Abstract

Purpose: A systematic and detailed examination of studies on diabetes can guide the evaluation of healthcare services, the identification of issues, and the implementation of preventive interventions. Therefore, this study aims to assess diabetes research in the primary health care (PHC) field using machine learning-based bibliometric methods.
Materials and methods: In this study, articles related to diabetes in the PHC field were obtained from the Web of Science Core Collection on March 15, 2024. These articles were analyzed using bibliometric methods and the Latent Dirichlet Allocation (LDA) topic modeling technique.
Results: The analysis of the studies revealed that 3.355 articles on diabetes in the PHC field were produced by researchers from 114 different countries, 4.226 institutions, and 12.426 individual researchers. Recent years have shown a growing interest in topics such as obesity, hypertension, chronic diseases, exercise, and physical activity within the PHC field. Topic modeling identified eight distinct latent topic clusters: (1) Diabetes management in PHC, (2) Risk factors and management of diabetes in PHC, (3) Acute and chronic complications of diabetes in PHC, (4) Gestational diabetes, (5) Prediabetes and Type 1 diabetes, (6) COVID-19 and diabetes, (7) Quality of life, awareness, and health education, (8) Current treatment methods and guideline resources.
Conclusion: Primary Care Diabetes and Journal of Family Medicine and Primary Care are the leading journals in PHC-related diabetes research. The studies show a significant overlap between diabetes research and topics such as hypertension and obesity. Future studies in the PHC field are recommended to focus on diabetic retinopathy and diabetic wound research.

Keywords:  Primary health care, diabetes, data analysis, science mapping, topic modeling