Rationale: Mali is one of the poorest countries in the world with a GDP of 260 USD per capita. The public health expenditure was 10.2% of the GDP in 1999. 59% of the population has access to health facility within 15 km (1). In Mali there is since 1995 an institutional framework that regulates the practice of traditional medicine (2). In the favor of this law healers and herbalists are allowed to open traditional clinics and traditional medicines shops. To improvetheir state of health, people use both conventional and traditional medicines. Traditional medicine, being a significant element in the cultural patrimony, still remains the main resource for a large majority of people. The accessibility to conventional drugs is however increasing, especially in urban areas. This development can lead to the combination of the two types of medicines. According to the literature herbal medicines can interact with conventional drugsin many ways. It is therefore important to study the knowledge of herbconventional drug interactions among traditional practitioners. On the other hand one of the objectives of the Department of Traditional Medicine is to develop new medicines (ITMs) from natural plants. The traditional healers and herbalists are the main informants for the DMT in the production of ITMs; therefore tostudy their knowledge of herb-herb interactions is necessary.General objectives: The general objective was to determine the level of knowledge and the practices regarding herb-herb and herb-conventional drug interactions of the traditional practitioners (healers and herbalists) registered by the Department of Traditional Medicine (DMT) and operating in Bamako.Study Design: Cross sectional descriptive study.Materials and Methods: A total of 256 healers and herbalists were registered by the DMT. Out of them 123 were operating in Bamako. The sample was chosen by convenience and the participants were asked for their verbal informed consent. Interviews using a semi-structured questionnaire were performed with 22 healers and 26 herbalists from September to November 2001 supplemented by 36 consultations using a checklist with 10 healers and two herbalists. The level of knowledge of herb-herb and herb-drug interactions was categorized as low, moderate and high according to the effects reported as results herb-herb and herb-conventional drug interactions. Chi-square, Mann-Whitney and Kruskall-Wallis tests were used for statistical analysis of the data. The level of significance was set at 0,05.Results: Healers and herbalists used four categories of medicines (herbs, ITMs, mineral elements and animal products). The herbs most frequently used were Cassia sieberiana DC, Mitragyna inermis (Willd.) O. Ktze. and Trichilia emetica Vahl. All the practitioners (48/48) were aware of herb-herb interactions. 69% (n=48) of the practitioners were categorized with low level of knowledge of herb-herb interactions. 58% (n=48) of the respondents were aware of herb-drug interactions. The majority 83% (n=48) of the practitioners however, scored low level of knowledge. The healers and herbalists reported thirty-two herb-herb combinations used with T. emetica with Anogeissus leiocarpa (DC) Guill. et Perrott as the mostly used. Swartzia madagascariensis Desv. and Securidaca longepedunculata Fresen. were the herbs that should never be used together orwith other herbs mostly cited. Thirteen herbs and eleven drugs were reported as those that the healers and herbalists told patients to take together, but at different times. M. inermis with antimalarial drugs and A. leiocarpa also with antimalarial drugs were the mostfrequently cited. S. madagascariensis and S. longepedunculata were the herbs reported as not to take with any conventional drugs.Conclusion: The study showed that healers and herbalists have low level of knowledge of both herb-herb and herb-conventional drug interactions. The effects that they reported as results of herb-herb or herb-drug interactions were mainly the positive aspects of the interaction. However there is a room of improvement because the healers and herbalists themselves recognized that their knowledge of herb-drug interactions was low and also they were willing to getmore knowledge about interactions. Recommendations: Based on the fact that the level of knowledge of herb-herb and herb-drug interactions is low and the fact that healers and herbalists reportedpositive effects of interactions, we recommend: - Training healers and herbalists about the possible consequences of herb-druginteractions as well as herb-herb interactions - To do specific study about the interaction that healers and herbalists reported when combining herbs and herbs with conventional drugs.