Commentary: psychological treatment of physical symptoms: who should get what?

asthma, juvenile diabetes mellitus, or cancer. As described by the authors, most studies have focused on main effects, with varied results. Either investigators did not consider potential moderating variables or the samples in these studies were too small to test meaningfully for interactions. With increasingly limited resources in the age of “managed care,” individual differences in responses to interventions become a salient issue and of practical importance (Barr, Boyce, & Zeltzer, 1994; Boyce, Barr, and Zeltzer, 1992; Zeltzer, Barr, McGrath, & Schechter, 1992). Thus, as the authors note, one of the key directions for future research is to design studies that will better address not only “what works better than what else” but also “who should get what” for optimizing positive outcomes. Unless the identification of individual differences is embedded in intervention studies, results will likely be diluted and not as useful to the clinician. From an historical perspective, it appears that various psychological interventions were in fashion over time as the topic of study until newer, more “hip” interventions took their place. For example, intervention shifts took place spanning the spectrum from psychoanalysis to family systems approaches to relaxation training to hypnotherapy to social skills training, to name a few. However, the focus has been on testing an intervention strategy rather than on studying likely mechanisms of psychobiologic interface with subsequent development and testing of theory-driven interventions designed to affect this interface in very specific ways. For example, salient moderating variables on asthma severity, diabetic control, or chemotherapy-related Journal of Pediatric Psychology, Vol. 24, No. 4, 1999, pp. 333–334