In modern medical care in which Kampo and Western drugs are often combined, it is extremely important to clarify drug–drug interaction (DDI) to ensure safety and efficacy. However, there is little evidence of DDI in Kampo medicines. Therefore, as part of our studies to clarify the DDI risk for Kampo medicines, we evaluated the effects of five Kampo medicines [yokukansan (YKS), rikkunshito (RKT), shakuyakukanzoto (SKT), hangeshashinto (HST), and goshajinkigan (GJG)] that are widely used in Japan, on drug transporter P-glycoprotein (P-gp) using a Caco-2 permeability assay. These Kampo medicines inhibited the P-gp transport of digoxin through a Caco-2 cell monolayer. The IC50 values were 1.94–10.80 mg/ml. Of the five Kampo medicines, YKS showed the strongest inhibition (IC50 = 1.94 mg/ml), which was attributed to Uncariae Uncis Cum Ramulus. Unfortunately, we could not find the active ingredients responsible for its action. Finally, the Igut/IC50 values for the five Kampo medicines were calculated, and the DDI risk was objectively evaluated according to the criteria in the DDI guidance issued by the Japanese Ministry of Health, Labor, and Welfare and the US Food and Drug Administration. The Igut/IC50 values for the five Kampo medicines were ≤3.4. As these values were <10, they were evaluated as having a weak P-gp inhibitory effect that does not require further verification in humans, suggesting that the DDI risk due to P-gp inhibition for these Kampo medicines is low. The results should provide useful clinical information on the safety and efficacy of the combined use of Kampo and Western medicines.Electronic supplementary materialThe online version of this article (10.1007/s11418-018-1222-x) contains supplementary material, which is available to authorized users.