Modern diagnostic strategies for pulmonary embolism diagnosis almost all rely on an initial assessment of the pretest probability. Clinical prediction rules are decision-making tools using combinations of easily available clinical predictors to define the probability of a disease. The assessment of the clinical probability of pulmonary embolism has an important impact on the diagnostic strategy and on therapeutic management. Clinical prediction rules provide accurate and reproducible estimates of clinical probability. They should be derived and validated following strict methodological standards. The use of clinical prediction rules should be encouraged, since their implementation in local guidelines for pulmonary embolism diagnosis has been shown to improve patients' outcomes.