Authorship Inequity in Global Health Research Conducted in Low- and Middle-Income Countries and Published in High-Income Country Family Medicine Journals

INTRODUCTION

Although there have been calls to increase the inclusion of researchers from low- and middle-income countries (LMICs) and to promote bidirectional partnerships in global health research, there has been slow progress toward author representation from these countries.^1-4 As stated by Juliane Chaccour in a Lancet Global Health editorial, authorship in collaborative research can be a proxy for research leadership.⁵ Authorship equity can therefore be a measure of strengthened research capacity in LMICs and a step toward global health equity.^6,7 Although publication disparities between LMICs and high-income countries (HICs) have been found in other specialties, authorship representation in family medicine journals has yet to be analyzed.^2,8,9

Bibliometric analyses have found that authors from LMICs are underrepresented in global collaboratives conducted in the author’s own country, especially in the more prominent first and last (senior) authorship positions.^10-12 In global emergency medicine, reproductive health, and oncology, nearly one-half (47% to 48%) of the senior authors are from HICs.^8,9,13 The disparity is particularly evident in journals that have high impact factors and that are published in HICs.^2,11,14 This pattern influences the research’s dissemination, noted by the higher relative citation ratio of articles with authors from countries having a high income.¹⁵ LMIC authorship is also less likely when the research has funding from HICs.^7,8 With the majority of the funding originating from institutions in countries having a high income, global health priorities and research agendas are often influenced by those institutions’ research leaders and eventual authors.¹⁶

The number of publications and prominent authorship positions can contribute to researchers’ promotions and the ability to secure future funding.^17,18 Subsequently, the disproportionate opportunities for researchers from HICs, including time allotted for research, funding sources based in these countries, editorial bias for western authors, and the academic pressures noted above, can exclude the advancement of researchers from LMICs.^12,19 Academic institutions, funding agencies, and journals should be held accountable for the promotion of equitable research partnerships to ensure that the LMIC researcher can be a main stakeholder in the research agenda and publication for the work completed in their country.^11,20,21 As Kaufman et al wrote in the conclusion of their bibliometric analysis of global health research, “Global health partnerships cannot be equitable without changing authorship trends between HIC and LMIC institutions.”¹⁰

To explore the current status in a field not yet studied, we investigated authorship inequities for research that was conducted in LMICs and published in family medicine journals based in HICs.

METHODS

RESULTS

Article Selection and Characteristics

A total of 14,707 articles were published in the HIC journals between 2018 and 2023, of which 1,030 articles mentioned low-income countries or LMICs, or Cochrane Review low-resourced or underresourced terms (Figure 1).²⁵ Twelve duplicate articles were removed. A total of 1,018 article abstracts from 17 journals were reviewed for inclusion and exclusion criteria. Of the 585 articles that underwent full-text review, 154 were removed, most often because they were editorials. One journal, American Family Physician, did not have any articles that met the final inclusion criteria.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

PRISMA Flow Diagram for Article Identification, Screening, and Inclusion

PRISMA = Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Note: Diagram is based on the PRISMA 2020 statement.²⁵

This process left 431 relevant articles published in 16 family medicine journals with editorial offices in HICs (Table 1). The United States (6 of 16 journals) and Australia (3 of 16 journals) were the top 2 locations for the journals’ editorial offices. The median impact factor of the journals in 2023 was 2.0, with a range from 1.1 to 5.3. More than one-half of the articles reported studies conducted in upper-middle-income countries (55.9%) with the large majority of first authors and last (senior) authors coming from LMICs (78% and 68%, respectively). Less than one-half of the articles (44.1%) were funded.

View this table:

• View inline
• View popup

Table 1.

Article Characteristics by Family Medicine Journal (N = 431 Articles)

Publication and Author Trends

Over time, there was an increase in publication of articles focused on research conducted in LMICs in family medicine journals (Figure 2, left and right panels). There were corresponding increases in representation of first authors from these countries from 2018 (59%) to 2023 (85%) (left panel), and also in representation of senior authors from these countries from 2018 (55%) to 2023 (74%) (right panel).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Trends in First and Senior Author Representation Over Time

HIC = high-income country; LMIC = low- or middle-income country (includes low-income, lower-middle-income, and upper-middle-income countries).

Article and Authorship Country Associations

Only 22 articles reported research conducted in low-income countries. Uganda was the most represented country in this category; the other countries, including Malawi and Ethiopia, are shown along with upper-middle-income countries in Figure 3, left and right, top maps. Articles in this category had the highest proportion of first and senior authors from HICs when only single country income levels were considered (Figure 4).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Heat Maps of Study Country Income Level With the Country Income Level of the Associated Articles’ First and Senior Authors

Studies conducted in low-income countries and lower-middle-income countries

Note: Powered by Bing (Microsoft Corp). Copyright Australian Bureau of Statistics, GeoNames (Unxos GmbH), Microsoft Corp, NavInfo (China Siwei Surveying & Mapping Technology Co, Ltd), Open Places, OpenStreetMap (OpenStreetMap Foundation), TomTom (TomTom NV), Zenrin (Zenrin Co Ltd).

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

First and Senior Author Country Income Level by Study Country Income Level

Research conducted in lower-middle-income countries accounted for about one-third of the articles, but less than one-quarter of the first and senior authors (Table 2). India, Iran, and Nigeria were the most represented in study location among lower-middle-income countries (Figure 3, left, top map). With respect to senior authorship, Australia, Iran, India, and the United Kingdom were the countries most represented (Figure 3, left, bottom map).

View this table:

• View inline
• View popup

Table 2.

Article Characteristics by Author Country Income Level

Most of the research and authors were from upper-middle-income countries (Table 2). China, Brazil, South Africa, and Turkey were the most represented in both study location and first authorship (Figure 3, right, top and middle maps). The United States, however, was in the fourth spot for most-represented senior author’s country affiliation after China, Brazil, and Turkey (Figure 3, right, bottom map).

There did not appear to be any association between research funding status and authorship (Table 2). But there was a trend toward higher average citation rate for articles having HIC first and senior authors compared with articles having LMIC first and senior authors.

DISCUSSION

Our bibliometric analysis of research conducted in LMICs published in family medicine journals provides an objective assessment of authorship equity within global family medicine research. HIC first and senior authors were represented in about one-quarter of the research conducted in LMICs overall and published in family medicine journals, and this representation was even higher for research done in low-income countries and lower-middle-income countries. Articles with HIC first and senior authors were cited more times on average than those with first and senior authors in LMIC countries. These findings are consistent with those of other studies that examined authorship equity among research published in other specialties and by other institutions and national research agencies.^8,15,19,26

Compared with studies of LMIC author representation in medicine generally and in specific specialties,^{8,11,12,14,27} our study found a higher rate of LMIC author representation in family medicine journals. Authorship equity and publication of research conducted in LMICs increased over time, trends that have been seen in other studies as well.^2,10 There are also shared trends of disparities, however, including a paucity of articles conducted in low-income countries, lower authorship representation by country income level, greater HIC authorship representation on multicountry studies, and an association of HIC authorship with higher rates of citation.^2,13,15 We did not find a trend between funding status and authorship representation in family medicine journals, but more than one-half of the studies were unfunded, a situation that differs from that for other specialties.^8,18

As other authors have highlighted, authorship is a complex topic that incorporates many factors. With respect to senior authorship, our study did not find as large of a change over time when compared to first authorship. As described by Kaufman et al¹⁰ in a bibliometric analysis of global health research published out of University of California San Francisco, the “institutional research enterprise” functions in such a way that senior authorship is often a reflection of the research agenda itself. As the authors suggested, institutions could consider incentivizing authors to relinquish or share research leadership, and therefore senior authorship. Furthermore, it is important to recognize and name the industrial complex of academic dissemination as a problematic system that continues to perpetuate authorship inequity and ultimately leadership inequity in research collaborations. Journals have become intertwined with the institutional expectations for faculty promotion criteria and funders’ expectations for open access publications, which also benefits the journals themselves.

Some suggested solutions include joint authorship, which would encourage HIC authors to ensure that their partners are collaborating on the dissemination process and that teams are creating early dissemination plans.^28,29 More importantly, though, HIC authors need incentives to relinquish control of research collaborations and to ensure that the research question, study design, and dissemination process are led by the local partner. Many HIC authors aim to collaborate equitably; however, their institutional promotion requirements prioritize first and senior authorship positions as well as lead roles on grants awarded. One solution, therefore, is for HIC authors to work with their institutions to change promotion criteria to reward individuals who transition away from their traditional leadership roles and who recognize the value of mentorship and facilitation roles.¹⁷ Finally, equitable research collaboration will also require HIC authors dedicating time to ensuring research mentorship is available for local partners. This will require culturally humble mentorship, however, to create a space that (1) promotes local partners speaking up when research structures do not fit the local context and (2) encourages HIC authors to recognize when traditional research structures may be promoting harm and when they need to step back and listen.^30,31 Other recommendations for HIC researchers to promote allyship can be found in a recent publication by Pai et al¹⁶ on shifting power in global health.

Our study has several major limitations. Most importantly, we recognize the irony that most of the authors on this article are based at HIC institutions, and these authors have been first or senior authors on collaborative global health research. This project was conceptualized at a global health conference located in an HIC, demonstrating how the industrial complex of academic dissemination continues to perpetuate inequity in global health research collaboration. Furthermore, we have had additional training in scholarly work, have access to a strong institutional library with individuals who can support bibliometric analyses, and ultimately have more time than our LMIC collaborators. As a result, we are defining inequity from an HIC lens. Second, the articles included in this bibliometric analysis were all written in English given that it was the shared spoken language of our team. Ten out of the 27 WONCA journals with editorial offices in HICs were in non-English languages (eg, French, Portuguese, German) and therefore not analyzed here. The WONCA website was last updated in January 2022, so some new journals may have been excluded (e-mail communication, Diarmuid Hayes, WONCA Senior Communications Officer; Nov 12, 2024).³¹ We also excluded articles about refugees from LMICs living in HICs, which could have excluded authors from those countries as well. Although this study focused on authorship equity, we did not account for or investigate equity within the country and regions mentioned. After initiating a bibliometric analysis for all family medicine journals, regardless of editorial office location, we recognized the need for a more nuanced discussion to account for other regional authorship inequities (eg, other identity considerations such as tribe, gender, race); therefore, we focused here only on HIC journals to clarify the current state of equity in authorship within the academic industrial complex with its disproportionate opportunities and pressures specific to HICs. Future studies should include additional investigators from LMICs to further analyze articles published in journals from these countries with consideration to regional inequities, as well as expand on inclusion of other languages.