Systematic review and meta-analyses of studies analysing instructions to authors from 1987 to 2017

Scholarly reporting practices vary across disciplines and even within journals of the same subdiscipline. Instructions to Authors (ItAs) provide guidelines for manuscript submission, including requirements regarding research integrity. Despite the long history of scholarly publishing, research on the evolution and variability of ItAs is limited. This study aimed to synthesize findings from all studies analyzing ItAs from multiple journals to assess changes over time and identify factors associated with the coverage of research integrity topics. The focus was on six key research integrity topics: authorship, conflicts of interest, data sharing, ethics approval, funding disclosure, and ICMJE Uniform Requirements for Manuscripts (URM). The researchers hypothesized that the coverage of these topics would vary across journals based on factors such as publication year, country, database indexation, impact factor, discipline, and sub-discipline.

The introduction cites several studies that highlight the variations in research reporting across disciplines (references 1-8). These studies indicate differences in manuscript format, structure, and level of detail. The authors also note the multiple roles of ItAs, including outlining journal requirements, describing review processes, promoting standards, and addressing research integrity (references 9-19). The scarcity of research focusing on ItA evolution and variability across disciplines is emphasized (reference 20).

This study followed PRISMA guidelines. A systematic review was conducted using MEDLINE, Scopus, and Web of Science databases, supplemented by Google Scholar and reference lists. The inclusion criteria encompassed all studies analyzing ItAs from more than one journal. A total of 153 studies published between 1987 and 2017 were included. Data extraction included journal characteristics (number of journals analyzed, sampling method, discipline, country, year of ItAs, etc.), topics analyzed, and the percentage of journals addressing each topic. Meta-analyses were performed for the six research integrity topics using the logit transformation method, with random-effects models used to estimate summary percentages. Subgroup analyses and meta-regressions were conducted to explore the heterogeneity.

The study identified 153 eligible studies. The number of studies analyzing ItAs increased sharply after 2002. Over 100 different topics were analyzed, grouped into 32 major topics. Research ethics and reporting guidelines were the most commonly analyzed. Narrative synthesis revealed 12 primary study objectives; the most frequent being to assess the presence/absence of specific topics in ItAs and to analyze the relationship between ItA coverage and reporting in published papers. Time-trend analysis (across 11 studies) showed an overall increase in topic coverage, especially for data deposition, peer review process description, and CONSORT guideline recommendations. Meta-analyses revealed large between-study heterogeneity for all six research integrity topics. Six factors explained this heterogeneity: 1) Time: overall increase in topic coverage. 2) Country: significant differences in coverage across nations. 3) Database indexation: journals indexed in higher-impact databases tended to address topics more often. 4) Impact factor: a weak association between higher impact factors and coverage of authorship, conflicts of interest, and URM. 5) Discipline: Health Sciences journals showed considerably higher coverage. 6) Sub-discipline: general journals typically showed better coverage than subdisciplinary journals. Additional analyses, though limited by data availability, explored associations with language, publishers, and endorsements of ICMJE, CONSORT, and WAME. Figure 3 illustrates the trends over time for the six research integrity topics.

The increase in ItA coverage over time likely reflects improvements in scholarly methods, enhanced teaching of standards, and increased attention to research integrity. The higher coverage in Health Sciences journals is likely due to stricter regulations on human and animal experimentation and increased focus on editorial processes. However, significant variations in coverage remain across countries and sub-disciplines, indicating that many journals still lag in adopting best practices. The study highlights the need for a comprehensive database to compare ItA requirements and monitor adherence to guidelines. Limitations of the study include the focus on research integrity topics, selection bias (mostly Health Sciences journals), and the absence of tools for assessing the risk of bias. The binary assessment of topic addressing (present/absent) neglects the nuances in how topics are addressed. The limited number of studies analyzing country, language, and discipline effects also represents a limitation.

The study demonstrates an overall increase in the addressing of six research integrity topics in ItAs over time. However, substantial heterogeneity remains across countries and disciplines. A comprehensive database comparing ItA requirements and monitoring publication adherence is needed to improve research quality and transparency. Future research should explore additional factors influencing ItA content and delve into the nuances of topic implementation.

The study's limitations include the focus on research integrity topics, the prevalence of Health Sciences journals in the analyzed sample, and the lack of a standardized risk of bias assessment tool for studies analyzing ItAs. The binary approach to analyzing topic coverage may oversimplify the complexity of how individual topics are addressed within ItAs. Additionally, there is limited data available for some of the factors examined, especially concerning country, language, and discipline effects.