Association between body composition and calcaneal ultrasound t-score in middle-aged and older adults from four ethnic minorities in Guangxi, China

This study aimed to investigate the association between body composition and calcaneal ultrasound T-score (a surrogate indicator of bone mineral density, BMD) in middle-aged and older adults from four ethnic minorities (Maonan, Mulao, Miao, Yao) in Guangxi, China, and to explore ethnic differences in body composition and bone status to provide evidence for targeted osteoporosis prevention strategies. A cross-sectional study was conducted with 1652 participants aged ≥ 45 years recruited via random cluster sampling from ethnic autonomous counties in Guangxi. Body composition indices (e.g., fat-free mass, muscle mass, fat mass) were measured using a bioelectrical impedance analysis (BIA) analyzer, and right calcaneal T-score was measured by quantitative ultrasound (QUS); note that QUS-derived T-score is not equivalent to DXA-measured BMD. Descriptive statistics, Pearson correlation analysis, one-way analysis of variance (ANOVA), multivariable linear regression, and propensity score matching (PSM) were used for statistical analysis, with a two-tailed p < 0.05 considered statistically significant. The study population was predominantly female (56.05%), with 31.48% aged 56–65 years and 87.71% having an educational level of junior high school or below. Ethnic stratification revealed significant heterogeneity in anthropometric and body composition indices: the Mulao group had the highest average height and fat-free mass, the Miao group had the highest BMI and fat mass, and the Yao group had the highest calcaneal T-score while the Maonan group had the lowest. Pearson correlation analysis showed that calcaneal T-score was positively and significantly correlated with most body composition indices (all p < 0.01), with lower limb muscle mass exhibiting the strongest correlation (r = 0.223); no significant correlations were found between T-score and visceral fat mass, waist-to-hip ratio, trunk fat mass, or upper limb fat mass (p > 0.05). Multivariable regression confirmed that lower limb muscle mass remained an independent correlate of higher T-scores after adjusting for confounders (β = 0.152, p < 0.001). One-way ANOVA indicated significant ethnic differences in 19 of the 20 measured indices (all p < 0.05), with only skeletal muscle index showing no inter-ethnic difference (p > 0.05). PSM analysis demonstrated acceptable overall matching quality, with most covariates achieving effective balance after matching, while fat mass and weight remained inadequately balanced (p = 0.077 and p = 0.047, respectively). Body weight and lean mass, particularly lower limb muscle mass, are positively associated with higher calcaneal T-scores in middle-aged and older adults of ethnic minorities in Guangxi, with stronger associations observed in women. Significant ethnic differences exist in body composition and bone status among the Maonan, Mulao, Miao, and Yao groups. Maintaining adequate body weight and preserving lower limb muscle mass through targeted exercise may be beneficial for bone health in this population, and ethnic- and gender-specific screening and intervention programs should be developed based on the observed disparities.