Background: We conducted a systematic review and meta-analysis to determine the efficacy of psychological interventions for premenstrual syndrome. Methods: We systematically searched and selected studies that enrolled women with premenstrual syndrome in which investigators randomly assigned them to a psychological intervention or to a control intervention. Trials were included irrespective of their outcomes and, when possible, we conducted meta-analyses. Results: Nine randomized trials, of which 5 tested cognitive behavioural therapy, contributed data to the meta-analyses. Low quality evidence (design and implementation weaknesses of the studies, possible reporting bias) suggests that cognitive behavioural therapy significantly reduces both anxiety (effect size [ES] = -0.58; 95% confidence interval [CI] = -1.15 to -0.01; number needed to treat [NNT] = 5), and depression (ES = -0.55; 95% CI = -1.05 to -0.05; NNT = 5), and also suggests a possible beneficial effect on behavioural changes (ES = -0.70; 95% CI = -1.29 to -0.10; NNT = 4) and interference of symptoms on daily living (ES = -0.78; 95% CI = -1.53 to -0.03; NNT = 4). Results provide much more limited support for monitoring as a form of therapy and suggest the ineffectiveness of education. Conclusions: Low quality evidence from randomized trials suggests that cognitive behavioural therapy may have important beneficial effects in managing symptoms associated with premenstrual syndrome. Copyright © 2008 S. Karger AG.