B Han1, B J Small, W E Haley. 1. Special Populations Research Branch, Division of Programs for Special Populations, Bureau of Primary Health Care, Health Resources and Services Administration, Bethesda, MD 20814, USA. bhan@hrsa.gov
Abstract
OBJECTIVE: To examine whether depressive symptomatology is a third fundamental component of the structure of self-rated health, in addition to two other components (physical disease and functional disability) among community-dwelling older adults with stroke. DATA SOURCES AND STUDY SETTING: A total of 591 community-dwelling older adults with stroke were identified from the 1993 Asset and Health Dynamics among the Oldest-Old (AHEAD) national survey of community-dwelling older adults. STUDY DESIGN: A cross-sectional study. Structural equation modeling was applied to compare a widely used two-factor model of self-rated health with a model adding depression as a third possible factor. PRINCIPLE FINDINGS: The hypothesized three-factor model explained additional 21% more variance of self-rated health of older adults with stroke (R2 = 79%, NNFI = 0.95, CFI = 0.96, RMSEA = 0.04) as compared with the two-factor biomedical model (R2 = 58%, NNFI = 0.95, CFI = 0.98, RMSEA = 0.05). The three-factor model was statistically different from the two-factor model. CONCLUSIONS: Greater attention should be given to the theoretical structure of self-rated health of older adults with stroke, particularly, the significant impact of depression on their self-rated health.
OBJECTIVE: To examine whether depressive symptomatology is a third fundamental component of the structure of self-rated health, in addition to two other components (physical disease and functional disability) among community-dwelling older adults with stroke. DATA SOURCES AND STUDY SETTING: A total of 591 community-dwelling older adults with stroke were identified from the 1993 Asset and Health Dynamics among the Oldest-Old (AHEAD) national survey of community-dwelling older adults. STUDY DESIGN: A cross-sectional study. Structural equation modeling was applied to compare a widely used two-factor model of self-rated health with a model adding depression as a third possible factor. PRINCIPLE FINDINGS: The hypothesized three-factor model explained additional 21% more variance of self-rated health of older adults with stroke (R2 = 79%, NNFI = 0.95, CFI = 0.96, RMSEA = 0.04) as compared with the two-factor biomedical model (R2 = 58%, NNFI = 0.95, CFI = 0.98, RMSEA = 0.05). The three-factor model was statistically different from the two-factor model. CONCLUSIONS: Greater attention should be given to the theoretical structure of self-rated health of older adults with stroke, particularly, the significant impact of depression on their self-rated health.
Authors: Nancy E Mayo; Susan C Scott; Mark Bayley; Angela Cheung; Jayne Garland; Jeffrey Jutai; Sharon Wood-Dauphinee Journal: Qual Life Res Date: 2013-12-19 Impact factor: 4.147