Body mass index and physical fitness in university students: sex differences and nonlinear patterns in a large cross-sectional study
Qiang Wei, Lei Wang, Chun Yang, Wei Sun, Siqin ShenBackground
Body mass index (BMI) is routinely used in university health surveillance, but its sex- and grade-specific association with multidimensional physical fitness, including nonlinear patterns, remains unclear.
Methods
Data from 16,292 university students (8,505 females; 7,787 males; aged 17–23 years) who completed the national standardized fitness assessment were analyzed in this cross-sectional study. Vital capacity, 50-m sprint, standing long jump, sit-and-reach, 800/1,000-m run, and muscular endurance were standardized within sex to derive a Physical Fitness Index (PFI), with time-based tests reverse-scored. Two-way analyses of variance examined associations of sex, academic grade, and BMI category with PFI. Sex- and grade-specific quadratic models were used to characterize BMI–PFI curves and estimate the model-derived BMI value associated with peak predicted PFI. Pearson correlations assessed associations between PFI and individual fitness components.
Results
Academic grade was associated with PFI, whereas no overall sex effect was observed (sex: F (1, 16,284) = 1.51, p = 0.219, partial η 2 = 0.000093; grade: F (3, 16,284) = 74.54, p < 0.001, partial η 2 = 0.0136). The sex × grade interaction was statistically significant but small ( F (3, 16,284) = 20.44, p < 0.001, partial η 2 = 0.00375). Within both sexes, normal-weight students showed the highest PFI; underweight students had higher PFI than overweight and obese peers, while PFI was lower in the overweight and obese groups. The BMI–PFI association was stronger in males (partial η 2 = 0.1061) than in females (partial η 2 = 0.0287). All quadratic models showed a concave BMI–PFI pattern, with lower predicted PFI at higher BMI values and model-derived peak estimates varying across sex–grade subgroups. These estimates should not be interpreted as physiological optima or recommended BMI targets. Correlations indicated that PFI integrated cardiorespiratory fitness, lower-limb power, flexibility, muscular endurance, and running performance.
Conclusion
BMI was associated with multidimensional physical fitness in this large cohort of university students, with lower PFI observed at both low and high BMI values. BMI showed a stronger association with PFI than academic grade, particularly in males. These findings support BMI-informed and sex-sensitive fitness monitoring, but should be interpreted as cross-sectional associations because body-composition and behavioral measures were unavailable.