The news, therefore, is somewhat pessimistic. Empiric estimate appears to be fairly good in stratifying risk groups for PE but suffers from a lack of standardization, as well as poor sensitivity and specificity. Multiple attempts to standardize pretest risk have used several different but usually overlapping variables, and none appears yet to be much better than simple clinical estimate--the very scale they were trying to replace. The clinician therefore must assess whether, on the basis of what has been published so far, there is any convincing reason to move from an empiric to a score-based prediction system. Apart from the clear advantage of standardization, the score-based systems are varied, not agreed on, and usually clumsy to use in a busy ED. Until one prediction system has been shown to be both easy to remember and use in clinical practice, and to be more accurate than an empiric score, emergency physicians should continue to use their own practice-based empiric system. DVT can be safely, quickly, and accurately diagnosed in the ED, and for most cases there is no need to use a prediction rule. Suspected PE continues to be a complex entity to evaluate and diagnose in the ED, and none of the objective prediction rules have yet been able to improve the difficult task of the ED evaluation of the patient with suspected PE.