Aims: Differences between racial and ethnic groups in subjective measures of health are well documented. However, subjective assessments also differ by confounding variables related to race/ethnicity as well as other potential sources of systematic bias, for example, language and/or culture. Hence, some of the most widely used subjective measures may not be valid across all sub-populations in the United States. This study examines measurement equivalence properties of the SF-36 (subjective measure) and a performance based measure (PBM) index of physical functioning (""""""""gold standard"""""""") among several diverse racial/ethnic groups, addressing the following questions: 1) are there structural equivalence differences (measurement invariance) in the subjective measure and PBM across these populations; 2) are there indications of differential item functioning (DIP) within the subjective measure by racial/ethnic group; 3) does the subjective measure predict the performance-based measures equally across each group (metric and scalar equivalence), after controlling for other variables; and 4) do any observed differences vary according to gender? Methods: Approximately 330 subjects age 45-64 from each of the 4 racial/ethnic groups (non- Hispanic whites, non-Hispanic African-Americans, English-speaking Hispanics, Spanish-speaking Hispanics; N=1320) will be recruited from 4 general medicine clinics and other community centers in Cleveland and Chicago. The SF-36, a short demographic survey, and 11 physical tasks (upper and lower extremity strength, endurance, agility/dexterity, gait, and stair climbing) will be administered to each subject using trained clinical research assistants. Medical charts will be reviewed and chronic conditions and medications recorded. Data analysis using structural equation modeling, confirmatory factor analysis, multivariate regression techniques, and item response theory (IRT) will be used to investigate the above aims. Significance: This study will be the first to validate subjective measures of physical functioning and general health status using a performance-based """"""""gold standard"""""""" including middle-age English and Spanish speaking Hispanics, adding to the knowledge of health status measurement of our fastest growing ethnic population. Regardless of outcome, results will provide heuristic value in directing further research, and will provide guidance in evaluating changes to current surveys. ? ? ?
| Sudano, Joseph J; Perzynski, Adam; Love, Thomas E et al. (2011) Measuring disparities: bias in the Short Form-36v2 among Spanish-speaking medical patients. Med Care 49:480-8 |
| Ruo, Bernice; Baker, David W; Thompson, Jason A et al. (2008) Patients with worse mental health report more physical limitations after adjustment for physical performance. Psychosom Med 70:417-21 |
