Johnson R J, Wolinsky F D
Kent State University.
J Health Soc Behav. 1993 Jun;34(2):105-21.
This article builds on earlier conceptualizations of the structure of health status to propose a more complex, multiequation model that examines the interrelationships among its various dimensions. As such, the focus is not merely on the identification of the direct effects of a variety of factors on perceived health status, but on how the constructs of disease, functional limitation, and self-rated health interrelate. In so doing, we expose the inherent problems of several complex and contaminated items routinely included in applications of established functional health status measures. The source of these problems lies in the lack of specificity or conceptual clarity for the individual items in these established measures. Potential biases are discussed, and several methods and strategies for addressing these problems are explored. Alternative scales are constructed and their psychometric properties are presented. Researchers who rely on public use data bases containing these scales should be aware of the potential biases and either modify the scales or use other appropriate methodologies to control for the measurement contamination.
