Is the 2010 Affordable Care Act Minimum Standard to Identify Disability in All National Datasets Good Enough for Policy Purposes?

Using linked 2009 Current Population Survey (CPS)-Annual Social and Economic Supplement/Social Security Administration records data and a definition of disability based on the six-question disability sequence (6QS) in the CPS-Basic Monthly Survey, we perform a face validity test that shows that the 6QS captures only 66.3 percent of those who administrative records confirm are receiving Social Security benefits based on their disability. Adding a work-activity question to the 6QS increases our capture rate by another 23.1 percentage points for a total of 89.3 percent. We find little difference in the distribution of conditions between those who only report a 6QS-based disability and those who only report a work activity-based disability. The four function-related questions in the 6QS do a relatively good job of capturing those receiving benefits based on these conditions. But the work-activity question does a far better job of capturing those receiving benefits than the two activity-related questions in the 6QS