Background <p>Osteosarcopenia, the coexistence of sarcopenia and low bone mass, raises the risk of falls, fractures, and functional decline, yet its diagnosis depends on dual-energy X-ray absorptiometry (DXA) and muscle function testing, which are difficult to apply routinely in the community. Whether gait variability, a marker of neuromuscular control, is associated with osteosarcopenia and adds information beyond mean gait measures is unknown in Korean community-dwelling older women. We examined this association and the discriminative ability of gait variability for osteosarcopenia.</p> Methods <p>In this cross-sectional study, 190 community-dwelling women aged 65 years or older were classified by DXA and the 2019 Asian Working Group for Sarcopenia criteria into normal (<i>n</i> = 32), osteoporosis (<i>n</i> = 118), and osteosarcopenia (<i>n</i> = 40) groups. Spatiotemporal gait parameters and their trial-to-trial variability, expressed as the coefficient of variation (CV), were measured during overground walking with the OptoGait system. Binary logistic regression and receiver operating characteristic (ROC) analyses were performed.</p> Results <p>From the normal to the osteosarcopenia group, gait speed and stride length decreased while stride length CV increased. After adjustment for age and confounders, shorter stride length, higher stride length CV, and lower body mass index (BMI) were each independently associated with osteosarcopenia (all <i>p</i> ≤ 0.013). A combined model of these three variables discriminated osteosarcopenia with an area under the curve of 0.83 (sensitivity 75.0%, specificity 81.3%).</p> Conclusions <p>In this cross-sectional sample, gait variability, combined with stride length and BMI, was associated with osteosarcopenia and may offer a practical, noninvasive screening adjunct to prioritize older women for confirmatory DXA assessment. Prospective validation is needed.</p>