AIM: This study aimed to develop appendicular skeletal muscle mass (ASM) estimating formulas that also consider the presence of paralysis for older adults and people with disabilities. METHODS: This retrospective study analyzed 315 consecutive patients, post-stroke, aged ≥65 years, in a rehabilitation hospital. Six different ASM estimating formulas were developed using a five-fold cross-validation method and compared with the measured ASM obtained from bioelectrical impedance analysis. These formulas included age, gender, height, weight, arm circumference, triceps skinfold, calf circumference and presence of paralysis. Using Pearson's correlation coefficients (r) and intraclass correlation coefficient (ICC), we examined the correlation between the formulas and the measured ASM. The accuracy of the ASM estimating formula for detecting decreased muscle mass was evaluated using the F-value and Matthew's correlation coefficient. RESULTS: Patients' mean ± SD age was 79.0 ± 8.1 years, and 51.4% of them were men. The mean ± SD bioelectrical impedance analysis-measured ASM was 13.7 ± 4.3 kg. Furthermore, 241 (76.5%) patients had decreased measured ASM. The mean adjusted R2 of the developed six formulas was 0.861-0.871. In all formulas, the r and ICC of the estimated ASM for the measured ASM were strong (r = 0.936-0.930 and ICC = 0.928-0.934). These formulas revealed excellent sensitivity (86.0-88.2%), specificity (72.5-81.1%), accuracy (0.838-0.870), F-value (0.899-0.918) and Matthew's correlation coefficient (0.509-0.612) for measured ASM depletion. CONCLUSIONS: We successfully developed ASM estimating formulas using anthropometric measurements considering the presence of paralysis. Thus, these formulas are beneficial for diagnosing sarcopenia in older adults, without requiring any special equipment. Geriatr Gerontol Int 2020; 20: 1145-1150.