The dataset contains records for 197,302 registered persons with disabilities gathered in 2020 via telephone interviews supplemented by in-person visits. It comprises 155 variables across seven domains: demographics; economic conditions; internet access and use; employment and skills; medical and rehabilitation services and assistive products and technology; accessible environment; and cultural and sports participation. Anonymized data, raw tables, and a detailed codebook are openly available on Harvard Dataverse 15. The resource enables fine-grained analyses of urban-rural and district heterogeneity, functional-limitation patterns, and gaps between required and received services. It also supports the construction of equity-sensitive indicators, multivariable and multilevel models of service utilization, and benchmarking for programmed monitoring. The cross-sectional design and a focus on certificate holders should be considered when interpreting results, yet the scale and coverage make the dataset valuable for comparative studies and careful linkage with public statistics under appropriate safeguards.
Guo et al. (2026) studied this question.