In his best known contribution to the field of psychooncology, the late Dr Bernard H. Fox applied his breadth of scholarship in biopsychosocial cancer epidemiology to address the question of whether and to what extent stress and other psychosocial factors may contribute to cancer risk. Less well known but equally important to the field is his incisive critique of the 1989 study by Spiegel et al. on survival time of patients with metastatic breast cancer following a psychosocial intervention. This essay represents an attempt to take Fox's line of thought to the next logical level of rethinking research on psychosocial interventions in biopsychosocial oncology. Following an analysis of the inadequacy of randomized clinical trials (RCT) to evaluate the causal effects of psychosocial interventions on cancer outcomes and distinguish these from mere prediction, an integrated RCT design is suggested to take into account the psychogenicity of a given intervention, potential mediating mechanisms, and individual differences that could help illuminate hypothesized causal processes linking an experimental intervention and cancer outcomes.