Replication of the multidimensionality of activities of daily living. (Fitzgerald, Smith, Martin, Freedman, & Wolinsky, 1993)

Fitzgerald, J. F., Smith, D. M., Martin, D. K., Freedman, J. A., & Wolinsky, F. D. (1993). Replication of the multidimensionality of activities of daily living. Journal of Gerontology, 48(1), 4.

Department of Medicine, Indiana University School of Medicine.

Abstract: We attempted to replicate the three-dimensional factor structure of a previously proposed ADL scale and demonstrate an association between the advanced ADL dimension and cognitive function. Data used in these analyses were baseline assessments of health and functional status of hospitalized patients enrolled in a randomized controlled trial of case managers as a means of reducing health care utilization. We submitted 14 items from the OARS to a two-stage process of principal components factor analysis. Four significant dimensions emerged that were remarkably similar to the advanced, basic, and household ADL dimensions reported by Wolinsky and Johnson (1991). In this sample of hospitalized patients, however, incontinence emerged as a weak fourth dimension. Multiple regression of SPMSQ mental status examination scores on these ADL dimensions demonstrates the association between cognitive function and the advanced ADL dimension. These data confirm that the underlying structure of ADLs consists of at least three separate dimensions, one of which is aligned with cognitive capacity.

http://geronj.oxfordjournals.org/cgi/reprint/48/1/S28

My notes:

Some authors thought Kat's et al.'s ADL and Lawton & Brody's IADL should be modeled separately. Spector et al & Kempen and Suurmeijer suggested a unidimensional structure. Wolinsky et all suggested a three dimensional model: basic ADL, Household ADL and Advanced ADL.

The authors of this study used principal Component analyses with a different dataset, trying to replicate the same 3 factor structure obtained in Wolinsky & Johnson (1991). The initial results of their analyses resulted in a 4-factor structure. In addition to Basic ADL, Household ADL and Advanced (cognitive) ADL factors, the fourth factor concerning incontinence emerged in their analysis.

Although results of this report did successfully replicate the 3-factor structure of ADLs as proposed by Wolinsky & Johnson (1991), one comment I have is that, for the purpose of this study, wouldn't confirmatory factor analysis (CFA) be more pertinent? At the same time, based on the analyses I ran using NLTC survey data, I also came upon the cognitive factor...