Stress and metabolic control in diabetes: psychosomatic evidence and evaluation of methods.

Psychosomatic research findings correlating psychologic stress with diabetic control fail as yet to provide valid conclusions. Investigators have presented many contradictory findings. The two major pathways by which stress could affect control, a) changes in compliance behavior, and b) a neurohumoral axis, have not been clearly distinguished from each other. The study of adolescent cohorts is associated with the problem of heterogeneity, limiting the application of results to other diabetic populations. Methods of determining diabetic control have been incomplete and the definition and measurement of stress have major inadequacies for the analysis of such a complex psychosomatic problem. Existing evidence is comprehensively reviewed and evaluated. The authors use a modified definition of stress to construct a stress scale specific to pregnant diabetics.