Comparing estimates of disability prevalence using federal and international disability measures in national surveillance

The National Health Interview Survey (NHIS) is changing the annual inclusion of standardized disability identifiers, reinvigorating the priority to examine existing disability question sets. These sets include questions developed by the United States (U.S.) National Center for Health Statistics in conjunction with the U.S. Census Bureau (the American Community Survey questions, ACS) and United Nations (the Washington Group Short Set questions, WGSS), that are policy relevant, comparable across populations, and short enough to be included in censuses and surveys across countries. To compare disability prevalence estimates from federal and international standardized disability questions across demographic factors. Bivariate analysis of disability question sets asking adults about vision, hearing, ambulation, cognition, and self-care difficulties and demographic factors using secondary data from supplements in the 2010 and 2013-2015 NHIS. Our study found substantial and statistically significant differences in the percentage of disabilities (overall and by type) based on comparable ACS and WGSS questions across demographic categories. Dependent on response coding, WGSS-based disability prevalence was consistently and significantly larger or smaller than ACS-based disability prevalence. Overall disability prevalence using ACS and two different WGSS response combinations were 16.3% and 9.2% or 39.4%, respectively. ACS and WGSS measures identify predictably different sized populations of adults with disabilities. Further, with some exceptions, ACS and WGSS questions identify populations with disabilities with relatively consistent demographic factors. Additional research is recommended to understand the comparability of disability prevalence and health disparities and inequities people with disabilities experience when using these measures.