Although interest in clinical guidelines has never been greater, uncertainty persists about whether they are effective. The debate has been hampered by the lack of a rigorous overview. We have identified 59 published evaluations of clinical guidelines that met defined criteria for scientific rigour; 24 investigated guidelines for specific clinical conditions, 27 studied preventive care, and 8 looked at guidelines for prescribing or for support services. All but 4 of these studies detected significant improvements in the process of care after the introduction of guidelines and all but 2 of the 11 studies that assessed the outcome of care reported significant improvements. We conclude that explicit guidelines do improve clinical practice, when introduced in the context of rigorous evaluations. However, the size of the improvements in performance varied considerably.