Context: Rhodiola rosea L. (Crassulaceae) products are popular natural remedies with a worldwide distribution. Recent studies have revealed potent CYP inhibition by R. rosea extracts both in vitro and in vivo, but information on in vitro CYP inhibition by commercial products are lacking. Variations in commercial R. rosea product quality have also been published. Objective: This study evaluates the variation of in vitro CYP inhibition potential and product quality of six commercially available R. rosea products. Materials and methods: Human CYPs isolated from baculovirus-infected cell system were incubated with testosterone (CYP3A4), dextromethorphan (CYP2D6) or phenacetin (CYP1A2). Positive CYP inhibitors ketoconazole (CYP3A4), quinidine (CYP2D6) and b-naphtoflavone (CYP1A2) were used as controls. Quantification of rosavin, rosarin, rosin, tyrosol and salidroside were used to evaluate R. rosea content. Results: IC 50 values ranged from 7.2-106.6 lg/mL for CYP3A4, 13.0-186.1 lg/mL for 2D6 and 10.7-116.0 lg/mL for 1A2. The tincture formulation of R. rosea was the strongest inhibitor giving the lowest IC 50 values of 7.2 ± 0.7, 13 ± 1.7 and 10.7 ± 5.6 lg/mL, respectively. CYP3A4 was significantly more inhibited by the different products than CYP1A2 (p < .05). One of the six products did not contain any rosavin, rosarin or rosin and is not a R. rosea product. Constituent concentrations were not linked to enzyme inhibition. Discussion and conclusion: The present results show a large variation in inhibitory potential between the products. Several of the products demonstrate similar inhibition levels as the product Arctic Root already proven to inhibit CYP enzyme activity in man.ARTICLE HISTORY