The aims of our study were to: (1) develop the Disability Grading Decision Support System (DGDSS) and to (2) compare the new International Classification of Functioning, Disability, and Health (ICF)-based disability determination tool (ICF-DDT) with the diagnosis-based disability determination tool (D-DDT). A total of 9357 patients recruited from 236 accredited institutions were evaluated using the ICF-DDT and the D-DDT, and the presence, severity, and category of the disability identified using the two determination tools were compared. In the DGDSS, the ICF-DDT consisted of four models comprising nine modules to determine the presence and the severity of the disability. The differences between models (modules) are the different combinations of World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0) and Scale of Body Functions and Structures. Compared with the D-DDT, more patients were determined to be disability-free when using the ICF-DDT. Module 1-1 had the highest profoundly severe rate, and module 2-2 had the highest mild and moderate disability rates. Module 2-1 had the highest severe disability rate. Module 1-1 resulted in the smallest difference, and module 3-1 resulted in the largest difference, compared with the D-DDT. Feedback from users suggested that the DGDSS is a robust system if the original data are accurate. The presence, severity, and category of the disability determined using the ICF-DDT and the D-DDT were significantly different. The results of the DGDSS provide information for policymakers to determine the optimal allocation of social welfare and medical resources for people with disabilities.