Does psychosocial intervention improve survival in cancer? A meta-analysis

Background: There is growing evidence that positive psychosocial intervention improves the wellbeing of cancer patients. Two meta-analyses conducted to date confirmed a significant small-to-moderate effect on quality of life. Previous randomized trials reported that psychosocial intervention also improved survival. However, more recent trials failed to detect a difference in survival. A systematic review of randomized trials that have examined the effectiveness of psychosocial intervention in cancer patients in terms of survival prolongation was conducted. Methods: Randomized trials published between 1966 and June 2002 were identified through the databases of MEDLINE, EMBASE, CancerLit, CINAHL, Cochrane Library and reference lists of relevant articles. Relevant data were abstracted. The results of randomized trials were pooled using meta-analyses to estimate the effect of treatment on overall survival at one and four years in all cancer patients and also in breast cancer patients with metastases. Results: Eight trials, which involved a total of 1062 patients (all cancer histologies), were identified. One- and four-year overall survival rates were obtained from eight trials and six trials, respectively. There was no statistically significant difference in the overall survival rates at one and four years [P = 0.6; RR 0.94 (95% Cl 0.72, 1.22)] and [P = 0.5; RR 0.93 (95% Cl 0.77, 1.13)], respectively. Four trials examined 511 metastatic breast cancer patients. Again, there was no statistically significant difference in the overall survival rates at one and four years [P = 0.3; RR 0.87 (95% Cl 0.67, 1.14)] and [P = 0.3; RR 0.91 (95% Cl 0.76, 1.10)], respectively. Conclusions: Psychosocial intervention does not prolong survival in cancer. This meta-analysis can not rule out small effect sizes because of the small number of trials and small trial sizes.