Scientometric study of learning circles in medical education: Trends, collaborative networking, and citation dynamics

Learning circles in medical education are collaborative, peer-led study groups designed to enhance learning through discussion and shared experiences. These circles create a supportive environment where medical students and professionals can engage in active learning, critical thinking, and problem-solving. This scientometric study aims to analyze the evolution and trends in research on learning circles and medical education. A quantitative methodology was employed to evaluate scientific production and collaborations in this field.

This study included articles published between 2005 and 2024 that addressed topics related to "learning circle," "group lesson," "teaching cycle," "class coaching," "collaborative learning," "team-based learning," "interactive learning," and their application in "medical education" or "health professions education." The studies were indexed in the Web of Science database. Articles that did not meet these thematic criteria or were not available in the mentioned database were excluded. Duplicates and studies without complete data for bibliometric analysis were also excluded. The search was conducted in the Web of Science database on October 12, 2024, using the specialized search formula. The selection, extraction, and analysis processes were carried out using R Studio and the Bibliometrix package. Key bibliometric metrics were calculated, and CiteSpace was used to visualize trends and patterns.

The analysis covered the period from 2005 to 2024, with 50 documents from 35 sources. An annual growth rate of 8.84% was observed, with an average document age of 6.18 years and an average of 20.38 citations per document. A total of 182 authors were identified, with an average of 3.74 co-authors per document and 10% international co-authorship. Document types included 35 articles, two early access articles, one conference paper, two book reviews, one correction, two editorial materials, one letter, two meeting abstracts, and four reviews.

The analysis highlights the concentration of productivity in a few key journals, as revealed by Bradford's law, and the common trend of single-document authorship, as shown by Lotka's law. The dual map overlay graph indicates continuity in the field, with recent articles strongly connected to classic references, while also pointing to emerging areas of innovation.