Since the early 90s, monitoring forest biodiversity has become a challenging issue. At an international scale however, coordinated actions are scarce, and have been to date mostly focused on assessing the sustainability of forest management as a whole (including wood production, forest health) rather than monitoring biodiversity per se. Thus, state and dynamics of forest biodiversity at the European scale still remain poorly known, or based most exclusively on indirect indicators such as deadwood volume. The aim of this paper is to synthesize the schemes used to monitor biodiversity in various European countries and propose solutions to develop an international, coordinated monitoring framework. Currently, two systems are used to monitor biodiversity: 1. Forest data included in general biodiversity monitoring schemes: in France, data concerning specialist forest birds are extracted from the national Breeding Birds Survey and several attempts are made to explain the observed trends. Although this survey provides nationwide biodiversity data, it is limited in terms of explanatory factors as complementary environmental protocols are not developed on the same sampling design; 2. Biodiversity data sampled within National Forest Inventories is the most commonly used scheme in Europe. Forest inventory data are used to complete the European biodiversity monitoring schemes such as Forest Europe and SEBI. As a consequence, these indices are mainly based on indirect biodiversity indicators, for which the link with biodiversity is not always scientifically robust. In addition, direct biodiversity indicators based on forest inventory data concern only a few taxa (trees, vascular plants). Improving forest biodiversity monitoring could hence be based on: (i) adding environmental measures to biodiversity monitoring (e.g. forest structure), or (ii) adding direct biodiversity sampling to national forest inventories. This latter method has been developed in Switzerland and will be used as an example. Finally, recommendations on how an "ideal" monitoring scheme should be developed will be discussed.