The efficacy of psychosocial intervention for pain in breast cancer patients and survivors: a systematic review and meta-analysis

Persistent pain after breast cancer treatment is prevalent, and not all patients respond sufficiently to pharmacological treatment. Pain is recognized as a multi-dimensional phenomenon, which includes psychological and social components, and several clinical trials have investigated the efficacy of psychosocial interventions on pain in cancer patients and survivors. Our aim was to systematically review and quantify the existing research on the effect of psychosocial interventions on pain in breast cancer patients and survivors. Two independent raters reviewed 474 abstracts for eligibility, leading to the identification of 26 independent and eligible studies published between 1983 and 2012, which were assessed for their methodological quality and subjected to meta-analytic evaluation. A total of 1786 participants were included in the analyses. A statistically significant and robust overall effect size was found across all included studies (Hedges g = 0.37, 95 % CI: 0.20-0.40; p < 0.001). However, the effect size was considerably smaller (0.21), when adjusted for possible publication bias. Furthermore, the results were heterogeneous, and when exploring the sources of heterogeneity, studies of higher methodological quality were found to yield a more conservative effect size (g = 0.21, 95 % CI: 0.02-0.41) than studies of poorer quality (g = 0.65, 95 % CI: 0.25-1.04). The results also indicated that patient educational approaches yielded a larger effect size (g = 0.64) than relaxation-based interventions (g = 0.31, 95 % CI: -0.05-0.67) and supportive group therapy (g = 0.17, 95 % CI: 0.02-0.32). Taken together, while suggestive of psychosocial intervention as an effective tool in the management of pain among breast cancer patients and survivors, the results should be interpreted as preliminary. The methodological quality of the existing research varied considerably, and only few studies had selected patients on the basis of the presence of pain and included pain as the primary outcome.