THE ARTICLE by Horowitz and colleagues1 is most helpful in defining and illustrating some of the difficulties facing clinical researchers who are investigating psychodynamic constructs and psychotherapy. Horowitz et al rightly point out the gap between symptoms and the in-therapy focus on psychotherapeutic intervention. Even when the final goal of treatment is symptom reduction, eg, reduction in depression, the various schools of psychotherapy approach this final goal with quite different mediating goals of treatment. How to measure these mediating goals is extremely important to the advancement of the field. Horowitz and colleagues attempt to bridge dynamic and cognitive-behavioral constructs and to measure these constructs in a reliable manner.
See also pages 625, 633, 637, 639, 642, 646, 649, 651, and 654
Horowitz and colleagues' own work highlights some of the difficulties they and others have faced. Is their method reliable? The data suggest that it is reliable, although one