Background: Whether St. John’s wort extract is effective for depression is still a matter of discussion outside German-speaking countries. In the 2005 update of our Cochrane review, we had the impression that trials from German-speaking countries yielded findings which were more favourable to hypericum than trials from other countries. Therefore, we decided to include a formal analysis of the influence of country of origin into our 2008 update. Objectives: To investigate whether the results of trials performed in German-speaking countries differ from those performed elsewhere. Methods: The 28 double-blind randomized controlled trials comparing mono-preparations of St. John’s wort (Hypericum perforatum L.) extract with placebo (17 comparisons) or standard antidepressants (17 comparisons) in patients meeting criteria for major depressive disorder included in the 2008 version of the review were categorized according to country of origin (Germany, Austria, and Switzerland vs. other). Univariate and multiple meta-regression analyses were performed to investigate the influence of study precision, baseline HAMD (Hamilton Depression Scale) values, and country of origin on responder rate ratios and post-treatment HAMD scores. Results: Placebo-controlled trials from German-speaking countries yielded significantly more positive results than trials from other countries in univariate and multivariable analyses, both regarding response rate ratios (p-values 0.02 and 0.03, respectively) and HAMD post-treatment scores (< 0.001 and < 0.001). While placebo response rates were similar in trials from German-speaking countries and from elsewhere, response rates in hypericum groups tended to be higher in trials from German-speaking countries. In trials comparing hypericum and standard antidepressants, a significant effect of country of origin was only observed in the univariate analysis on responder rate ratios. Conclusions: Trials from German-speaking countries report more favourable results on hypericum than trials from other countries. Potential explanations for these findings might be the inclusion of different patients (in spite of the same diagnostic classification) or differences in the design and performance of trials.