Orchid AB
How Does the Way the Construct of Disability Is Measured Affect the Estimates Obtained in Selected National Surveys? (303625)
*Frances M. Chevarley, Agency for Healthcare Research and QualityDavid W. Keer, Disability Research Consultant
Keywords: disability, AHRQ, MEPS, NHIS, ACS, DHHS, ACA
Prevalence estimates of persons with disabilities are important for public health planning and other purposes. However it is often difficult to compare estimates derived across data sources due to differences in survey purpose, scope, and target populations; differences in the data collection process including questions and response categories; and differences in analysis where specific measures are constructed post-data-collection based on the availability of questions. There has been a great deal of work done to date to reach consensus on disability measures. In order to include more disability data in the Agency for Healthcare Research and Quality’s National Disparities and Quality Reports (NHDR/QR) a process was initiated in 2005 that culminated in the development of comparable disability measures for the Medical Expenditure Panel Survey (MEPS), National Health Interview Survey (NHIS), and Medicare Current Beneficiary Survey (MCBS) that have been used in the NHDR/QR since 2007. On a separate track, another process was started in 2002 to review the existing American Community Survey (ACS) disability questions and this culminated in the development of the 2008 ACS disability questions which are still in use. In response to the Affordable Care Act’s (ACA’s) requiring data collection standards for disability, the DHHS Secretary mandated in 2011 the use of the 2008 ACS disability questions in all HHS sponsored population based surveys. Previous work explored and evaluated how the 2008 ACS disability questions and estimates compared with prior disability measures using data from the MEPS, the NHIS, the MCBS, and the ACS. Current analyses compare the 2008 ACS disability question estimates across the MEPS, the NHIS, and possibly other surveys.