The 131st Annual Meeting (November 15-19, 2003) of APHA

Abstract #57617

Susan Kinne, PhD, Center for Disability Policy and Research, University of Washington, 146 N Canal Street #313, Seattle, WA 98103, (206) 685 4769, susaki@u.washington.edu

In 2001, the Washington State Behavioral Risk Factor Surveillance Survey (BRFSS) included two items to identify respondents with disabilities. The BRFSS does not sample on or weight by disability, so respondents with disabilities may not be representative of adults with disabilities in the population. It is important to identify sample biases and their possible effects in estimating differences in health behaviors of people with and without disabilities. This analysis compared weighted data on age, sex, education, race and ethnicity, employment and household income of 2001 BRFSS respondents with disabilities (23.1% of the sample age= 21) to 2000 Census data on Washington adults with disabilities (21.7% of the population age =21), bearing in mind that the definitions of "disability" differ in the two sources. Relative to the Census, the BRFSS disability sample had more education, higher employment, and lower proportions of Hispanics and Asian Americans. Education and employment are associated with better health behaviors and greater access to health services, and often indicate less severe impairment among people with disabilities. Over-representation of people with these traits in the disability sample will tend to reduce estimated health and behavior differences between people with and without disabilities. The presentation will conclude with examples of health behavior differences and ways in which the apparent sample bias might be treated.

Learning Objectives:

• Identify differences in BRFSS and Census 2000 methodology and definitions of disability
• Identify ways in which BRFS and Census estimates of adults with disabilities differ for a state population
• Understand how sample composition differences can translate into underestimates of health and behavior differences between adults with and without disabilities

Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: Washington State Department of Health/Behavioral Risk Factor Surveillance Survey
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.