In recent decades, there has been an exponential increase in the number of biomedical publications and, at the same time, concern has grown about the quality of their reporting. In the 1990s, a series of initiatives started to be developed with the aim of improving the quality of the content of scientific publications, at first focused on randomised controlled trials. These initiatives took shape in a series of guidelines that offered structured recommendations on the information that should be included in a paper depending on the study design.

The English term used to refer to them, reporting guidelines, has been translated to Spanish in different ways: “listas-guía”, ‘lista de directrices”, “directrices para la comunicación”, “directrices de presentación de informes”… This has given rise to some confusion as to what they are and what their purpose is, which is to improve reporting in published studies. They are also frequently referred to with the broad term of “checklists”, confusing them with other types of guidelines or tools whose purpose is to assess bias and the applicability of previously published reports. There are multiple tools for these purposes. One example would be the tools used by the authors of systematic reviews to assess the risk of bias of the primary studies included the review (Quadas-2,1 RoB-2,2 Newscastle-Otawa scale3…). There are also the tools intended for the readership, such as the user’s guides to the medical literature of the JAMA (Journal of the American Medical Association),4 the guidelines of the BMJ (British Medical Journal), based on the recommendations of the Evidence-Based Medicine Working Group, or the CASPe (Critical Appraisal Skills Programme Español) tools.5 The GRADE (Grading of Recommendations, Assessment, Development, and Evaluations) approach goes a step further for, in addition to the assessment of the quality of the evidence, grading the strength of recommendations in the development of clinical practice guidelines, systematic reviews or assessment of health care technologies.6

When it comes to reporting guidelines, which in the Spanish version of this article we called “guías para la elaboración de manuscritos” (based on their definition),7 their main goal is to guide authors in providing the best possible description of the methods used in each of the phases of research accurately and transparently. They are meant for manuscript authors, to be applied during the drafting of the paper, and also for journal editors and reviewers. It would be fair to say that the intent of these tools is to promote excellence in scientific reporting, regardless of the quality of the study. There can be studies that have an optimal design, implementation and analysis of results, but which are poorly reported because the published content is missing key information regarding the methods and results. This can create problems at a later stage in the assessment of study bias or applicability and in the inclusion of the study in a systematic review, all of which could have been easily prevented with a more detailed description.

The first reporting guideline, which is also the best known today, was the CONSORT Statement8 (CONsolidated Standards Of Reporting Trials, focusing on randomised trials). It was followed by others such as STARD9 (Standards for Reporting Diagnostic accuracy), PRISMA10 (Preferred Reporting Items for Systematic reviews and Meta-Analyses), STROBE11 (Strengthening the Reporting of OBservational studies in Epidemiology)… and extensions of these guidelines focusing on specific aspects within the type of study addressed by the guideline.

The proliferation of this type of guidelines spurred yet another initiative, EQUATOR (Enhancing the QUAlity and Transparency Of health Research, https://www.equator-network.org/).12 Active since 2006, its mission is to collect all valid reporting guidelines for any care setting, promote their application and assess the scientific literature on the subject (Table 1). It is the first attempt to coordinate efforts to address inadequate publication practices at a global scale. It encompasses researchers, medical journal editors and reviewers, guideline developers, research funding groups and other collaborators interested in improving the quality of scientific research and publications. In addition, the “Toolkit” section of the website13 offers practical help and resources to support professionals in different activities.

In this article, we present the guidelines that we consider most relevant based on the type of study and the frequency of their use for authors who seek to publish articles, peer reviewers and readers of scientific publications overall.

Well-designed and implemented randomised controlled trials (RCTs) are the gold standard for producing primary evidence to guide clinical decision-making regarding health care interventions.

Alarmed at the overwhelming evidence that the quality of RCT reporting tended to be far from optimal, a group of researchers and editors developed the Consolidated Standards of Reporting Trials (CONSORT) statement to improve the quality of reporting of RCTs for publication. The first version of the statement was published in 1996. Two revisions have been published since, the CONSORT 2001 and the CONSORT 2010 versions, with additional publication of several extensions. In 2022, very interesting extensions were published that delved in more detail on the reporting of intervention outcomes (The CONSORT-Outcomes 2022 Statement)14 and the adverse events (CONSORT Harms 2022 Statement).15

The CONSORT 2010 Statement, checklist, participant flow diagram and the indispensable explanation paper (CONSORT 2010 Explanation and Elaboration) are available at the EQUATOR8 website.16 The checklist comprises 25 items (some with subitems) that, due to their significant relevance, should always be included in any article reporting on a RCT. They are structured into 6 sections: Title and abstract, Introduction, Methods, Results, Discussion and Other information. The CONSORT Outcomes 2022 extension adds 17 more subitems to the CONSORT 2010 checklist, for greater detail in the Results (item 6a), Sample size (item 7a), Statistical methods (item 12a) and Ancillary analyses (item 18) sections.

Table 2 presents the CONSORT 2010 checklist, the Spanish translation of which is available in the Spanish version of this article.

Although a randomised controlled trial is the best possible method to demonstrate the efficacy of an intervention, it is not always feasible. We often encounter articles in the scientific literature reporting results obtained in nonrandomised clinical trials. These studies are conducted when carrying out a RCT would not be ethical, when the objective of the study is to analyse effectiveness in real-world clinical practice or the cost-effectiveness of a health care intervention or when the causal chain of the effect of the intervention is highly complex. Under these circumstances, it is appropriate to conduct nonrandomised trials, whose value has yet to be fully recognised. To be sure, RCTs offer the significant advantage of controlling biases or systematic errors that were not taken into account in the study design due to lack of previous awareness, but the lack of randomisation in nonrandomised trials can be compensated with Bayesian methods, used to quantify the degree to which the findings could be explained by potential selection biases.17

Many of the decisions made in our daily clinical practice or in public health policy are based on evidence from nonrandomised trials. Therefore, transparency, thoroughness and clarity are essential in reporting these studies. The Transparent Reporting of Evaluation with Non Randomized Designs (TREND) was developed to provide guidance on the subject.18,19

The first version of TREND was published in 2004 in the American Journal of Public Health17 on the initiative of the Center for Disease Control and Prevention (CDC)20 and several journal editors and representatives. The purpose was to provide recommendations for authors, reviewers and editors on the information that should be included in the reporting of nonrandomised studies. The TREND checklist comprises 22 items structured into 5 sections: Title, Introduction, Methods, Results and Discussion. Table 3 presents the most recent update of the checklist.

TREND is considered a complement to CONSORT (the guideline for randomised clinical trial reporting), and the two checklists have some overlapping elements and some distinct elements specific to each of them. The information presented in this section was updated by the CDC in September 2018 and by EQUATOR21 in November 2021.

One of the proposed definitions for epidemiology is “study of distribution and determinants of health-related states among specified populations and the application of that study to the control of health problems”. The aim of epidemiological studies is to answer questions regarding the frequency and trends in a given disease, which preventive interventions are most effective or the benefits of modifying certain diagnostic or therapeutic approaches, among others.

To answer these questions, the field of epidemiology has developed a methodology of its own that is based on observation. Thus, the design of an epidemiological study depends on the question that researchers ask regarding the population and the topic under study (disease, risk or protective factors, health care interventions, etc).22 For instance, the most appropriate design to study the prognosis of a condition would be a longitudinal observational study, while in aetiological research it would be a cohort or a case-control study.

Observational studies, by design, are intrinsically likely to suffer from systematic errors and biases that affect their internal validity, which in turn (in addition to other factors) may affect the external validity or generalizability of their results. Furthermore, articles reporting on observational studies frequently fail to present sufficiently clear and detailed information on the conducted research, which hampers the assessment of the evidence.

The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement was developed with the aim of improving the reporting of observational studies, issuing a series of recommendations on the information that should be provided in a complete and accurate report of an observational study. The STROBE Statement was published in 2007 and it is available at the EQUATOR website11 or its own site.23

It provides checklists for the main study designs of analytical epidemiology: cohort, case–control, and cross-sectional studies. A combined checklist has also been published that can be used to any of the three designs (Table 4),24 as well as an article explaining each item in detail, offering a rationale and published examples (STROBE: explanation and elaboration).25

Each of the checklists comprises 22 items structured in 6 sections: Title and abstract, Introduction, Methods, Results, Discussion and Other information (Funding). In the combined checklist, 18 items are common to all designs and 4 are specific to each design.

Reports of diagnostic accuracy studies should present all the key information readers may need be able to appraise the importance and credibility of their conclusions, in addition to a complete and rigorous description of their results. Only when all the necessary information is provided in the publication is it possible to assess the potential biases of the study and its generalizability at a later time. Yet, multiple reviews have evinced that diagnostic accuracy study reports frequently fail to report essential and adequate information on the selection of participants or the study design or the full results. In addition, the conclusions regarding diagnostic accuracy may be overly optimistic.26

In 2003, the Standards for Reporting Diagnostic Accuracy (STARD) statement was published following the example of the CONSORT statement. This guideline was published in prominent general and specialised medical journals. It included a 25-item checklist, a flow diagram and an explanatory document clarifying the meaning and rationale of each of the items and a short summary of the available evidence they were based on.27 Its publication prompted several editorials recommending its acceptance by journal editors and review committees.

In 2015, the statement was updated to include new evidence on the sources of bias and variability and to facilitate its use.

Table 5 presents the 30 items of the STARD 2015 checklist, of which the translation to Spanish is available in the Spanish version of this article. The items are structured into the usual sections found in original articles: Introduction, Methods, Results and Discussion, in addition to one section at the start (Title and abstract) and another at the end (Other information).

Among the explanation and elaboration papers of the STARD 2015 statement, available at the EQUATOR website,12 the reader can find the original article by Cohen et al.28 with commented examples and a detailed explanation of each of the 30 items in the checklist. This information is provided to facilitate the writing of diagnostic accuracy reports and ensure they include all the relevant information. It is also meant to help journal reviewers and editors as well as readers to ensure that submitted and published articles are sufficiently detailed.

The systematic review (SR) and meta-analysis is a type of study that synthesises the available evidence on a specific topic. The methodology used in this type of studies follows a clear, standardised and reproducible protocol to ensure the quality, consistency and transparency of the review process.29 When reporting results, it is very important to describe the development process in full and in detail, including why it was done, how and what was found. The PRISMA 2020 Statement provides recommendations to ensure that this happens, reflecting the most recent advances in methods to identify, select, appraise, and synthesise studies eligible for inclusion in systematic reviews.

All the information on PRISMA 2020 can be found in its own website,30 including translations to Spanish of some of the documents. Both its website and the EQUATOR website31 offer access to the PRISMA 2020 Statement, the 27-item checklist, the flow diagram template and, for those interested in more in-depth information, a more thorough and detailed version of the checklist and the PRISMA 2020 explanation and elaboration paper.32

The 27 items of the verification checklists are structured into 7 sections: Title, Abstract, Introduction, Methods, Results, Discussion and Other information (including financial support).

Table 6 presents the PRISMA 2020 checklist33 and Fig. 1, a template PRISMA flow diagram.

Figure 1.

PRISMA 2020 flow diagram template.

Source: Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ. 2021;372:n71.

(0.61MB).

The process of publishing a scientific paper culminates when it reaches the target readership. Readers, however, should not be mere uncritical recipients of the information at their disposal. If the article has been written following the recommendations of the applicable reporting guideline, the reader will have sufficient information for, upon reading it critically, appraise the potential for bias in the evidence and whether the results can be applied locally.

Multiple tools are available to support the critical appraisal of published reports. The Critical Appraisal Skills Programme (CASPe)5 offers workshops on critical appraisal skills and tools to aid critical appraisal in Spanish, available at the website https://redcaspe.org/materiales/. At this site, the reader can find checklists composed of relevant questions that can guide the reading of reports for most of the study designs addressed in this article. These questions have only 3 possible answers: “yes”, “no” and “can’t tell”. A “can’t tell” answer usually means that the necessary information to answer the question is not available in the report. In some instances, it is due to deficiencies in the study methodology, but in others the problem stems from deficient reporting. The widespread application of reporting guidelines would reduce uncertainty and facilitate the appraisal of research publications.

With this article, the Working Group of Evidence-Based Paediatrics encourages future authors to use reporting guidelines to report their findings in a complete and structured manner. This will facilitate the subsequent publishing process of the article in addition to its critical reading by the audience.

The authors have no conflicts of interest to disclose.