Understanding Within-Group Variability of Everyday Cognition in Aging Black/African American Adults: A Mimic (Multiple Indicators, Multiple Causes) Model Approach

Abstract

Background/Study Context: Everyday cognition represents the ability to solve problems within domains that are representative of issues faced by adults on a daily basis. The current study examined individual differences in everyday cognitive ability among aging Black/African American adults.

Methods: Demographic data on age, gender, education, physical functioning, chronic illnesses, self-reported health, and depression were collected from 248 African American adults (mean age = 67.8 years, standard deviation = 8.47 years). A multiple indicators, multiple causes (MIMIC) modeling approach was used to examine the associations of individual characteristics with latent everyday cognitive ability and composite score indicators.

Results: Age, depressive symptoms, and number of chronic illnesses were negatively related to latent everyday cognition. The individual characteristics of age, depressive symptoms, self-rated health, and education were directly associated with composite indicators of latent everyday cognition. This suggests that within this sample of older Black/African American adults that certain composite scores (i.e., telephone use, food preparation, and finances) may be particularly sensitive to these individual characteristics.

Conclusion: These results identify specific sources of variability in everyday cognitive ability among aging Blacks/African Americans. These individual differences should be accounted for when studying everyday cognition among Blacks/African Americans and when comparing the everyday cognitive ability of Blacks/African Americans with other groups.

Notes

Note. CES-D = Centers for Epidemiological Studies Depression Scale (CES-D; Radloff, 1977); IADL = Instrumental Activities of Daily Living (Lawton & Brody, 1969); EPT = Everyday Problems Test (Willis & Marsiske, 1993).

¹Model modifications were also tested using an iterative stepwise procedure where all direct effects were tested at each stage and the direct effect that improved model fit to the greatest degree remained in the model. The results were identical to results based on the modification indices.

Note. IADL = Instrumental Activities of Daily Living (Lawton & Brody, 1969); CES-D = Centers for Epidemiological Studies Depression Scale (CES-D; Radloff, 1977); EPT = Everyday Problems Test (Willis & Marsiske, 1993). Only significant direct effects are presented, the remaining direct effects were constrained to equal zero.

*p < .05.

^†Parameters are fully standardized with respect to the means and variances of the covariates and dependent variables.

§Parameters are standardized with respect to the mean and variance of the dependent variable. n.e. = not estimated.

Note. df =degrees of freedom; CFI = comparative fit index; RMSEA = root mean square error of approximation.