BACKGROUND: Focusing on the relationship between frail older people and gait speed is vital to minimize the need for long-term care or increased support. The relationship between gait speed, respiratory muscle strength, and dynamic balance, is not well understood in older people requiring long-term care or support. Therefore, this study aimed to provide new insights into the relationship between gait speed, respiratory muscle strength, and dynamic balance in community-dwelling older people who required long-term care or support and used a daycare center. METHODS: This was a cross-sectional study of 49 community-dwelling older people (21 men, 28 women) aged ≥65 years who were certified as requiring long-term care or support under the Japanese system. The participants' maximal inspiratory pressure (PImax), maximal expiratory pressure (PEmax), walking speed (maximal and normal walking speed), and maximal double-step length test (MDST) results were recorded. The measurement data were evaluated using Pearson's correlation coefficient and multiple regression analysis. RESULTS: Pearson's correlation coefficient revealed correlations between PImax and the following: maximal walking speed (r = 0.606, p < 0.001), normal walking speed (r = 0.487, p < 0.001), and MDST (r = 0.435, p = 0.002). Correlations were also observed between PEmax and the following: maximal walking speed (r = 0.522, p < 0.001), normal walking speed (r = 0.467, p < 0.001), and MDST (r = 0.314, p = 0.028). Moreover, a correlation was found between MDST and both maximal walking speed and (r = 0.684, p < 0.001) and normal walking speed (r = 0.649, p < 0.001). The effect size was 0.379. Multiple regression analysis using a forced entry method with maximal walking speed as the dependent variable showed that maximal walking speed was significantly associated with MDST (p < 0.001) and PEmax (p = 0.036), with an effect size of 0.272. The model's adjusted coefficient of determination was 0.593 (p < 0.001). Multiple regression analysis using a forced entry method with normal walking speed as the dependent variable showed that normal walking speed was significantly associated with MDST (p < 0.001) and PEmax (p = 0.021), with an effect size of 0.272. The model's adjusted coefficient of determination was 0.497 (p < 0.001). Multiple regression analysis using a forced entry method with MDST as the dependent variable showed that MDST was significantly associated with PImax (p < 0.025), with an effect size of 0.243. The model's adjusted coefficient of determination was 0.148 (p = 0.017). CONCLUSIONS: Respiratory muscle strength and dynamic balance were related to walking speed in older people requiring long-term care or support.