in its field, evaluating its contribution to the occupational hazard prevention field; if appropriate, by citing the appropriate references.
• The adequacy of the title and summary. The title may be too ambiguous, or too long. There may be inconsistencies between the data in the summary, the text, and the tables. Essential parts of the abstract may be missing, such as an articulation of the work's objective, the presentation of numerical data, if relevant; or the conclusions of the work, which are often not entirely consonant with or supported by that which is conveyed in the ext.
• Formal aspects. Is the text easy to read? Is the wording of the sentences and paragraphs clear and easy to follow? Are there sentences that are difficult to understand? The evaluator should check whether the text respects the usual structure; for example, in the case of an original article, whether the contents of the Introduction, Methods, Results and Discussion are properly ordered. Tables and figures must be understandable without turning to the text, and complement the information in it, not repeat it. It is also worth assessing the length of the text. If excessive, the authors may be encouraged to delete expendable paragraphs or parts of the text. The manuscript must observe, throughout, the "Instructions for authors" of Archivos de Prevención de Riesgos Laborales.
• Ethical aspects, including both ethical issues of the work itself (for example, the privacy of the study's subjects, informed consent, and other ethical principles affecting medical research on human beings) and the ethical aspects of its publication (for example, related to repetitive publication or potential conflicts of interest).
• The content and validity of the scientific information.

Objectives: Are they properly presented and defined?

Methods: Does the study's design allow it to achieve its stated objectives? Is the field, population and sample used in the study described sufficiently? Are they suitable for the purposes defined? Are the tools used to collect the information sufficiently described? Are they adequate for the purposes defined? Is the statistical analysis performed adequate?

Results: Are they credible? Has the information to be presented been properly selected? Is the information relevant to answering the research question presented? Is superfluous or unnecessary data presented?

Interpretation and conclusions: Are they adequately derived from the data presented? Are the results of the study evaluated relative to those of other studies? Are the limitations of the study discussed? Are the implications of the work for future research in its field suggested?

Bibliographical references: Are they relevant? Are they current? Are they cited correctly? Are there any important omissions? Are they insufficient? Are there too many?

1. The drafting of the evaluation report. In addition to completing all the relevant parts of the evaluation sheet, depending on the type of manuscript in question, the evaluator must write his comments for the editor and for the authors, taking into account the following guidelines:
   • In the comments for the editor opinions related to the manuscript may be freely expressed. These are comments that are not to be sent to the authors, but that one thinks the editors should receive. For example, a recommendation about whether or not to publish the manuscript should be made, exclusively, in the comments to the editor. A harsh criticism of a certain aspect of the manuscript may be more freely expressed to the editor, it being appropriate to temper the tone of this same criticism in the comments to the authors. Should a situation of serious ethical conflict be detected, this should also be placed, exclusively, in the comments to the editor. The evaluator may also come across particular aspects of the work that he cannot adequately evaluate because he does not have the necessary knowledge or experience (for example, a rare statistical analysis). The right thing to do in these cases is, again, report them in the comments just for the editor.

• In the comments for the authors it is a good idea to include a first paragraph briefly summarising the work (one or two sentences at most), together with an evaluation of its relevance and originality in the field of occupational safety and health, and its aptness for the readers of Archivos. In the evaluation report itself, some evaluators prefer to divide their comments into major and minor, which is often useful for both the editor and authors. The major comments refer to important limitations in the design or content of the work, while the minor ones address issues that are easier to resolve, such as structural problems, or the wording of the text. In general, a series of separate comments is preferable to very long paragraphs dealing with various aspects. Whenever relevant, the location in the text of those aspects that are commented on should be specified, referring to the page, paragraph and phrase and/or section where the problem is found. Numbering each comment greatly facilitates the editor's assessment and the authors' response. It is very important to justify each and every one of the comments made, whether based on specific contents in the manuscript, or on other information, such as evidence from previous studies. Thus, when necessary, bibliographic citations will be included in the evaluation to reinforce arguments made. Finally, a respectful and constructive style should always be used in comments to authors. The function of the evaluator is not only to influence the acceptance or rejection of the manuscript. Rather, a good evaluator can also educate and assist the authors, who are sometimes less experienced, and can benefit from the comments and recommendations in the evaluation, even if the journal decides not to publish the work. Thus, whenever possible it is preferable to say how a problem should be corrected, or could have been corrected, rather than simply indicating its existence.