Aggregation of Psychosocial Risk Factors: Models and Methods

Abstract

Many studies of psychosocial influences on cardiovascular disease (CVD) report associations of one risk or protective factor. However, correlations among such risks are pervasive. Traditional approaches to managing the ambiguities created by these intercorrelations rely on statistical control in a search for “independent” risk factors which, while informative, are imperfect. Alternative quantitative approaches are available to model more directly the interconnections among person characteristics, interpersonal processes, and larger social environments or contexts. In developing a more accurate and refined psychosocial epidemiology of everyday life that is useful in our efforts to understand, prevent, and manage CVD, we should supplement the traditional emphasis on independent risks with conceptual models and methodologies that address aggregated risks more directly as they naturally operate and unfold.