Research advances in physical activity and insomnia: a bibliometric and thematic evolution analysis

Background Insomnia affects approximately 16.2% of the global population, constituting a substantial public health burden associated with diverse adverse health outcomes. Although physical activity (PA) represents a modifiable health behavior with demonstrated potential for sleep improvement, a systematic mapping of research hotspots and evolutionary trajectories in this domain remains lacking. Methods We systematically retrieved English-language publications from the Web of Science Core Collection (WoSCC) from inception through May 20, 2026. A comprehensive search strategy combining subject headings with free-text keywords yielded 2,224 included records. We utilized the Bibliometrix package in R and VOSviewer software to visualize publication patterns across countries, institutions, journals, and authors; constructed thematic evolution maps; and identified research hotspots and emerging trends through strategic diagram analysis. We retrieved a supplementary Scopus dataset for cross-database validation. Web of Science served as the primary data source for analyses and visualizations, while Scopus verified trend consistency. Results The annual growth rate of publications reached 15.12%. The United States (n = 397) and China (n = 462) emerged as the predominant contributing nations. High-impact institutions included Harvard University and the University of California system. Among journals, Sleep Medicine exhibited the highest h-index. Thematic evolution revealed four distinct phases: 2010–2018 focused on metabolic associations and measurement instruments; 2019–2021 shifted toward pandemic-related stress and sleep-emotion comorbidities; 2022–2023 established a “monitoring-intervention-evaluation” closed-loop paradigm; and 2024–2025 entered a “genetic causality-digital intervention” era. Mendelian randomization(MR) and wearable devices have emerged as cutting-edge frontiers. Cross-database validation with Scopus corroborated a high degree of concordance in publication trends and core thematic foci. Conclusion The PA-insomnia research field has completed a paradigm shift from metabolic association validation to genetic causal precision, currently positioned at a critical juncture of multidisciplinary integration and digital transformation. Future research priorities might include, developing artificial intelligence-enabled wearable systems for real-time sleep-emotion-metabolism digital twin modeling, and advocating for the integration of multi-component “sleep hygiene plus exercise” interventions into clinical practice guidelines to address escalating sleep health challenges amid global population aging. Cross-database validation with Scopus corroborated a high degree of concordance in publication trends and core thematic foci.