Aim: The purpose of this study was to clarify the frequencies and sociodemographic and other characteristics around use of herbal medicine as a remedy for mental health problems in Japan. Methods: Data from the World Mental Health Japan (WMHJ) Survey and US National Comorbidity Survey Replications were analyzed. The WMHJ was conducted in 2002 to 2006, with 4129 respondents. National Comorbidity Survey Replications was conducted in 2002 to 2003, with 9282 respondents. The interview asked the respondents about their use of several types of herbs for mental health problems. Frequencies of use of herbal medicine were compared between Japan and the United States. Multiple logistic regression analyses were conducted to determine sociodemographic and mental health-related correlates of 12-month herbal medicine use. Relevant sampling weights were used to adjust for the sampling designs. Results: The proportion for use of herbal medicines as a remedy for mental health problems in the past 12 months was lower (0.4%) in Japan than that in the United States (3.7%). Low education in both countries (P <.05) was significantly associated with nonuse of herbal medicine. Any anxiety disorder in Japan was significantly associated with herbal medicine use (P <.01), while any mental disorder categories were significantly associated in the United States (P <.01). Discussion: The frequency for use of herbal medicine among patients with mental health problems in the past 12 months was much lower in Japan compared to the United States. Persons with high educational attainment and anxiety disorders used herbal medicine as a remedy for mental health problems more frequently in Japan.