Our study demonstrated that journals from the countries in the Eastern Mediterranean had an important presence in global science as measured by their visibility – indexing in major bibliographical databases such as MEDLINE, SCOPUS and WoS, and their impact – citations received by articles they published. It is not possible to compare the scientific output of Eastern Mediterranean journals to journals from other world regions, as this is, to the best of our knowledge, the first study assessing regional journals rather than the scientific output of researchers in the region or individual countries.
Citations to articles came predominantly from the journals outside of the Eastern Mediterranean region (76.8% in WoS and 75.4% in SCOPUS), indicating that research published in the Eastern Mediterranean did not generate only local but rather a much wider interest of the scientific community. The number of journals in the Eastern Mediterranean countries (419 in 22 countries) and the volume of articles published by local researchers confirms the finding of previous studies that local journals have important role in bridging the gaps between research and practice in developing countries [4
]. Our study showed that they also made a contribution to the global research community. Publishing in English (80% of the journals in this study) certainly contributed to more effective communication of research to the global scientific community. Actual circulation of the printed journals could not be assessed because most of the journals (96.3%) did not publish this information. Information about other important editorial and publishing standards, such as requirements about the patient’s privacy rights, policy on managing conflicts of interest, and ethical conduct in clinical and animal research could not be found in more than half of the journals, indicating that IMEMR journals need to work on reaching the excellence in editorial and publishing standards in medical publishing. Indexing in major bibliographical and citation databases was associated with significantly better adherence to publication and editorial standards. This could mean that databases preferentially indexed journals with higher-quality editorial and publishing processes. As we did not follow the editorial and publishing standards in time, it is also possible that indexing in a bibliographical database brought prestige to the journal, which stimulated better editorial and publishing output.
The limitations of the study include its cross-sectional design and the availability of the journal for citation and publication standards analysis. While we were able to assess database indexing of all journals in IMEMR, we had access to 41% of the journals for citation analysis and 58% for the assessment of publication standards, reflecting the regularity and completeness of journals’ submission to IMEMR by publishers or editors. As most of the missing data were from small, local journals, the results presented may be an overestimation of the full IMEMR and present only the visibility of the higher-quality journals in IMEMR. Indexing and citation analysis was also limited by the information recorded in the databases and the practices of individual databases. The major problem we encountered was the inconsistency in journal title indexing. Different databases had different variations of the journal titles, mostly because of inconsistent transcription and transliteration, especially in cases when the title included the name of an institution. Another problem for locating individual articles from IMEMR journals in databases was often a significant delay in indexing, which could underestimate the indexing data obtained in this study. Delays may have been the consequence of irregularity in publishing or in submitting published issues to the database, or an omission or delay by a database producer. Another problem specific for IMEMR journals was in locating individual articles using authors’ names because of differences in transcription, transliteration, and indexing of personal names in different databases. The common practice of database producers to index authors by surname and first (and middle) name initial(s) created problems for complex Arabic names, which do not follow the same logic as most western names [22
]. The final limitation of the study was the fact that the evaluation of publication and editorial standards was inherently subjective in nature and may have influenced the results, especially for journals in Arabic or Farsi languages. Also, the checklist developed by INASP [17
] included standards that were not universal for all biomedical journals, such as the requirement to publish authors’ contributions on the first page of a journal article. Many journals, including JAMA, N Eng J Med, Lancet and BMJ [24
] publish this information at the end of the article body text, before references. Such variations in publishing styles should be kept in mind in interpreting our results.
Problems in reaching global research community
Although our results cannot be generalized outside of the Eastern Mediterranean, they illustrate common problems of journals from developing areas of the world in reaching the mainstream scientific community. We are not aware of similar comprehensive studies of all journals indexed in large regional index in the developing world, although there are studies of single journals [25
] or studies of research outputs from countries [7
] in the Eastern Mediterranean; other world regions [9
] or individual medical fields [26
]; general publication output from countries [27
], continents such as Africa [9
], or developing countries in general [11
Implications for stakeholders in scientific publishing
Despite the limitations of our cross-sectional analysis of IMEMR journals, the results may have important implications for all stakeholders in health research and publishing in the developing world:
, both from the Eastern Mediterranean and other parts of the world, should be aware of the importance of the journals in the region and the research they publish. They should not be afraid to publish in good journals from this region or search IMEMR for specific topics unique for Eastern Mediterranean, the region covering 22 countries with different health care systems, unique diseases, environmental issues and sociodemographic specificities [28
2. Journal editors
could use the metrics from this study to evaluate their own journal in comparison to those that have reached the level of excellence recognized by major international and highly selective databases. They should pay attention to all details of the journal, including the ISSN as perhaps the most important element for the identification of a journal – it should be clearly displayed on the front page and on the masthead page or its equivalent in the electronic edition of the journal. The title should be unique and should be written consistently, as registered with the ISSN registry.Practical advice for journals published in a national language is that they should also clearly register and submit to the indexing databases a parallel official and unique title in English for indexing purposes instead of letting indexing databases resort to free translations. Abbreviations should also be avoided in the titles as they create additional problems because there is no consistency how different databases handle such items.Editors should keep in mind that their work does not end with the indexing into a desired database: they need to periodically make sure that the journal is indexed correctly. Timeliness of indexing, either in IMEMR or international databases, is crucial and is the best way to deal with irregularities and delays in indexing dynamics. The cases we observed where the number of the volume from one year continued to the next year should not happen to journals striving for visibility in and impact on the international scientific community. The quality of the publication and editorial work is also the responsibility of the editor. As our survey showed, publishing excellence can be achieved and maintained, and indexing in international databases is a stimulus for the improvement in adherence to publishing and editorial standards. This requires considerable effort and a high degree of professionalism on the side of journal editors, who are mostly performing their editorial duties on a voluntary basis. However, they can rely on editorial associations and their useful guidelines for good publishing practices [30
]. Editors in the Eastern Mediterranean have a particular advantage of very active regional editorial organization, Eastern Mediterranean Association of Medical Editors (EMAME) and should use their collective strength to improve their individual journals [31
3. Journal owners, which are mostly academic institutions or organizations, should recognize their journals as important tools in the knowledge translation for better health care. They should ensure that editors have resources to produce a journal that would be recognized and respected by the local and global scientific community.
4. National policy makers could use the methodological approach described in our study to evaluate their national journals and develop their strategies how to best support journals for national and international knowledge translation. National activities could range from establishing national editorial and publishing standards for medical journals and ensuring that journals follow them to the best of their ability, to providing training in different aspects of medical publishing and quality assurance.
5. International policy makers, such as WHO, should continue to organize, develop and use locally produced journals and other health materials as a reliable source of information about health in developing countries. For the Eastern Mediterranean Region, good relations with all ministries of health and leading academic and research institutions in the region put the WHO EMRO in a unique position to lead the generation, translation and uptake of research evidence for improved policy and management decisions. Assessment exercises in journal visibility, impact and quality, such as performed in this study, should be repeated every two to five years to follow the development of the journals and develop strategic actions for their improvement.
6. Producers of indexing databases should be aware of the specificities of journals from different regions and develop standards for specific issues. They should also collaborate with other stakeholders to ensure that the communication with the scientific journals is not unidirectional, i.e. journals submitting their journal to the database, but to provide the framework for interaction and continual external quality control.