Finding what is not there: unwarranted claims of an effect of psychosocial intervention on recurrence and survival.

In a recently published randomized trial, Andersen et al. claimed to observe a reduced risk of recurrence and improved survival among women with early stage breast cancer who were assigned to a psychological intervention versus an assessment-only group. Anderson et al. concluded that "psychological intervention, as delivered and studied here, can improve survival." The current commentary challenges that conclusion on methodological and statistical grounds, noting that the study by Andersen et al. was not designed to assess survival and used methods that capitalized on chance, making it highly unlikely that their claims could be replicated. No other study designed to assess whether psychosocial intervention provides a survival benefit for cancer patients has ever demonstrated such an effect; and, currently, there is no support for the hypothesis that survival benefits can be attributed to psychosocial intervention. The authors of this commentary argue that much needs to be learned at the more basic biobehavioral level about the impact of stress or psychological factors on tumor biology before even considering whether large clinical trials are warranted. In addition, making the claim that psychological interventions improve survival is not evidence-based, is potentially harmful, and detracts from the potential significant benefits of psychological interventions related to quality of life.