Response specificity revisited: stress and health status.

Abstract

Numerous authors have obtained significant, but low correlations between life events and specific health status measures. Even with the inclusion of moderating variables, such as coping, the relationship between life stress and health status measures has rarely exceeded .50. It has been argued that failure to enhance the correlation has occurred because response specificity has not been controlled. A representative community sample of 184 adults from London, Ontario, was surveyed with respect to life stress variables, stress-response specificity, and various illness behaviors. First-order factor analysis of the stress-response specificity measure indicated an emotional/psychological response factor, an eating response factor, a drug/alcohol response factor, and several physical response factors. Second-order factor analysis indicated a psychological response and a physical response factor. Each of the illness behaviors was regressed, in a stepwise procedure, upon the life stress measures and related first- and second-order stress-response specificity factors. As predicted, significant relationships were found between the related stress-response specificity and illness behavior measures. These results support the importance of stress-response specificity as a moderator variable in investigations of the relationship between life stress and illness behavior.