North-South relations in scientific publications: editorial racism?

Abstracts

O objetivo do estudo foi comentar a possível existência de preconceito editorial entre editores de revistas científicas de países do Norte contra autores do Sul. Destacou-se que em estudo por métodos bibliométricos ficou evidenciada a existência de um importante desequilíbrio entre a produção científica de pesquisadores de países de alta renda ("Norte") e daqueles trabalhando em instituições de países de renda média ou baixa ("Sul"). Há uma percepção generalizada entre autores brasileiros de que, em parte, isso seria devido a preconceito de editores de revistas internacionais contra autores do Sul - 75% de uma amostra de 244 autores que responderam a inquérito acreditam que exista preconceito. Essa impressão é reforçada pela observação de uma minoria dos membros de conselhos editoriais das principais revistas na área de saúde proveniente do Sul. Embora o preconceito possa explicar parte do problema, há também questões especificas e remediáveis que podem aumentar a probabilidade de publicar no exterior. Essas incluem investir na qualidade do texto e da redação, e mostrar empatia com editores e leitores, sinalizando claramente a contribuição que o artigo pode trazer para a literatura internacional. Finalmente, é abordada a questão de onde publicar: em periódicos nacionais ou internacionais. Foram propostos seis tópicos que devem ser levados em conta nessa opção: idioma e público-alvo; tipo de contribuição ao conhecimento; capacidade de generalização; índice de citações; velocidade de publicação; e acesso livre. O aumento rápido de publicações brasileiras em periódicos internacionais mostra que o preconceito editorial, embora existente, pode ser efetivamente vencido por trabalhos com metodologia sólida e apresentação de qualidade.

Viés de publicação; Políticas editoriais; Saúde pública; Epidemiologia

North-South relations in scientific publications: editorial racism?

Cesar G Victora; Carmen B Moreira

Programa de Pós-Graduação em Epidemiologia. Universidade Federal de Pelotas. Pelotas, RS, Brasil

^{Correspondence}

ABSTRACT

The aim of the present study was to comment on the possible existence of editorial prejudice among the editors of scientific journals from Northern countries against Southern authors. We highlight that a study using bibliometric methods documented an important imbalance in terms of the international scientific production of health researchers from high-income countries (the "North") and those from low and middle-income countries (the "South"). In a survey of Brazilian researchers, three in every four blamed this imbalance, at least in part, on prejudice among international editors. This is supported by the fact that a very small percentage of editorial board members of international journals come from the South. Although prejudice can explain part of the imbalance, there are also specific measures that may increase the likelihood of a paper from the South being accepted in international journals. These include the need to invest in the quality of the written text, and to show empathy with editors and readers, emphasizing the contribution of the manuscript to the international literature. Finally, we discuss whether research carried out in the South should be published in national or international journals, and suggest that there are at least six dimensions to this choice. These include language and target audience; type of contribution to knowledge; generalizability; citation index; speed of publication; and open access. The rapid growth in the number of Brazilian contributions to the international health literature shows that editorial prejudice, although often present, can be effectively offset by research with solid methodology and good-quality presentation.

Keywords: Publication bias. Editorial policies. Public health. Epidemiology.

THE PROBLEM

A comprehensive international review of the distribution of scientific publications in the health field was recently published, analyzing entries in the major bibliometric databases between 1992 and 2001. This review showed that 90.4% of published articles were from high-income countries, 7.9% from middle-income countries (including Brazil, with 0.7%), and 2.7% from low-income countries.⁶ These conclusions have been confirmed by similar studies.¹⁰

Such marked imbalance may be explained by a number of factors. First, scientific production in low and middle-income countries (or "the South") is inarguably inferior to that of high-income countries ("the North"). This is due to a variety of factors, including less funding for research, fewer researchers and laboratories, and the migration of scientists to rich countries, among others. Second, pressure to publish in certain Southern countries is – or at least used to be – much less intense than in Northern countries. A third explanation may be the greater use, especially in the field of public health, of qualitative and observational designs – such as ecological and cross-sectional studies¹ – which are less valued in the international literature than experimental studies.¹⁴ A fourth factor is that the share of the scientific production of the South published in national journals is not adequately assimilated by international databases. Finally, there is the possibility of editorial prejudice, which is the subject of the present article.

There seems to be a generalized perception among Brazilian scientists that editors of journals published in Northern countries – especially English-speaking countries, which concentrate the majority of high-impact journals – may be prejudiced against articles from the South. Reports such as the one transcribed below are frequent:

• A renowned journal published an article based on a North-American cohort, showing that breastfeeding protected against overweight in children and adolescents aged 9-14 years.

  ³

  This survey did not use a population-based sample, and information on breastfeeding were obtained retrospectively from mothers when children were in the age range studied. Information on weight and height were also reported by mothers, and a single outcome (body mass index) was evaluated. Response rate in this study was below 50%. The journal also published an editorial highlighting the importance of breastfeeding for the prevention of obesity. After a few weeks, we submitted to the same journal an article based on a Brazilian population-based birth cohort, in which over 2000 18-year olds were weighed, measured, and had their body composition evaluated by our team using standardized procedures. Information on the duration of breastfeeding had been collected on occasion of three visits taken place during the first four years of life, and 80% of original cohort members were traced at age 18 years. Our study showed no association between breastfeeding and eight different indicators of overweight, obesity, body composition, and height. The article was refused immediately. We appealed to the editor, without success. The article was eventually published in another international journal, whose editors were more receptive.

  ¹³

In academic life, few things are more frustrating than receiving a letter stating that "we believe your article to be more adequate for publication at the local or regional level." Or instead: "we receive a much greater number of articles than we can publish, and the priority rating of your article was not sufficiently high to warrant its inclusion..." The author often feels like Snoopy in Figure 1.

As an attempt to quantify this collective experience, we carried out a survey including the 351 first authors of articles published in 2005 in Brazilian journals Revista de Saúde Pública, Cadernos de Saúde Pública, and Revista Brasileira de Epidemiologia. We asked authors whether, in their opinion, "are the editors of journals produced in rich countries prejudiced against articles submitted by researchers from less developed countries such as Brazil?" Of the authors contacted, 30% did not respond after two attempts. The distribution of valid answers is presented in Figure 2. Three-quarters of respondents agreed completely (25%) or partially (50%) with the statement. Only 8% said to disagree completely with the statement.

During the survey, several colleagues expressed their opinions on the issue. Most commentaries included statements such as "I agree that there is some degree of prejudice, but our articles also show problems and peculiarities that render them insufficiently attractive." Problems most frequently mentioned include deficiencies in English writing and the overall poor quality of some articles. Peculiarities included the study of health problems of local importance, with little international relevance; the fact that many results are very specific to the Brazilian settings (such as, for instance, in evaluations of services and programs); and the interest of Southern countries in issues related to inequities in healthcare, a subject of little interest to many Northern journals.

The perception of this problem is not a recent phenomenon. In 1995, Breilh,² from Ecuador, indicated that the scientific production of Latin America was virtually "invisible" in North America and Europe: "there is an Olympian ignorance regarding books, studies, and instrumental innovations generated in the heart of Latin America."

In the 2002 International Congress of Epidemiology, in Montreal, there was a round table conference on this subject, which included the editors of some international journals. At this meeting, the first author of the present article presented a review of articles published in the International Journal of Epidemiology (IJE) and American Journal of Epidemiology (AJE) in 2000 and 2001. In the former, 11% of first authors of published articles were from Southern countries, versus 6% in the latter. These results were compared to the percentage of members of the editorial boards of these journals that, at that time, worked in Southern countries: none of the 65 associate editors of AJE, and two out of 16 in IJE. None of the 30 board members of Epidemiology were from the South. The scenario was slightly better for the American Journal of Public Health, where two out of nine associate editors were from low or middle-income countries.

This important discussion gained in intensity in 2003, when Saxena et al⁹ published a letter in Lancet showing that, of the 530 members of the editorial boards of the 10 most important psychiatric journals, only four were from the South. This letter was followed by an inflamed commentary by the journal's editor, highlighting the fact that only eight of the 111 board members of the five most important medical journals were from the South.⁴ This editor used the expression "institutional racism," and his commentary was followed by letters and articles demonstrating such imbalance in the fields of tropical health,⁵ pediatrics,¹¹ and psychiatry.¹²

In short, it seems beyond doubt that there is a certain degree of prejudice against Southern articles, but it does not seem fair to blame this prejudice entirely on editors. In the following sections, we shall discuss certain recent trends in international publishing that may contribute to reduce this problem.

WHAT'S NEW?

The aforementioned review of health-related scientific production from 1992 to 2001⁶ also reported a few positive findings. The major time trend detected in the period was a marked increase in the proportion of publications coming from Brazil, China, and the Republic of Korea. Despite this increase, Brazil accounted for only 0.73% of publications in the studied period, still placing behind small countries such as Finland and Denmark.⁶

Much of the Brazilian progress must be credited to the notable expansion in the country's postgraduate and research sector taken place in the last few decades. The fact that many Brazilian authors are breaking the barriers of international publication proves that these obstacles are not insurmountable.

On the other hand, there is also evidence of changes in attitude in some international journals. We list a few of them below, again in a somewhat anecdotic fashion.

The British Medical Journal, when requesting an editorial from a North-American colleague, pointed out that "we are anxious that our editorials should have as much international appeal as possible because of our international readership. A global view of the subject is therefore essential for our readers this editorial should have a coauthor who is from a developing country."

The editor of the American Journal of Public Health, when inviting the first author of the present article to be an Associate Editor, highlighted that "I am looking for an international associate editor to ensure that critical public health research from outside of the United States is brought to the attention of our readers."

Recently, Lancet began to indicate a body of 100 editorial consultants, of which 25 are to be from Europe, 25 from North America, and the other 50 from the rest of the world. Two Brazilians have already been indicated. This journal has been giving priority to issues related to diseases affecting poor countries, and practically all numbers include at least one article on this subject, a fact which has been confirmed by an independent bibliometric evaluation.¹⁰

The measures above, although still timid, may contribute to reduce publication bias against the South. In the next section, we shall discuss what can be done by Brazilian authors and institutions.

WHAT WE CAN DO ABOUT IT

This section is elaborated based on the first author's experience as an editorial consultant and reviewer of international journals, when evaluating a manuscript from the South.

The first problem is the quality of writing. Many manuscripts submitted by Brazilian authors are very poorly written. Poor writing leads the editor or reviewer a priori to tend towards rejection. Writing well in scientific English is very difficult, even for researchers with postgraduate training outside Brazil. Revision by a qualified translator may help, but if the latter is not experienced in writing scientific articles, his or her help may make the text even worse. Research groups that wish to increase their international projection should seriously consider hiring specialized scientific writers, if possible native English speakers.

The second problem is the lack of empathy with the reader (or, in initial stages, with the editor and reviewers). When describing a Brazilian study in the international literature, the author must be sure that the text is intelligible to a reader from outside Brazil. This applies to specific issues, such as, for instance, explaining why minimum wages are used to measure family income, or detailing the location and characteristics of the geographical area where the study was conducted. This applies also to more general issues: what would be the interest of an international journal in publishing a Brazilian study? Answering this question requires a discussion of the external validity of the study, including a thorough characterization of the studied population, and speculation as to whether these results would or not be generalizable to other contexts. Brazilian studies often show advantages in relation to surveys conducted in the North, such as the possibility of studying diseases or exposures that are specific of poor populations.

The third issue is format. Every editor likes to receive articles that rigorously follow the journal's manuscript preparation norms. For instance, whether the abstract must be structured or unstructured, the information to be included in the front page, reference format, section titles, among others. Editors do not like to receive manuscripts that seem to have been previously submitted to and rejected by other journals, and then submitted to their journals without alterations in format. Of course, any editor of a journal that is not absolutely first tier will be aware that many of the manuscripts submitted to him or her have been previously refused, but it is essential to reformat the manuscript specifically to the journal to which it is being submitted.

Still regarding format, the size of the article is an important issue. In general, editors prefer to publish short articles, since this allows for the inclusion of a greater number of articles into a fixed number of pages. For cultural reasons that are beyond the scope of the analysis article, it seems that authors of Latin origin tend to be unduly prolix, at least from the Anglo-Saxon perspective. This leads, for instance, to extremely long introduction and discussion sections, and often to exceeding the word limit recommended by the journal. This may prove a fatal mistake, since the editor may have doubts about the appropriateness of sending such a long article for review, often opting for immediate refusal.

Most high-quality international journals currently refuse over 90% of manuscripts immediately, without external review. Editors reach this decision very fast, without reading the entire article. Criteria for this decision include the apparent quality of the article (partly based on writing and format, but also on methodology and length), its presumed interest to the journal's readership, and, without doubt, the articles authors; authors who are known, or affiliated to renowned institutions are undoubtedly more likely to pass this initial evaluation. In the aforementioned survey, a colleague spontaneously referred that "the problem is not being Brazilian, it's being unknown." It is unquestionable that editorial prejudice may important for the immediate selection of articles to be sent for review.

Editors usually start by reading the manuscript's cover letter. This letter is often underestimated by authors, but this is actually the best opportunity the author has to convince the editor that his or her article is appropriate for that journal. It also provides an opportunity to argue how a study carried out in Brazil may contribute to international literature. The editor then proceeds to reading the abstract and methods. After reading these three items, it is possible to decide whether the article should or not be sent to external reviewers.

In conclusion, it is important to consider the formal aspects of the article, and to show empathy with the target audience, if one whishes to publish in Northern journals. The personal experience of the present first author is that, after an author becomes experienced and well known, the percentage of refusals drops rapidly.

WHERE TO PUBLISH?

When commissioning the present article, the editors of the Revista de Saúde Pública mentioned that "maybe you could also discuss the issue of different readerships for different scientific productions, in other words, what type of study is aimed at the national public and what research could be of interest to international readers. It seems to us important to eliminate the false and simplistic notion that robust research should go to international publications whereas poor research is destined to the internal public." This is a complex, multidimensional issue, which has undergone profound change in recent years. When deciding on where to publish, authors must consider the following issues:

• Language and target audience. When submitting a manuscript, it is important to consider the target audience for the results at hand. For example, the evaluation of a Brazilian health care program must be published in Portuguese, given its importance for decision making in this country. Not with standing, a good evaluation is also of interest to the international public – the Mexican

  Programa de Educación, Salud y Alimentación (PROGRESA), for instance, which inspired the Brazilian

  Bolsa-Família, has been the object of high quality evaluations,

  ⁷ which, if restricted to the Mexican public, would have little or any access to readers of other countries. The same is true for the study on revaccination (REVAC), recently published by a Brazilian group.

  ⁸ A solution for this problem are journals that offer bilingual publication, such as is currently the case with the

  Revista de Saúde Pública.

• Type of contribution to knowledge. When a Brazilian study produces highly relevant results, its authors will normally attempt to publish them in a high-impact international journal in order to reach a wider public. On the other hand, confirmatory studies of associations that are previously known but which need to be replicated in our settings are more appropriate for national journals.

• Generalizability. Studies investigating predominantly biological outcomes, which can more easily be generalized to other contexts, are appropriate for international publication. On the other hand, studies of the prevalence of health-related problems within our population, evaluations of national services or programs, or studies of inequities in health are of more local interest. In contrast with those of basic science, the results of research in the field of public health tend to be highly dependent on the local context, and are therefore more appropriate for national journals.

• Citation indexes. Unfortunately, the citation indexes of Southern journals are still low when compared to the majority of international publications. Since this is an important evaluation criterion for our post-graduate programs, there is a strong incentive for publication in international journals. In the field of public health in Brazil, this problem is not as impeditive, since the two major Brazilian journals in this field are considered as international for evaluation purposes. However, in other fields of knowledge, this is an important incentive to publish abroad.

• Publication speed. There is widespread belief that the publication of articles in national journals is slow, frequently taking over a year, whereas in international journals this would be a more agile process. We are unaware of comparative data that confirm such a difference. Another point in favor of sending articles to first tier international journals is that, due to the large number of submissions, rejection is almost immediate for most articles, the author thus wasting less time when compared to journals in which the majority of manuscripts are forwarded to external review, and in which refusal may take months to be formalized. Journals such as Lancet include a "fast track" option, which reduces time between submission and publication to less than a month. However, this option is restricted only to extremely innovative articles. Brazilian journals could consider such a possibility as an incentive to high-quality articles.

• Open access. A great advantage of national journals is open access to published work, which is not the case for a large number of international journals. There is growing pressure to expand open access; for instance, the National Institutes of Health and the Wellcome Trust no longer allow articles generated by research financed by them to be published in restricted access journals.

In short, the decision regarding where to publish is complex, and depends on the factors mentioned above. With the dissemination of electronic and bilingual publications, this issue is evolving rapidly towards a reduction of the distance between national and international journals.

CONCLUSIONS

In the present commentary we have discussed the marked imbalance, in international scientific journals, between research originating from scientists from the North and the South. We have shown that there is a generalized perception among Brazilian researchers that such imbalance is due at least partly to prejudice of international editors against Southern authors, which is supported by the fact that a very small percentage of editorial board members of international journals come from the South. Although prejudice can explain part of the imbalance, there are also specific measures that may increase the likelihood of a paper from the South being accepted in international journals. These include the need to invest in the quality of the written text, and to show empathy with editors and readers, emphasizing the contribution of the manuscript to the international literature. Finally, we discuss whether research carried out in the South should be published in national or international journals, and suggest that there are at least six dimensions to this choice. These include language and target audience; type of contribution to knowledge; generalizability; citation index; speed of publication; and open access. The rapid growth in the number of Brazilian contributions to the international health literature shows that editorial prejudice, although often present, can be effectively offset by research with solid methodology and good-quality presentation.

REFERENCES

Received: 3/31/2006

The first author is Associate International Editor of the American Journal of Public Health and an Editorial Consultant of Lancet, as well as a member of the editorial boards of other national and international journals, including the Revista de Saúde Pública.

• 1. Almeida-Filho N, Kawachi I, Filho AP, Dachs JN. Research on health inequalities in Latin America and the Caribbean: bibliometric analysis (1971-2000) and descriptive content analysis (1971-1995). Am J Public Health 2003;93:2037-43.
• 2. Breilh J. Epidemiology's role in the creation of a humane world: convergences and divergences among the schools. Soc Sci Med 1995;41:911-4.
• 3. Gillman MW, Rifas-Shiman SL, Camargo CA, Jr., Berkey CS, Frazier AL, Rockett HR, et al. Risk of overweight among adolescents who were breastfed as infants. Jama 2001;285:2461-7.
• 4. Horton R. Medical journals: evidence of bias against the diseases of poverty. Lancet 2003;361:712-3.
• 5. Keiser J, Utzinger J, Tanner M, Singer BH. Representation of authors and editors from countries with different human development indexes in the leading literature on tropical medicine: survey of current evidence. BMJ 2004;328:1229-32.
• 6. Paraje G, Sadana R, Karam G. Public health. Increasing international gaps in health-related publications. Science 2005;308:959-60.
• 7. Rivera JA, Sotres-Alvarez D, Habicht JP,Shamah T, Villalpando S. Impact of the Mexican program for education, health, and nutrition (Progresa) on rates of growth and anemia in infants and young children: a randomized effectiveness study. Jama 2004;291:2563-70.
• 8. Rodrigues LC, Pereira SM, Cunha SS, Genser B, Ichihara MY, de Brito SC, et al. Effect of BCG revaccination on incidence of tuberculosis in school-aged children in Brazil: the BCG-REVAC cluster-randomised trial. Lancet 2005;366:1290-5.
• 9. Saxena S, Levav I, Maulik P, Saraceno B. How international are the editorial boards of leading psychiatry journals? Lancet 2003;361:609.
• 10. Sumathipala A, Siribaddana S, Patel V. Under-representation of developing countries in the research literature: ethical issues arising from a survey of five leading medical journals. BMC Med Ethics 2004;5:E5.
• 11. Tutarel O. How international are leading general paediatric journals? Arch Dis Child 2005;90:816-7.
• 12. Tyrer P. Combating editorial racism in psychiatric publications. Br J Psychiatry 2005;186:1-3.
• 13. Victora CG, Barros F, Lima RC, Horta BL, Wells J. Anthropometry and body composition of 18 year old men according to duration of breast feeding: birth cohort study from Brazil. BMJ 2003;327:901-4.
• 14. Victora CG, Habicht JP, Bryce J. Evidence-based public health: moving beyond randomized trials. Am J Public Health 2004;94:400-5.

Publication Dates

• Publication in this collection
  01 Dec 2006
• Date of issue
  Aug 2006

History

• Received
  31 Mar 2006