Proxy-Reports in the Ascertainment of Disability Prevalence with American Community Survey Data.

Abstract

Population estimates on disability prevalence inform policy makers and public health professionals. Understanding how factors capable of affecting measurement (e.g., proxy-report) vary in the population is important for establishing level of confidence in sample-derived population estimates. To establish how use of proxy-reports varies by six disability types stratified by sex, race-ethnicity, and age group. Specific aim is achieved by investigating the number of proxy-reports used amongst the disable population. Cross-sectional study using American Community Survey (ACS) Public Use Microdata Sample (PUMS) 3-year file collected during 2009-2011 survey period. Community-dwelling population in continental United States (US). The unweighted count of 6,003,183 individuals in the microdata are said to represent about 193,277,485 individuals in the continental US population. Stratified disability period estimates are computed. Amongst the disable: the number of proxy-reports; allocations; and Person Inflation Ratios (PRIs) are presented by disability type. Half of all the reported disabilities are derived through the use of proxy-report. In addition, high rates of item-allocation and PRIs are generally found in race-ethnic minorities. Proxy-report use and PRIs are lower for those aged > 65-but not allocation rates. Although use of proxy report in the ascertainment of disability varies in complex ways, data suggest prevalence of proxy reports is lowest amongst Non-Latino-Black females ages 21 to 64. Efforts toward providing clinicians with high quality descriptive epidemiology should continue as a reliable thermometer for measuring disability in the population is needed.