Multilevel analysis of morbidity in athletes and reduction of health risks

Objective. To study the relationships between diseases in elite athletes of various disciplines for development of effective strategies for prevention and health management. Material and methods. The study included medical examination data from 15.464 athletes of Russian national teams for the period 2021—2023. We analyzed data on gender, age, body mass index, blood biochemistry, sport and diseases classified according to ICD-10. Statistical analysis was carried out using the R programming language and appropriate software packages. Network metrics such as the Phi coefficient (φ), average vertex degree, clustering coefficient, average path length, and network density were used to assess disease relationships. Correlations were assessed using contingent tables and adjustments for multiple comparisons using the Benjamini-Hochberg method. Results. There are significant gender differences in disease associations among athletes of various sports. In women, rhythmic gymnastics and wrestling show more structured disease networks with high average vertex degree and clustering coefficient. This indicates high probability of disease co-occurrence. In men, high values of these metrics are observed in fencing and sailing. Analysis of average path length and network density showed more dispersed disease networks among women in diving and figure skating. This indicates lower probability of co-occurrence. In men, similar dispersal is observed in hand-to-hand combat and handball. Conclusion. Our results highlight the role of gender-oriented approach in preventive and therapeutic measures for athletes. Differences in disease connectivity require specific prevention and treatment strategies taking into account gender differences and type of sport. This will improve health of athletes and reduce the risks of diseases associated with professional activities.