Clinical librarianship (CL), currently receiving renewed interest world-wide, seeks to provide quality-filtered information to health professionals at the point of need to support clinical decision-making. This review builds upon the work of Cimpl (Bulletin of the Medical Library Association 1985, 73, 21-8) and attempts to establish the evidence base for CL. The objectives were to determine, from the literature, whether CL services are used by clinicians, have an effect on patient care, and/or clinicians' use of literature in practice and/or are cost-effective. The methodology used was a systematic review of the literature, following, where possible, the NHS Centre for Reviews and Dissemination (CRD) framework. Modifications to this methodology included the resources searched, and the critical appraisal checklist (CriSTAL) used. Two hundred and eighty-four unique references were retrieved. Seventeen (16 unique) evaluative and a further 33 descriptive studies met the inclusion criteria. The quality of reporting of the literature was generally poor. CL programmes appear to be well-used and received by clinicians. However, there is insufficient evidence available on their effect on patient care, clinicians' use of literature in practice, and their cost-effectiveness, thus highlighting the need for further high-quality research.