Regional disparities and scientific collaboration in Brazilian surgical research: a national bibliometric studyABSTRACT
Introduction: To analyse Brazilian scientific production in surgery from a bibliometric perspective, with a particular focus on regional distribution, patterns of scientific collaboration, and the impact of publications in both national and international arenas.
Methods: We conducted a retrospective, national, ecological observational study using bibliometric data from the SciVal platform (Elsevier), derived from the Scopus database. The study included full scientific articles with at least one Brazilian affiliation published between 2014 and 2023. We evaluated key bibliometric indicators, including publication volume, national and international collaboration, citation counts, and Field-Weighted Citation Impact (FWCI). Analyses were stratified by geographical region, with indicators normalised by resident population and the total number of medical doctors.
Results: We identified 14,160 scientific articles in the surgical field, accounting for about 8.5% of Brazilian medical research output. Publication volume grew steadily throughout the decade, followed by a recent stabilisation. Output was heavily concentrated in the Southeast and South regions, which also recorded the highest scientific impact (FWCI up to 0.82) and the greatest integration into international collaboration networks (23%). In contrast, the North region exhibited the lowest output and impact (FWCI = 0.48). International collaboration was consistently associated with superior bibliometric performance. The most frequent research topics included spinal surgery, orthopaedics, hernias, and obesity, with the latter associated with the highest scientific impact (FWCI = 1.80).
Conclusion: Surgical research output in Brazil has shown sustained growth, though it is marked by pronounced regional disparities and a strong dependence on established academic centres and international collaborative networks. Expanding scientific capacity in underrepresented regions and strengthening collaborative strategies with robust national leadership are key priorities for the balanced and sustainable development of surgical research in the country.
Keywords:
Bibliometrics; Surgical Research; Scientific Production Indicators; International Collaboration; Health Disparities
RESUMO
Introdução: Analisar a produção científica brasileira em cirurgia sob a perspectiva bibliométrica, com ênfase na distribuição regional, nos padrões de colaboração científica e no impacto das publicações nos cenários nacional e internacional.
Métodos: Realizado estudo observacional retrospectivo, baseado em dados da plataforma SciVal (Elsevier), provenientes da base de dados Scopus. Foram incluídos artigos científicos completos com pelo menos uma afiliação brasileira, publicados entre 2014 e 2023. Avaliaram-se indicadores bibliométricos, incluindo volume de publicações, colaboração nacional e internacional, citações e Field-Weighted Citation Impact (FWCI), com análise estratificada por região e com indicadores normalizados pela população e pelo número de médicos.
Resultados: Foram identificados 14.160 artigos científicos na área cirúrgica, o que corresponde a 8,5% da produção médica brasileira. Observou-se crescimento progressivo ao longo da década, seguido de estabilização recente. A produção apresentou forte concentração nas regiões Sudeste e Sul, que também registraram maior impacto científico (FWCI de até 0,82) e maior inserção em redes de colaboração internacional (≈23%). Em contraste, a região Norte apresentou menor produção e menor impacto (FWCI = 0,48). A colaboração internacional esteve consistentemente associada a um desempenho bibliométrico superior. Os principais temas incluíram cirurgia da coluna, ortopedia, hérnias e obesidade, sendo esta última associada ao maior impacto (FWCI = 1,80).
Conclusão: A produção científica em cirurgia no Brasil apresenta crescimento sustentado, porém com marcadas desigualdades regionais e forte dependência de centros acadêmicos consolidados e de redes internacionais de colaboração. A ampliação da capacidade científica em regiões sub-representadas e o fortalecimento de estratégias de colaboração com protagonismo nacional emergem como prioridades para o desenvolvimento equilibrado e sustentável da pesquisa cirúrgica no país.
Palavras-chave:
Bibliometrics; Surgical Research; Scientific Production Indicators; International Collaboration; Health Disparities
INTRODUCTION
Scientific research in higher education institutions is a cornerstone of clinical practice improvement and professional medical training. Knowledge generated by integrating teaching, clinical care, and research guides public health decision-making, informs more effective policy design, and promotes more rational use of healthcare resources. Scientific output must be viewed not merely as an academic pursuit, but as a critical instrument to elevate clinical care and tackle health inequalities1.
Within medicine, surgical specialities occupy a central role in scientific production, combining highly complex clinical activities, technological innovation, and intensive human resource training. Surgical research helps refine operative techniques, incorporate new devices, and organise perioperative care2,3. This technical-scientific profile makes surgical research output a particularly sensitive metric for evaluating institutional capacity to generate impactful, innovative knowledge that directly informs clinical practice.
Nonetheless, surgical scientific output exhibits substantial geographic heterogeneity, particularly in countries characterised by profound regional disparities4. In Brazil, this phenomenon is particularly striking, with academic activity heavily clustered in a small number of states and institutions. This raises pertinent questions regarding the organisation of research, human resource training, and access to research environments structured for surgical investigation5. Despite the relevance of this issue, systematic analyses integrating territorial distribution, collaboration patterns, and the impact of publications within Brazilian surgical research remain scarce.
Against this background, this study aimed to analyse Brazilian scientific production in surgery using bibliometric indicators to describe its territorial distribution across federal states, national and international collaboration patterns, and the scientific impact of publications. We also evaluated the leadership role of Brazilian institutions as primary authors in studies conducted through international collaboration. Synthesising these dimensions provides a more precise understanding of the organisation of surgical research in Brazil and its regional asymmetries.
METHODS
This is a national, retrospective, ecological, observational study analysing Brazilian scientific production in surgery from 2014 to 2023. We retrieved bibliometric data using the SciVal platform (Elsevier), based on records indexed in the Scopus database6.
Full scientific articles published in Scopus-indexed journals and identified via SciVal between 1 January 2014 and 31 December 2023 were included. We identified records exclusively through the ‘Surgery’ thematic category (ASJC 2746) using SciVal’s structured filters. We extracted data using SciVal’s analytical modules on 10 March 2025, applying predefined filters for the period (2014-2023), document type (full scientific articles), and institutional affiliation, including publications with at least one Brazilian affiliation9.
The SciVal platform does not employ search strategies based on free-text query strings, as the data are pre-structured and indexed in Scopus. Thus, records were identified using combined filters for subject area, period, and institutional affiliation. The platform automatically excluded documents categorised as editorials, letters, notes, conference proceedings, book chapters, and other non-journal article types. We also discarded duplicate records and publications with incomplete affiliation details.
The following bibliometric variables were analysed: total number of publications, publication year, publishing journals, access type, number and names of authors, institutional affiliations, and authors’ countries of origin. We also assessed international collaboration, the number of institutions and countries involved per publication, and scientific impact metrics, including citation counts and Field-Weighted Citation Impact (FWCI). Journal characteristics, such as impact metrics and thematic classifications, were also recorded.
Co-authorship network analysis was also conducted using authorship data extracted from SciVal11. The network graph was constructed with authors represented as nodes and co-authorship relationships as edges, and subsequently analysed and visualised using Gephi software (version 0.10.1). First, the largest connected component was identified to isolate the primary core of the collaboration network and remove disconnected or low-representation components. We then filtered to retain only edges with a weight greater than five (corresponding to more than five shared publications between two authors) and nodes with a degree greater than five (corresponding to authors with more than five distinct collaborations). We evaluated different filtering thresholds during the exploratory phase; we selected these parameters because they preserved the overall network structure while highlighting the most relevant authors and collaborative relationships. Community structures were identified using the Louvain algorithm implemented in Gephi. In the graphical representation, node size was proportional to the author’s degree (number of collaborations), while edge thickness was proportional to connection weight (number of publications shared between two authors).
Demographic and healthcare contextual data were obtained from official national databases, including the Brazilian Institute of Geography and Statistics (IBGE)7 and the Federal Council of Medicine (CFM)8,9. Bibliometric data were initially organised and analysed in an aggregate manner by federal state, geographical region, and publication year. These production indicators were then linked to demographic, healthcare, and academic contextual data, allowing surgical scientific output to be examined in light of resident population, medical doctor distribution, availability of postgraduate research (stricto sensu) programmes, and economic and human development indicators.
To enable comparisons between states and regions with different structural characteristics, we evaluated results using both absolute numbers and standardised indicators, such as rates per population and per registered medical doctor. This approach permitted the exploration of the spatial and temporal distribution of surgical scientific production in Brazil, supported by graphical and cartographic representations, including maps, time series, and exploratory charts.
This study relied exclusively on publicly available secondary data from bibliographical databases and official national sources, analysed in an aggregated format without individual identification. The study did not access sensitive personal data and involved no direct human participation. Therefore, the study was exempt from review by a Research Ethics Committee, in accordance with current regulations for this type of research.
RESULTS
Between 2014 and 2023, Scopus identified 23,636,942 full scientific articles published in journals across all fields of knowledge globally. Medicine accounted for 6,158,248 (26.05%) of these articles. Brazilian medicine contributed 166,668 articles, representing 2.70% of global medical production. Across broad medical areas, surgery produced 14,160 journal articles, accounting for about 8.5% of Brazilian medical scientific output during the evaluated period. Figure 1 shows comparative national scientific production by region over the 10-year period. National scientific production stabilised in recent years, with a clear predominance of the Southeast and South regions in absolute terms.
When evaluating the main surgical subfields in national production during this period, the most frequent topics were treatment strategies in spinal health, followed by orthodontic implants and bone health, and surgical innovations in hernia repair. Table1 shows that publications addressing Obesity, Metabolic Health, and Quality of Life achieved the highest scientific impact, as evidenced by the highest FWCI.
When analysing the ten Brazilian institutions with the highest absolute number of publications, a heavy concentration of universities in the Southeast region is apparent (Table 2). Most of these publications originate from multicentre studies, including one on the global burden of diseases, injuries, and risk factors10.
Among journals with the most publications in the field during this period, national journals predominate (Table 3).
When scientific output is evaluated using criteria proportional to the population and the number of general surgeons, regional differences decrease, although the South and Southeast regions still predominate (Table 4). To evaluate regional bibliometric performance, we used normalised indicators, with data presented in Table. Regardless of the metric used, the North region shows lower academic output in both absolute and proportional terms. Furthermore, the bibliometric impact of this production is notably lower, as shown by its low FWCI score.
Comparing scientific collaboration between regions, we found that some geographically adjacent Brazilian regions, such as the North and Central-West, show little collaboration. Similarly, we observed no unidirectional collaboration between the Northeast and North regions. The Southeast region collaborates with all regions of the country, as does the South, albeit to a lesser extent. The Southeast and South regions produced the most articles resulting from international scientific collaboration, as shown in Table 5. As described above, articles from these regions have a higher FWCI and a larger percentage of highly cited publications.
After identifying the largest connected component and applying the filtering criteria (edges >5, nodes > 5), the final co-authorship network comprised 527 authors connected by 11,356 relationships. The Louvain algorithm identified collaborative communities. For graphical representation, we highlighted the eight largest communities, representing approximately 84% of the network’s authors (Figure 2). The main collaborative communities showed a striking predominance of the Southeast region.
Scientific co-authorship network in Brazilian surgical production (2014-2023). Nodes represent authors, and edges represent co-authorship relationships. Node size is proportional to each author’s degree (number of collaborations), while edge thickness is proportional to connection weight (number of publications shared between two authors). Colours represent the collaborative communities identified by the Louvain algorithm. The figure displays the eight largest communities, which account for approximately 84% of the authors in the network.
The largest community (red) showed strong geographic concentration, with 94% of authors affiliated exclusively with the Southeast region. The second-largest community (blue) showed greater regional integration, with 67% of authors associated with both the Southeast and South regions, and authors bridging these groups.
The third-largest community (green) also showed a predominance of the Southeast region (80%), with additional participation of authors linked to the South region (16%), indicating partially integrated collaboration hubs. The fourth community (purple) was characterised by a strong association between the Southeast and South regions (82%).
In contrast, the smaller communities showed greater geographic diversity. The fifth community (orange) included 61% of authors working across multiple regions, while the sixth (pink) and eighth (black) displayed a broader distribution across different regions. The seventh community (brown) had an intermediate profile, combining authors with a national reach and a partial concentration in the Southeast.
Overall, the largest communities in the network exhibited lower geographic diversity, whereas smaller communities showed greater integration across different Brazilian regions.
The United States had the highest number of international collaborations across all Brazilian regions. Western European countries also exhibited a collaboration profile proportional to that observed in Brazil. We also highlight African countries’ participation in collaborations with the South and Southeast regions, primarily through large multicentre studies and international consensus statements, such as the second international consensus guidelines for advanced breast cancer published in 2014 and updated in 201711,12. These publications involved multiple countries, reflecting broad international collaborative networks rather than specific bilateral partnerships between African nations and Brazilian regions.
DISCUSSION
This study revealed a continuous growth pattern in Brazilian surgical scientific production between 2014 and 2023, with relative stabilisation towards the end of the period, accompanied by marked geographic and institutional concentration. The findings show that surgical scientific output in Brazil is unevenly distributed and reflects the structural organisation of the national science and technology system, dominated by a few academic and clinical hubs. These results suggest that the capacity to generate knowledge in surgery is strongly associated with the historical concentration of funding, infrastructure, and medical training programmes in specific regions of the country.
A comparative analysis of Brazilian regions shows clear dominance by the Southeast and South in both absolute and relative terms, while the North underperforms significantly in both volume and scientific impact. Previous studies reinforce the hypothesis that scientific production follows the unequal distribution of universities, highly complex tertiary hospitals, and postgraduate research programmes, creating a cumulative process in which better-equipped regions concentrate greater production capacity and funding acquisition. This phenomenon, widely documented in international literature, can be understood as a structural concentration effect in science, wherein inequalities tend to perpetuate over time13.
Normalising the results using standardised indicators proportional to population and specialist density enabled a clearer analysis of these inequalities. While these adjustments partially reduced the regional gaps, asymmetries persisted, indicating that structural factors such as specialist density, the academic environment, and institutional capacity play a decisive role in scientific production. Consequently, these findings suggest that expanding surgical scientific output in underrepresented regions depends not only on demographic growth or an increase in the number of professionals, but also on consistent regional investment policies targeting research infrastructure and academic training14.
Regarding scientific impact, regions more integrated into collaborative networks, particularly international ones, achieved higher Field-Weighted Citation Impact (FWCI) scores, whereas regions with limited participation, such as the North, showed lower relative impact. These results suggest that participation in international collaborative networks is associated with higher bibliometric impact, possibly by enhancing scientific visibility, knowledge circulation, and publication reach. This interpretation aligns with evidence that impact metrics, such as citation counts, are influenced by network effects and the collective nature of scientific production15,16.
Analysis of scientific collaboration patterns reinforces this interpretation. Nationally, the Southeast and South regions serve as the primary hubs of articulation, connecting with all other regions, whereas the North and Central-West show minimal integration with each other. This pattern suggests a national research network that remains asymmetric and hierarchically structured, indicating reliance on established centres for scientific coordination.
Co-authorship network analysis deepens this interpretation by revealing an equally unequal structural organisation. The largest collaborative communities are predominantly concentrated in the Southeast, while smaller groups exhibit greater geographic diversity. This pattern suggests the coexistence of a highly productive core and more diverse, though less structured, groups, highlighted by highly collaborative authors who bridge researchers from different regions27.
Classic studies argue that contemporary science has become increasingly dependent on interdisciplinary teams and inter-institutional collaboration, which tends to favour centres with greater capacity and structural organisation. Consequently, regions with lower resource density face greater challenges in establishing independent collaborative networks, reinforcing asymmetric patterns of scientific integration.
In the international arena, there is a strong predominance of partnerships with high-income countries, particularly the United States and Western European nations, indicating that these regions exert considerable influence over Brazilian scientific production. While international collaboration is associated with greater scientific impact, this participation is not homogeneous. In many medical fields, these collaborations occur within large international multicentre studies, where middle-income countries’ involvement may reflect integration into pre-existing networks rather than scientific leadership or the establishment of structured, sustainable partnerships. The results indicate opportunities to expand international collaboration strategies that can strengthen Brazilian scientific leadership, especially through long-term partnerships that foster greater institutional and scientific prominence. This perspective aligns with studies highlighting the importance of structured international collaborations in expanding the visibility, sustainability, and impact of scientific production18.
Regarding research topics, the study observed a predominance of fields such as spinal surgery, orthopaedics, hernias, and obesity. This profile suggests some alignment between Brazilian scientific production and the country’s most prevalent clinical needs. The predominance of these areas suggests that Brazilian scientific output is partially tailored to common clinical demands. However, it remains crucial to balance engagement with global research agendas with attention to national healthcare priorities to foster local relevance and international scientific visibility. This perspective is consistent with studies emphasising the importance of balancing global collaboration with the capacity to respond to the specific needs of national healthcare systems19.
This study has limitations that readers should consider when interpreting the results. The analysis relied on Scopus data, which may underrepresent scientific output published in non-indexed journals. Additionally, the thematic classification and bibliometric indicators used depend on the database’s methodology, which may introduce indexing biases. Nevertheless, using a database with broad international coverage and standardised indicators ensures the robustness and comparability of the findings.
In summary, scientific production in surgery in Brazil remains structurally concentrated, with persistent regional inequalities and a heavy reliance on consolidated academic centres and national and international collaborative networks. These findings provide valuable evidence for planning strategies to reduce regional disparities and strengthen national research capacity.
CONCLUSIONS
Surgical scientific production in Brazil has grown over the last decade, yet remains marked by strong regional concentration and persistent inequalities in scientific impact. Participation in collaborative networks, particularly international ones, is associated with greater visibility and superior bibliometric performance, highlighting the importance of integration into collaborative networks for national scientific output.
The findings offer valuable evidence to support strategies to reduce regional disparities, strengthen national scientific capacity, and foster more balanced collaborations between institutions and regions, thereby contributing to an increasingly high-calibre international presence for Brazilian surgical research.
ACKNOWLEDGEMENTS
The authors thank the participating universities and research centres for their institutional support, as well as the technical support provided to access and analyse the bibliometric data used in this study.
REFERENCES
-
1 Ramsey PG, Miller ED. A single mission for academic medicine: improving health. JAMA. 2009;301(14):1475-6. doi: 10.1001/jama.2009.472.
» https://doi.org/10.1001/jama.2009.472 -
2 McCulloch P, Altman DG, Campbell WB, Flum DR, Glasziou P, Marshall JC, et al. No surgical innovation without evaluation: the IDEAL recommendations. Lancet. 2009;374(9695):1105-12. doi:10.1016/S0140-6736(09)61116-8.
» https://doi.org/10.1016/S0140-6736(09)61116-8 -
3 Hirst A, Philippou Y, Blazeby J, Campbell B, Campbell M, Feinberg J, et al. No surgical innovation without evaluation: evolution and further development of the IDEAL framework and recommendations. Ann Surg. 2019;269(2):211-20. doi:10.1097/SLA.0000000000002794.
» https://doi.org/10.1097/SLA.0000000000002794 -
4 Van Rossum M, Bosker BH, Pierik EGJM, Verheyen CCPM. Geographic origin of publications in surgical journals. Br J Surg. 2007;94(2):244-7. doi:10.1002/bjs.5571.
» https://doi.org/10.1002/bjs.5571 -
5 Chiarini T, Oliveira VP, do Couto e Silva Neto FC. Spatial distribution of scientific activities: an exploratory analysis of Brazil, 2000-10. Sci Public Policy. 2014;41(5):625-40. doi:10.1093/scipol/sct093.
» https://doi.org/10.1093/scipol/sct093 -
6 Elsevier. Scopus: content coverage guide [Internet]. Amsterdam: Elsevier; 2024 [cited 2026 Apr 25]. Available from: https://www.elsevier.com/solutions/scopus/how-scopus-works/content
» https://www.elsevier.com/solutions/scopus/how-scopus-works/content -
7 Instituto Brasileiro de Geografia e Estatística (IBGE). Estimativas da população residente para os municípios e unidades da federação [Internet]. Rio de Janeiro: IBGE; 2023 [cited 2026 Apr 25]. Available from: https://www.ibge.gov.br/estatisticas/sociais/populacao.html
» https://www.ibge.gov.br/estatisticas/sociais/populacao.html -
8 Scheffer M, Cassenote A, Guerra A, et al. Brazilian Medical Demography 2023 [Internet]. São Paulo: Brazilian Federal Council of Medicine; 2023 [cited 2026 Apr 25]. Available from: https://demografia.cfm.org.br/
» https://demografia.cfm.org.br - 9 Scheffer M, Almeida C de J, Cassenote AJF, Dias IWH, Moreira JP de L, Sousa J, et al. Demografia Médica no Brasil 2025. Caetano M, editor. Brasília (DF): Ministério da Saúde; 2025.
-
10 Vos T, Lim SS, Abbafati C, Abbas KM, Abbasi M, Abbasifard M, et al. Global burden of 369 diseases and injuries in 204 countries and territories, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2020;396(10258):1204-22. doi:10.1016/S0140-6736(20)30925-9.
» https://doi.org/10.1016/S0140-6736(20)30925-9. -
11 Cardoso F, Costa A, Norton L, Senkus E, Aapro M, André F, et al. ESO-ESMO 2nd international consensus guidelines for advanced breast cancer (ABC2). The Breast. 2014;23(5):489-502. doi:10.1016/j.breast.2014.08.009.
» https://doi.org/10.1016/j.breast.2014.08.009 -
12 Cardoso F, Costa A, Senkus E, Aapro M, André F, Barrios CH, et al. 3rd ESO-ESMO international consensus guidelines for advanced breast cancer (ABC 3). The Breast. 2017;31:244-59. doi:10.1016/j.breast.2016.10.001.
» https://doi.org/10.1016/j.breast.2016.10.001 -
13 Sidone OJG, Haddad EA, Mena-Chalco JP. A ciência nas regiões brasileiras: evolução da produção e das redes de colaboração científica. Transinformação. 2016;28(1):15-32. doi:10.1590/2318-08892016002800002.
» https://doi.org/10.1590/2318-08892016002800002 -
14 Scheffer MC, Cassenote AJF, Guilloux AGA, Dal Poz MR. Internal migration of physicians who graduated in Brazil between 1980 and 2014. Hum Resour Health. 2018;16(1):21. doi:10.1186/s12960-018-0286-8.
» https://doi.org/10.1186/s12960-018-0286-8 -
15 Li W, Zhang S, Zheng Z, Cranmer SJ, Clauset A. Untangling the network effects of productivity and prominence among scientists. Nat Commun. 2022;13(1):4907. doi:10.1038/s41467-022-32604-6.
» https://doi.org/10.1038/s41467-022-32604-6 -
16 Wagner CS, Whetsell TA, Leydesdorff L. Growth of international collaboration in science: revisiting six specialties. Scientometrics. 2017;110(3):1633-52. doi:10.1007/s11192-016-2230-9.
» https://doi.org/10.1007/s11192-016-2230-9 -
17 Wuchty S, Jones BF, Uzzi B. The increasing dominance of teams in production of knowledge. Science. 2007;316(5827):1036-9. doi:10.1126/science.1136099.
» https://doi.org/10.1126/science.1136099 -
18 McManus C, Baeta Neves AA, Maranhão AQ, Souza Filho AG, Santana JM. International collaboration in Brazilian science: financing and impact. Scientometrics. 2020;125(3):2745-72. doi:10.1007/s11192-020-03728-7
» https://doi.org/10.1007/s11192-020-03728-7 -
19 Guimarães R, Santos LMP, Angulo-Tuesta A, Serruya SJ. Defining and implementing a national policy for science, technology, and innovation in health: lessons from the Brazilian experience. Cad Saúde Pública. 2006;22(9):1775-85. doi:10.1590/S0102-311X2006000900002.
» https://doi.org/10.1590/S0102-311X2006000900002
