Reciprocal associations between International Physical Activity Questionnaire variables and sarcopenia-related markers: a hierarchical moderated regression analysis with age as a moderator

Background Sarcopenia, the age-related loss of skeletal muscle mass, strength and physical performance, is closely linked to habitual physical activity (PA). Whether higher PA is associated with more favourable sarcopenia-related markers, and whether more favourable markers are in turn associated with higher PA, has rarely been examined in both directions. We tested these reciprocal associations and whether they differ with age. Methods In this cross-sectional study, 356 community-dwelling adults (112 men and 244 women; aged 19–85 years) completed the Korean version of the International Physical Activity Questionnaire short form (walking, moderate and vigorous PA, and sitting time). Sarcopenia-related markers were gait speed (foot-worn inertial sensors), handgrip strength (hydraulic dynamometer) and appendicular skeletal muscle mass indexed to height squared (ASM/HT 2 ; bioelectrical impedance). Hierarchical moderated regression was conducted in both directions (PA→markers and markers→PA), adjusted for height, weight and sex, with linear, quadratic and cubic age terms and predictor × age interactions. PA variables were square-root transformed, and heteroscedasticity-consistent standard errors were used. Results Beyond body size, sex and age, PA explained little variance in the markers ( Δ R 2 ≤ 0.019). Age moderated the association between vigorous PA and ASM/HT 2 ( β = −0.07, 95% CI −0.12 to −0.02, p = 0.008), which was positive only in young adults. In the reverse direction, muscle status (the composite of handgrip strength and ASM/HT 2 ) was positively associated with walking PA ( β = 0.34, 95% CI 0.05 to 0.63, p = 0.021), and age moderated its association with vigorous PA ( β = −0.12, 95% CI −0.22 to −0.02, p = 0.018): higher muscle status was associated with more vigorous PA in young adults ( β = 0.73, 95% CI 0.21 to 1.24, p = 0.006) but not in middle-aged or older adults. Moderate PA showed a weak positive association with gait speed ( β = 0.12, p = 0.048). Conclusions In young adults, vigorous PA and muscle measures were associated in both directions, and this coupling was absent in later adulthood; muscle status was also positively associated with walking PA. Because the design is cross-sectional, these associations require confirmation in longitudinal or intervention studies.