In the May 1992 issue of the Archives, Perry et al1 reported on the relationship between psychosocial and immune variables in human immunodeficiency virus type 1 (HIV-1) seropositive individuals. This is the second Archives article devoted to this area that did not find significant associations on the whole.1,2 We are impressed by the efforts involved in the generation of these two studies. However, we are concerned that the conclusion of Perry et al1 suggesting no relationship between psychosocial and immune variables is premature. Thus, we raise the following four points.
First, there was no discernible theoretical rationale interrelating the broad set of psychosocial and psychiatric measures assessed and driving the statistical hypotheses tested. In other reports, a theoretically driven psychosocial model interrelating stressful life events, social support, and coping style has demonstrated associations with psychological distress,3 sures,4-11 and clinical health status.11,12 prior studies in